Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Development of a high-performance liquid chromatographic assay for digoxin in plasma using post-column… Kwong, Elizabeth 1984

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
831-UBC_1984_A1 K86.pdf [ 6.13MB ]
Metadata
JSON: 831-1.0096571.json
JSON-LD: 831-1.0096571-ld.json
RDF/XML (Pretty): 831-1.0096571-rdf.xml
RDF/JSON: 831-1.0096571-rdf.json
Turtle: 831-1.0096571-turtle.txt
N-Triples: 831-1.0096571-rdf-ntriples.txt
Original Record: 831-1.0096571-source.json
Full Text
831-1.0096571-fulltext.txt
Citation
831-1.0096571.ris

Full Text

DEVELOPMENT OF A HIGH - PERFORMANCE LIQUID CHROMATOGRAPHIC ASSAY FOR DIGOXIN IN PLASMA USING POST-COLUMN FLUOROGENIC DERIVATIZATION by ELIZABETH KWONG B . S c . F a c u l t y o f P h a r m a c e u t i c a l S c i e n c e s U n i v e r s i t y - o f B r i t i s h Co l umb i a V a n c o u v e r , B .C . Canada , 1980 A THESIS SUBMITTED IN PARTIAL FULFILMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY i n THE FACULTY OF GRADUATE STUDIES F a c u l t y o f P h a r m a c e u t i c a l S c i e n c e s D i v i s i o n o f P h a r m a c e u t i c a l C h e m i s t r y We a c c e p t t h i s t h e s i s as c o n f o r m i n g t o t h e r e q u i r e d s t a n d a r d THE UNIVERSITY OF BRITISH COLUMBIA Augus t 1984 © E l i z a b e t h Kwong . 1984 In p r e s e n t i n g t h i s t h e s i s i n p a r t i a l f u l f i l m e n t o f the requirements f o r an advanced degree a t the U n i v e r s i t y o f B r i t i s h Columbia, I agree t h a t the L i b r a r y s h a l l make i t f r e e l y a v a i l a b l e f o r r e f e r e n c e and study. I f u r t h e r agree t h a t p e r m i s s i o n f o r e x t e n s i v e copying of t h i s t h e s i s f o r s c h o l a r l y purposes may be granted by the head o f my department or by h i s or her r e p r e s e n t a t i v e s . I t i s understood t h a t copying or p u b l i c a t i o n o f t h i s t h e s i s f o r f i n a n c i a l gain s h a l l not be allowed without my w r i t t e n p e r m i s s i o n . Department of ?(l^mu!s*M aJl Cfa^h&y The U n i v e r s i t y of B r i t i s h Columbia 1956 Main Mall Vancouver, Canada V6T 1Y3 Date Jefr C, n<H-E-6 (3/81) ABSTRACT Digoxin i s a cardiac glycoside obtained i n purified form from the leaves" of Digitalis lanata. The molecule i s composed of a sugar portion and an aglycone (genin) portion. Chemically i t is. the-drug of choice for the treatment of congestive heart failure and certain disturbances of cardiac rhythm. The therapeutic significance of digoxin, and particularly the relatively narrow margin between a therapeutic and toxic dose, has warranted the development of a specific and sensitive analytical method for their quantitation i n biological samples. The principal method employed by most researchers and hospital laboratories for the assay of digoxin i s based on the use of a radioimmunoassay (RIA) procedure. Present reports have shown that the digoxin metabolites, digoxigenin mono-digitoxoside, digoxigenin bis-digitoxoside and digoxigenin cross-react extensively i n RIA. Dihydrodigoxin cross-reactivity depends on the RIA k i t used. It was also shown that the RIA procedure gave false-positive results in specimens taken from non-digitalized renal patients. Such discrepancies raise questions regarding the "true" plasma digoxin concentrations measured by the current RIA method. Therefore, to ascertain the analytical r e l i a b i l i t y of available assays i n measuring digoxin levels, this thesis reports the development of an HPLC separation of digoxin from i t s metabolites and some commonly co-administered drugs which, when coupled with post-column derivatization, w i l l quantify i i the levels of intact digoxin i n plasma. Essentially this procedure i s based on a novel method developed by Gfeller et a l . (9). The post-column detection i s based on the reaction of hydrochloric acid with the steroid portion of the cardiac glycosides. Fluorescence of the derivative i s further enhanced by the addition of a hydrogen peroxide/ ascorbic acid mixture. The fluorogenic reactants are also employed i n the method uti l i z e d i n the United States Pharmacopeia and hence are considered to be reliable. We have undertaken several modifications to the a i r segmentation post-column derivatization step developed by Gfeller's group (9). To increase the sensitivity of detection, a non-segmented post-column reactor i s used. The glycosides are i n i t i a l l y separated on a 15 cm octadecylsilane, 3u packing column, using a combination of methanol, ethanol, isopropanol and water as the mobile phase at a flow rate of 0.3 mL/min. A solution of dehydroascorbic acid and hydrochloric acid are added to the column effluent at 0.23 mL/min to form the fluorophore using a 10 m knitted reactor controlled at a temperature of 79°C. The glycosides react with HC1 and dehydroascorbic acid to form the fluorophore which i s monitored by a Waters fluorescence detector equipped with a 360nm excitation f i l t e r and a 425 nm emission f i l t e r . The quantitation of digoxin in plasma i s carried out by the incorporation of an internal standard, digitoxigenin. The method involves the extraction of the compound from plasma with methylene chloride containing 2% propanol. Endogenous substances, commonly c o - a d m i n i s t e r e d d r u g s , a n d m e t a b o l i t e s o f d i g o x i n d o n o t i n t e r f e r e w i t h t h e m e t h o d . T h e t o t a l c h r o m a t o g r a p h i c t i m e , i n c l u d i n g t h e p o s t - c o l u m n d e r i v a t i z a t i o n s t e p , i s a b o u t 4 0 m i n u t e s . T h e d e t e c t i o n l i m i t o f t h e m e t h o d i s 0 . 5 n g / m L o f d i g o x i n i n p l a s m a . T h e a v e r a g e e x t r a c t i o n r e c o v e r y i s 7 0 % o v e r t h e t h e r a p e u t i c c o n c e n t r a t i o n r a n g e . D e t e c t o r r e s p o n s e i s l i n e a r f r o m 0 . 5 n g / m L t o 3 . 3 n g / m L o f p l a s m a . U s i n g H P L C s e p a r a t i o n a n d p o s t - c o l u m n d e t e c t i o n w i t h p a r a l l e l q u a n t i t a t i o n b y r a d i o i m m u n o a s s a y , w e e x a m i n e d t h e p l a s m a o f d i g i t a l i z e d p a t i e n t s . T h e m e a n H P L C / R I A r a t i o s a r e 0 . 9 4 1 0 . 3 ( 3 - D . ) ( A c u t e C a r e U n i t o f t h e U n i v e r s i t y H o s p i t a l ) a n d 1 . 0 ± 0 . 3 4 ( 3 . D . ) ( V a n c o u v e r G e n e r a l H o s p i t a l ) . T h e s e r a t i o s a r e c o m p a r a b l e t o t h o s e f o u n d b y s p e c i f i c H P L C - R I A r e s u l t s r e p o r t e d b y o t h e r w o r k e r s . T h e o v e r a l l p e r f o r m a n c e d e m o n s t r a t e s t h a t t h i s s y s t e m h a s t h e s e n s i t i v i t y , l i n e a r i t y a n d s p e c i f i c i t y d e s i r e d f o r t h e d e t e r m i n a t i o n o f d r u g c o n c e n t r a t i o n s i n p l a s m a f r o m d i g i t a l i z e d p a t i e n t s . i v TABLE OF CONTENTS Chap te r Page ABSTRACT i i L I ST OF TABLES x L IST OF FIGURES x i SYMBOLS AND ABBREVIATIONS x i i i ACKNOWLEDGEMENT x v i INTRODUCTION - RATIONALE 1 1 LITERATURE SURVEY Gene ra l 4 1 . 1 . Pharmacodynamic S t u d i e s 4 1 . 1 . 1 . . C h e m i s t r y 4 1 . 1 . 2 . E f f e c t o f D i g o x i n on M y o c a r d i a l F u n c t i o n 5 1 . 1 . 2 . 1 . P o s i t i v e I n o t r o p i c A c t i o n 6 1 . 1 . 2 . 2 . C a r d i a c E l e c t r o p h y s i o l o g i c E f f e c t s 6 1 . 1 . 3 . P h a r m a c o k i n e t i c s 7 1 .1 .3 .1 . A b s o r p t i o n 7 1 . 1 . 3 . 2 . D i s t r i b u t i o n 8 1 . 1 . 3 . 3 . M e t a b o l i s m and E l i m i n a t i o n 9 1 . 2 . P lasma D i g o x i n L e v e l s 12 1 . 3 . Method o f A n a l y s i s 17 1 . 3 . 1 . Chemica l Methods 17 1 . 3 . 2 . Rad io immunoassay (R IA ) 19 1 . 3 . 3 . B i o l o g i c a l A s say s 23 1 . 3 . 4 . Ch r oma tog r aph i c Methods 24 1 . 3 . 5 . HPLC-RIA 26 1 . 3 . 6 . P o s t - c o l u m n D e r i v a t i z a t i o n 27 2 EXPERIMENTAL 30 2 . 1 . S u p p l i e s 30 2 . 1 . 1 . Chem i c a l s 30 2 . 1 . 2 . Reagents 30 2 . 1 . 3 . S o l v e n t s 31 2 . 1 . 4 . S u p p l i e s f o r E x t r a c t i o n s 31 2 . 1 . 5 . P o s t - c o l u m n D e r i v a t i z a t i o n S u p p l i e s 31 v Chap t e r Page 2 . 1 . 5 . 1 . Packed -bed R e a c t o r 31 2.1 . 5 . 2 . A i r - s e g m e n t e d Rea c t o r 32 2 . 1 . 5 . 3 . Non-segmented R e a c t o r 32 2 . 2 . Equ ipmeht 33 2 . 2 . 1 . H i g h - p e r f o r m a n c e L i q u i d Ch r oma tog r aph i c 33 (HPLC) Sys tem 2 . 2 . 2 . F low C e l l 34 2 . 2 . 3 . P o s t - c o l u m n R e a c t i o n Sys tem 34 2 . 2 . 4 . E x t r a c t i o n 37 . 2 . 2 . 5 . Gas Chromatography , 37 2 . 2 . 6 . HPLC - Mass S p e c t r o m e t r y 38 2 . 3 . S t a t i o n a r y Phase 39 2 . 4 . D e r i v a t i z a t i o n P r o c edu r e s 39 2 . 4 . 1 . For Gas Chromatography and E l e c t r o n - 40 Cap tu r e D e t e c t i o n 2 . 4 . 1 . 1 . Reagen t : N - m e t h y l - b i s - 40 t r i f l u o r o a c e t a m i d e (MBTFA) 2 . 4 . 1 . 2 . Reagen t : T r i - s i l Z 40 2 . 4 . 1 . 3 . Reagen t : H e p t a f l u o r o b u t y r y l - 40 i m i d a z o l e (HFBI ) 2 . 4 . 2 . I n - v i t r o F l u o r o m e t r i c A s s a y Method 40 2 . 4 . 2 . 1 . S u l f u r i c A c i d 41 2 . 4 . 2 . 2 . A c e t i c A n h y d r i d e / A c e t y l C h l o r i d e / 41 T r i f l u o r o a c e t i c A c i d 2 . 4 . 2 . 3 . Dansy l H y d r a z i n e 42 2 . 4 . 2 . 4 . Hydrogen P e r o x i d e / A s c o r b i c A c i d / 42 H y d r o c h l o r i c A c i d 2 . 5 . L i q u i d Chromatography and U l t r a v i o l e t D e t e c t i o n 42 2 . 6 . P r e p a r a t i o n o f S o l u t i o n s 43 2 . 6 . 1 . Reagents 43 2 . 6 . 1 . 1 . ' G l y c i n a m i d e / P o t a s s i u m F e r r i c y a n i d e 43 2 . 6 . 1 . 2 . I s o n i c o t i n y l h y d r a z i n e (INH) - 43 Aluminum C h l o r i d e ( A l ) S o l u t i o n 2 . 6 . 1 . 3 . D e h y d r o a s c o r b i c A c i d 43 2 . 6 . 2 . S t a n d a r d S o l u t i o n s 44 v i Chap te r Page 2 . 6 . 2 . 1 . D i g o x i n S t a n d a r d s 44 2 . 6 . 2 . 2 . I n t e r n a l S t a n d a r d S o l u t i o n s 44 2 . 6 . 2 . 3 . D i g o x i n M e t a b o l i t e s 45 2 . 6 . 2 . 4 . C o - a d m i n i s t e r e d Drugs 45 2 . 6 . 2 . 5 . M o b i l e Phase 46 2 . 7 . HPLC P o s t - c o l u m n D e r i v a t i z a t i o n P r o c edu r e 47 2 .7 .1 . Column S e l e c t i o n 47 2 . 7 . 2 . O p t i m i z e d HPLC Non-segmented P o s t - c o l u m n 47 D e r i v a t i z a t i o n Sys tem 2 . 7 . 3 . O the r A l t e r n a t i v e Reagents Employed f o r 49 Der i v a t i z a t i o n 2 . 7 . 4 . Pa cked -bed R e a c t i o n Sys tem 51 2 . 7 . 5 . A i r - s e g m e n t e d R e a c t i o n Sys tem 51 2 . 7 . 6 . O p t i m i z a t i o n P r o c edu r e f o r P o s t - c o l u m n 52 D e r i v a t i z a t i o n 2 . 7 . 6 . 1 . R e l a t i v e P r o p e r t i e s o f 53 F l u o r o g e n i c Reagents 2 . 7 . 6 . 2 . R e a c t i o n Tempe ra tu re s 54 2 . 7 . 6 . 3 . R e a c t i o n K i n e t i c s 54 2 . 7 . 6 . 4 . D e t e c t o r Wave l eng th 56 2 . 7 . 6 . 5 . C o n t r o l o f B a s e l i n e No i s e 56 2 . 7 . 6 . 6 . (J - c yc l o d e x t r i n A d d i t i o n 57 2 . 8 . R e p r o d u c i b i l i t y and L i n e a r i t y o f t h e A s say Method 57 2 . 9 . P lasma E x t r a c t i o n P r o c edu r e 58 2 . 9 . 1 . E x t r a c t i o n P r o c edu r e i n Water 58 2 . 9 . 1 . 1 . S o l v e n t - s o l v e n t E x t r a c t i o n o f 58 D i g o x i n i n Water 2 . 9 . 1 . 2 . S o l i d Phase C a r t r i d g e E x t r a c t i o n 58 o f D i g o x i n i n Water 2 . 9 . 2 . E x t r a c t i o n P r o c edu r e i n " S p i k e d " P lasma 59 2 . 9 . 2 . 1 . S o l v e n t - s o l v e n t E x t r a c t i o n o f 59 D i g o x i n i n P lasma 2 . 9 . 2 . 2 . S o l i d Phase C a r t r i d g e E x t r a c t i o n o f 60 D i g o x i n i n P lasma 2 . 9 . 3 . A c t u a l E x t r a c t i o n P r o c e d u r e o f D i g o x i n 62 i n P lasma 2 . 1 0 . Re cove r y and P r e c i s i o n o f E x t r a c t i o n 63 2 . 1 1 . C a l i b r a t i o n Curve and S e n s i t i v i t y o f E x t r a c t i o n 63 v i i Chap te r Page 2 . 1 2 . S p e c i f i c i t y 65 2 . 1 3 . Q u a l i t y C o n t r o l P r o c edu r e 65 2 . 1 4 . A n a l y s i s o f D i g i t a l i z e d P a t i e n t P lasma Samples 66 2 . 1 5 . C a l c u l a t i o n s 66 3 RESULTS AND DISCUSSIONS 68 3 . 1 . P r e l i m i n a r y S t u d i e s f o r t h e Deve lopment o f an 68 A n a l y t i c a l Sys tem f o r D i g o x i n 3 . 1 . 1 . On-co lumn I n j e c t i o n f o r C a p i l l a r y Gas 68 Chromatography (GC) 3 . 1 . 2 . L i q u i d Chromatography - Mass S p e c t r o m e t r y 69 3 . 1 . 3 . L i q u i d Chromatography and UV D e t e c t i o n 71 3.1 . 4 . F l u o r o m e t r i c D e t e r m i n a t i o n o f D i g o x i n 72 3 . 2 . O p t i m i z a t i o n o f t h e HPLC P o s t - c o l u m n D e r i v a t i z a t i o n 75 Method 3 . 2 . 1 . O t he r Reagents f o r D e r i v a t i z a t i o n 75 3 . 2 . 2 . F low C e l l , 7 8 3 . 2 . 3 . Cho i c e o f Ch r oma t og r aph i c Columns 83 3 . 2 . 4 . E f f e c t o f M o b i l e Phase 86 3 . 2 . 5 . Cho i c e o f R e a c t i o n D e t e c t o r 89 3 . 2 . 5 . 1 . Packed -bed R e a c t o r 89 3 . 2 . 5 . 2 . A i r - s e g m e n t e d R e a c t o r 90 3 . 2 . 5 . 3 . T u b u l a r Non-segmented R e a c t o r 93 3 . 2 . 6 . R e l a t i v e P r o p o r t i o n s o f Reagents 93 3 . 2 . 7 . O p t i m i z a t i o n o f R e a c t i o n Tempera tu re 96 3 . 2 . 8 . R e a c t i o n K i n e t i c s 96 3 . 2 . 9 . D e t e c t o r Wave l eng t h 104 3 . 2 . 1 0 . Band B r o a d e n i n g 104 3 . 2 . 11 . B a s e l i n e N o i s e 108 3 . 2 . 1 2 . S e n s i t i v i t y 109 3 . 2 . 1 3 . R e p r o d u c i b i l i t y and C a l i b r a t i o n Curve 111 3 . 2 . 1 4 . C h o i c e o f I n t e r n a l S t a n d a r d 111 3 . 3 . A p p l i c a t i o n o f t h e A n a l y t i c a l P r o c edu r e t o :'• , "117 P lasma Samples 3 . 3 . 1 . O p t i m i z a t i o n o f E x t r a c t i o n Method i n 117 S p i k e d Water 3 . 3 . 2 . O p t i m i z a t i o n o f E x t r a c t i o n Method i n P lasma 121 3 . 3 . 3 . Re cove r y and P r e c i s i o n o f t h e E x t r a c t i o n 130 Method v i i i Chap t e r Page 3 . 3 . 4 . C a l i b r a t i o n Curve and Method S e n s i t i v i t y 130 3 . 3 . 5 . S p e c i f i c i t y 135 3 . 3 . 6 . Q u a l i t y C o n t r o l P r o c edu r e 136 3 . 4 . D e t e r m i n a t i o n o f D i g o x i n i n P a t i e n t P lasma by 136 HPLC P o s t - c o l u m n D e r i v a t i z a t i o n 4 SUMMARY AND CONCLUSIONS 145 5 REFERENCES . 146 i x LIST OF TABLES T ab l e Page 1 R e s u l t s o f F l u o r o m e t r i c Methods 74 2 E f f e c t o f D i f f e r e n t A c i d s on t h e F l u o r e s c e n c e 77 I n t e n s i t y o f D i g o x i n U s i ng P o s t - c o l u m n D e r i v a t i z a t i o n 3 M o b i l e Phases P o l a r i t y and R e s o l u t i o n o f D i g o x i n 87 f rom D i h y d r o d i g o x i n 4 R e l a t i v e P r o p o r t i o n s o f Reagents . 94 5 F i l t e r C omb i n a t i o n s f o r Optimum D e t e c t o r Wave l eng th 106 6 Band B r o a d e n i n g o f R e a c t i o n Sys tem 107 7 Data f o r t h e S t a n d a r d Curve 112 8 S o l v e n t - s o l v e n t E x t r a c t i o n o f D i g o x i n i n Water 118 9 S o l i d Phase C a r t r i d g e E x t r a c t i o n o f D i g o x i n i n Water 120 10 S o l v e n t - s o l v e n t E x t r a c t i o n o f D i g o x i n f rom P lasma 122 11 S o l i d Phase C a r t r i d g e E x t r a c t i o n o f D i g o x i n i n P lasma 124 12 S o l v e n t - s o l v e n t E x t r a c t i o n o f D i g o x i n Combined w i t h 125 S o l i d Phase C a r t r i d g e I s o l a t i o n 13 Recove r y o f D i g o x i n U s i n g D i f f e r e n t N o n - p o l a r 128 S o l v e n t Washes 14 R e p r o d u c i b i l i t y and Re cove r y Data 132 15 Data f o r P lasma S t a n d a r d Curve 133 16 Data f o r Q u a l i t y C o n t r o l P r o c edu r e 138 17 D i g o x i n i n P lasma o f D i g i t a l i z e d P a t i e n t s 140 x L IST OF FIGURES F i g u r e Page 1 S t r u c t u r e o f D i g o x i n 4 2 M e t a b o l i c Pathway o f D i g o x i n 10 3 Chemica l P r o c edu r e s f o r t h e P h o t o m e t r i c and 17 F l u o r o m e t r i c D e t e r m i n a t i o n o f C a r d e n o l i d e s 4 The P r i n c i p l e o f t h e Rad io immunoassay Method 20 5 S chema t i c o f t h e N o i s e F i l t e r 35 6 S chema t i c D iagram o f t he R e a c t i o n D e t e c t o r U n i t 36 f o r F l u o r e s c e n c e D e t e c t o r s 7 F low D iagram o f t h e P r e s e n t P o s t - c o l u m n 48 D e r i v a t i z a t i o n Method 8 E x t r a c t i o n Scheme o f D i g o x i n f rom P lasma 64 9 Gas Chromatogram o f H F B - D i g o x i n U s i n g 70 On-co lumn I n j e c t i o n 10 HPLC Chromatogram o f Dansy l Hydrazone o f D i g o x i n 73 11 E x c i t a t i o n and E m i s s i o n S p e c t r a o f D i g o x i n 76 12 S e p a r a t i o n o f D i g o x i n and i t s M e t a b o l i t e s U s i n g 81 10 and 25 cm Columns 13 O p t i m i z a t i o n o f Ch r oma t og r aph i c R e s o l u t i o n 82 14 S e p a r a t i o n o f D i g o x i n and i t s M e t a b o l i t e s U s i n g 84 C-8 Column 15 O p t i m i z e d S e p a r a t i o n o f D i g o x i n and i t s M e t a b o l i t e s 85 16 E f f e c t o f S o l v e n t P o l a r i t i e s on R e s o l u t i o n 88 17 Low l e v e l D e t e c t i o n o f D i g o x i n U s i n g A i r 92 S egmen t a t i o n Sys tem 18 E f f e c t o f Hydrogen P e r o x i d e C o n c e n t r a t i o n i n 95 A s c o r b i c A c i d on F l u o r e s c e n c e 19 O p t i m i z a t i o n o f R e a c t i o n Tempe ra t u r e 97 x i F i g u r e Page 20 I n f l u e n c e o f Tempe ra tu re on t h e S e n s i t i v i t y 98 o f D e t e c t i o n 21 R e a c t i o n K i n e t i c s 100 22 E f f e c t o f R e a c t i o n C o i l L eng th on F l u o r e s c e n c e 101 23 Se conda r y F low P a t t e r n i n t h e C r o s s S e c t i o n o f 102 a C o i l e d Tube 24 O p t i m i z a t i o n o f HPLC F low Rate 105 25 E f f e c t o f t h e N o i s e F i l t e r on t h e S i g n a l / N o i s e 110 R a t i o 26 S t a n d a r d Curve o f D i g o x i n 113 27 S e p a r a t i o n o f D i g o x i n , i t s M e t a b o l i t e s and 115 C o - a d m i n i s t e r e d Drugs u s i n g P o s t - c o l u m n D e t e c t i o n 28 T y p i c a l Chromatogram o f a B l ank and S p i k e d P lasma 131 29 C a l i b r a t i o n Cu rve o f D i g o x i n i n P lasma Samples 134 30 Compar i son o f D i g o x i n D e t e r m i n a t i o n u s i n g HPLC-PC 137 and RIA Method on R a d i o a s s a y C o n t r o l Serum 31 C o r r e l a t i o n between P lasma D i g o x i n L e v e l s by 143 HPLC-PC and RIA Method x i i SYMBOLS AND ABBREVIATIONS ACU H e a l t h S c i e n c e A c u t e Care H o s p i t a l A l A luminum ' ' ' . ; A-V A t r i o - v e n t r i c u l a r C C e n t i g r a d e cc c u b i c c e n t i m e t e r CI Chemica l i o n i z a t i o n cm c e n t i m e t e r C V . c o e f f i c i e n t o f v a r i a t i o n DCM d i e h l o r o m e t h a n e DLI d i r e c t l i q u i d i n t r o d u c t i o n DRP d i g o x i n r e d u c t i o n p r o du c t ECD e l e c t r o n c a p t u r e d e t e c t o r ELISA e n z y m e - l i n k e d immunosorbent EMIT enzyme m u l t i p l i e d immunoassay t e c h n i q u e ETOH e t h a n o l g gram GC gas ch r oma tog r aphy H2O2 hydrogen p e r o x i d e HC1 h y d r o c h l o r i c a c i d HFBA h e p t a f l u o r o b u t y r i c a c i d HFBI h e p t a f l u o r o b u t y r y l i m i d a z o l e HPLC H i g h - p e r f o r m a n c e l i q u i d ch r oma tog r aphy HPLC-PC H i g h - p e r f o r m a n c e l i q u i d c h r o m a t o g r a p h y - p o s t - c o l ID i n t e r n a l d i a m e t e r i n o r " i n c h e s x i i i I i H i s o n i c o t i n y l h y d r a z i n e I n t c t I n t e r c e p t I n t . S t d . I n t e r n a l s t a n d a r d IPA i s o p r o p y l a l c o h o l K + p o t a s s i u m i o n LC-MS l i q u i d ch roma tog raphy -mass s p e c t r o m e t r y m me te r M m o l a r MBTFA N - m e t h y l - b i s - t r i f l u o r o a c e t a m i d e MeOH methano l mEq m i l 1 i e q u i v a l e n t mg m i l l i g r a m MHPLC m i c r o h i g h - p e r f o r m a n c e l i q u i d c h r oma tog r aphy m i n ( s ) m i n u t e ( s ) mL m i l l i l i t e r mM m i l l i m o l a r mm m i l l i m e t e r MS mass s p e c t r o m e t r y mV m i l l i v o l t s n # o f samp les n # o f pu re s o l v e n t i n t he m i x t u r e ( e q u a t i o n 2) N a + sod ium i o n ng nanogram nm nanometer OD o u t e r d i a m e t e r ODS o c t a d e c y l s i l a n e P' o r P ' e p o l a r i t y pg p i c og r am x i v P i s o l v e n t p o l a r i t y pa r ame te r o f i t h t e rm r c o r r e l a t i o n c o e f f i c i e n t RFI r e l a t i v e f l u o r e s c e n c e i n t e n s i t y RIA r ad i o immunoas say RP r e v e r s e d phase RPM r e v o l u t i o n pe r m inu t e Rs r e s o l u t i o n S . D . s t a n d a r d d e v i a t i o n t j and t j j r e t e n t i o n t ime o f d i h y d r o d i g o x i n and d i g o x i n TLC t h i n l a y e r c h r oma tog r aphy u or urn m i c r o n uL m i c r o l i t e r USP U n i t e d S t a t e s Pha rmacope ia UV u l t r a v i o l e t V v o l t s VGH Vancouve r Gene ra l H o s p i t a l Wj and W J J bandw id th o f t h e two bands a t b a s e l i n e E summat ion 0 vo lume f r a c t i o n o f component o f s o l v e n t sy s tem xv ACKNOWLEDGEMENT I wou ld l i k e t o thank D r . K e i t h McE r l a n e f o r h i s e x c e l l e n t g u i d a n c e and c o n s t a n t s u p p o r t t h r o u g h o u t my g r a d u a t e c a r e e r . H i s h e l p , bo th a c a d e m i c a l l y and p e r s o n a l l y , has been phenomenal and has e n a b l e d me t o c omp l e t e my s t u d i e s . The v a l u a b l e s u g g e s t i o n s o f f e r e d by t h e members o f my g r a d u a t e c o m m i t t e e , D r s . James O r r , F rank A b b o t t , and Dav i d God i n a r e g r a t e f u l l y a c k n o w l e d g e d . The f r i e n d s h i p shown by t he s t u d e n t s and c o l l e a g u e s i n t h e F a c u l t y o f P h a r m a c e u t i c a l S c i e n c e s , U n i v e r s i t y o f B r i t i s h Co lumb ia i s g r e a t l y a p p r e c i a t e d . I wou ld p a r t i c u l a r l y l i k e t o t hank Ms . B a r b a r a F r a s e r f o r a l l he r h e l p i n t he p r e p a r a t i o n o f . t h i s t h e s i s . I wou ld a l s o l i k e t o e x p r e s s my g r a t i t u d e to M r s . B a r b a r a M c E r l a n e f o r p r o o f r e a d i n g my t h e s i s . Above a l l , I a cknow ledge the a s s i s t a n c e g i v e n by D r . R i c h a r d Wa l l and M r . Ro l and B u r t o n who have a lway s shown t h e i r b r i l l i a n t i n s i g h t i n t o my t e c h n i c a l p r o b l e m s . I am d e e p l y i n d e b t e d t o t he Canad i an Hea r t F o u n d a t i o n f o r t h e i r f i n a n c i a l s u p p o r t . F i n a l l y , I wou ld l i k e t o expr.ess my d e e p e s t g r a t i t u d e t o each member o f my f a m i l y f o r t h e i r l o v e and c o n c e r n . x v i INTRODUCTION Digoxin i s the most popular cardiac glycoside presently used for cardiac insufficiency. Although i t i s known to be a highly beneficial drug, i t i s also very potent and exhibits a narrow margin of therapeutic efficacy and safety. The determination of the optimal digoxin dosage in c l i n i c a l practice i s complex. For'these reasons, plasma level and dosage schedules are frequently monitored and regulated. Problems with therapeutic efficacy continue to be reported i n the literature (1) and many of these are ascribed to the analytical methodology employed. Although a host of chemical, chromatographic and biological methods have been reported for measurement of digoxin in body fluids, the principal and most sensitive method employed i n hospital laboratories i s based on the use of radioimmunoassay procedures. Although these methods are sensitive, the antibodies involved have frequently been reported to be nonspecific and are subject to cross reactivity with some of the less cardioactive metabolites of digoxin, as well as with endogenous and exogenous substances with structural similarity to the aglycone portion of the digoxin molecule. A rapid, specific assay for digoxin using physico-chemical measurement i s therefore needed. High performance liquid chromatography (HPLC) combined with ultraviolet (UV) detection methods which are available (2,3) have been successful i n fa c i l i t a t i n g the separation of digoxin from i t s metabolites, but the 1 applicability of this method to analysis in biological fluids has not been investigated due to the inherent lack of sensitivity of the UV detector and low absorptivity of digoxin. To achieve the necessary sensitivity required for therapeutic drug monitoring, numerous studies (4-8) have combined HPLC and radioimmunoassay (RIA) methods. Most of the methods described (4-6) either involved the administration of t r i t i a t e d digoxin to the subjects or the use of tr i t i a t e d internal standards prior to HPLC separation. These methods are often tedious and are not readily adaptable for routine monitoring of serum samples. The two most recent HPLC-RIA methods (7,8) offer some convenience for the c l i n i c a l monitoring of digoxin plasma concentration i n digitalized patients, however, timed collection of eluates employed i n these methods i s very c r i t i c a l and may"introduce errors'"in the assay method. An efficient HPLC separation reported for digoxin and i t s metabolites coupled with highly sensitive post-column derivatization (9) offers a possibility to develop an' extremely selective, sensitive and accurate method of analysis for digoxin. The specific aims of the project were as follows: 1. To develop a highly selective and sensitive HPLC post-column derivatization procedure for the quantitation of therapeutic levels of digoxin without interference from i t s metabolites and other commonly co-administered drugs. 2. To develop suitable extraction procedures for effective recovery of digoxin from plasma. 2 3. To include other commonly co-prescribed drugs such as spironolactone, furosemide, quinidine etc. and to determine i f these interfere with either the chromatographic process or detection system. 4. To test the precision and accuracy of the developed method using spiked samples i n plasma obtained from the blood bank of the hospital and the Red Cross. 5. In collaboration with the hospital laboratories, to assay samples of plasma obtained from digoxin treated patients and to compare the results with those obtained by the hospital RIA method on the same sample. 3 1. LITERATURE SURVEY Digitalis glycosides have been used for thousands of years to treat human diseases (10) . The modern era of treatment with the glycosides begins with the work of William Withering in the eighteenth century. His papers introduced the successful use of d i g i t a l i s for the treatment of patients suffering from dropsy (10). The progress in administration of d i g i t a l i s glycosides can be seen today in the use of digoxin as the rational choice in the pharmacologic management of congestive heart failure and certain disturbances of cardiac rhythm. 1.1 Pharmacodynamic studies: 1.1.1 Chemistry Digoxin (Fig. 1), a d i g i t a l i s glycoside, consists of a genin and a sugar moiety bound by an oxygen atom. This glycoside can be isolated from the leaves of Digitalis lanata. The basic structure of F I G . 1 STRUCTURE OF DIGOXIN 4 the aglycone or genin i s a cyclopentaperhydrophenanthrene skeleton, which forms the base of a l l steroids- To the ring at C-17, i s attached a five-membered beta-unsaturated lactone ring. The combination of the steroid nucleus and the lactone ring i s known as the genin or aglycone. The sugar portion, a tridigitoxoside, i s attached to the steroid nucleus at position 3; subsequent acid hydrolysis cleaves the digitoxoses and yields the corresponding aglycones. Hydroxyl substitutions are present at position 12 and 14. Carbons 10 and 13 of the genin moiety have a methyl group. The following conditions must be f u l f i l l e d i f the glycoside i s to have cardioactivity (11): 1. The steroid nucleus has a cis - c i s configuration at C and D ring, which probably i s involved in binding to the membrane-ATPase. This determines the specificity and extent of action. 2. An unsaturated lactone moiety (A^-structure) attached to the nucleus at C-17 i s presumed to combine with the potassium (K+) center of the transport ATPase (12). 3. A sugar component i s bound through an ether-type bond at C-3 of ring A of the nucleus. This auxiliary sugar side-chain influences water solubility and pharmacokinetics of the molecule. 1.1.2 Effect of digoxin on myocardial function: Each of the essential properties of cardiac muscle— contractility, conduction, refractoriness and automaticity— are altered by digoxin. The principles of c l i n i c a l use of digoxin are 5 based simply on two major effects of the drug. 1.1.2.1 Positive inotropic action: It i s increasingly recognized that the effectiveness of digoxin in the treatment of congestive heart failure i s a consequence of i t s positive inotropic effects. A beneficial cycle i s set up i n which the increased force of myocardial contraction produces more efficient emptying of the heart chambers. The drug increases cardiac output, relieves the elevated ventricular pressure, pulmonary congestion and venous pressure, and the consequent reduction i n the heart size further enhances i t s contractility. Diuresis i s also brought about with r e l i e f of edema (13). The majority of evidence for the mechanism of this positive inotropic action to date suggests that this action i s related to the a b i l i t y to inhibit the Na-K ATPase i n the myocardial sarcolemma (14). This action leads to an increase i n intracellular sodium (Na*) (15) which i n turn results i n an increase in intracellular calcium (16). This enhances excitartion-contraction coupling i n the muscle cells and hence augmentation of contractile force. 1.1.2.2 Cardiac electrophysiologic effects: There i s general agreement as to the c l i n i c a l use of digoxin i n the prevention, control and conversion of supraventricular tachycardias (17). In a t r i a l f i b r i l l a t i o n and fl u t t e r , glycosides slow the ventricular rates. This i s achieved by the prolongation of the refractory period of the A-V node, which allows for fewer supraventricular impulses to reach the ventricles (17). In 6 paroxysmal supraventricular tachycardia, due to atrio-ventricular A-V node reentry, glycosides exert their vagally mediated effect to decrease the rate of impulse conduction as well as to prolong the refractory period in this structure (18,19). The cardiac electrophysiologic effects of digoxin are also believed to be due to the inhibition of the Na+pump which maintains intracellular electronegativity by actively transporting Na+out of the c e l l . An altered ratio of intracellular to extracellular Na+and K +results in a reduction of the resting transmembrane potential (20), which leads to enhanced automaticity as well as reduced velocity of impulse conduction, especially in those areas such as the AV node where resting membrane potential i s normally low (21). 1.1.3 Pharmacokinetics: 1.1.3.1 Absorption: When administered by the oral route, digoxin i s mainly absorbed at the level of the small intestine and to a lesser extent, i n the distal small intestine and colon. Absorption shows considerable inter-individual variation, but averages 85% for the e l i x i r and 60-80% for tablets (22) . Numerous manuscripts (23-26) appearing in the last few years on digoxin bioavailability, have clearly documented important differences among the various preparations and have noted that the bioavailability and rate of absorption are s t r i c t l y related to the rate of dissolution i n the gut. Postprandial administration and delayed gastric emptying decrease the rate of absorption and delay the time to peak serum 7 digoxin levels, but total quantity of drug absorbed i s minimally altered. Drugs such as propantheline and metoclopramide, modify gastric motility and hence alter digoxin absorption (27) . Other drugs may also interfere with digoxin absorption when administered' concomittantly. These include kaolin pectate (28) , cholestyramine (29) and neomycin (30). Disease states such as malabsorption syndrome also reduce the absorption of digoxin tablets (31). 1.1.3.2 Distribution: The time course of the serum digoxin concentration indicates that the pharmacokinetics of digoxin should be described by a model containing at least two distinct compartments (32). After oral administration, peak serum concentration i s reached at about 45 to 60 minutes. The next step i s the distribution phase where the serum h a l f - l i f e ranges from 32 to 48 hours, which makes toxic reactions easier to manage i f they occur (33) . An equilibrium, or maximum peak concentration, occurs 4-6 hours after oral administration. For this reason, a correct evaluation of steady-state concentrations may be obtained only i f sampling i s performed at least 6-8 hours after drug intake. Digoxin i s 23-40% bound to serum protein (34-36). Decreased plasma protein binding of digoxin has been found in patients with uremia (37) but i s insignificant for therapeutic effect. The glycoside i s also bound to tissue proteins; high concentrations usually being found i n the heart, kidney and liver (38-40). Quantitatively, however, the largest depot for digoxin i s skeletal 8 muscle (41). Digoxin i s poorly distributed to adipose tissue, therefore lean body weight i s most relevant to dosing considerations (38-40). The extensive distribution of digoxin in tissue i s reflected by the large volume of distribution (Vd) (5.1-7.3 L/Kg) i n healthy volunteers, but this drops to 2.6-4.3 L/Kg (32,42,43) in patients with renal failure. The Vd i n neonates and infants i s higher than that found i n adults (in infants 2-81 days old, Vd i s 9.9 L/Kg (44)). The basis for this difference seems to be due partially to an increased tissue binding of digoxin in the younger age group. /Among other possible factors i s a change with age i n body composition — i . e . , larger amount of total body water with a greater proportion in the extracellular water compartments and a reduced quantity of plasma protein available for drug protein binding. 1.1.3.3 Metabolism and elimination: According to several reports, digoxin i s mainly eliminated unchanged in urine, being subjected to limited metabolic degradation (45-47). From 67%-93% of a dose of digoxin i s recovered i n the urine, mainly as unchanged drug (48-50). However, mechanisms other than renal excretion of the unchanged compound may play a significant part i n the disposition of digoxin i n man i n several instances (51-53). Digoxin undergoes biotransformation by the l i v e r . The chemical transformations that occur consist of saturation of the lactone ring to produce dihydrodigoxin or i t s aglycone, dihydrodigoxigenin, and 9 hydrolysis of the sugar moieties (54) (Fig. 2). Luchi and Gruber (55) observed that in patients requiring unusually high amounts of digoxin, digoxigenin and dihydrodigoxigenin accounted for more than 50% of the extractable excreted material, while the remainder was s t i l l i n the unchanged form. Clark and Kalman (52) examined the distribution and excretion of metabolites in 50 patients and found that dihydrodigoxin was detectable in 50% of the patients and constituted between 1-48% of total glycoside. From 2% to 10% of the glycosides excreted in.the urine were polar, water soluble metabolites. +2H • DIGOXIN-(GENIN + 3 SUGARS) -2H * DIGOXIGENIN BIS-DICITOXOSIDE ( 2 SUCARS ) * DIGOX1CENIN M0N0-DIGIT0X0S1DE ( I SUGAR ) •DICOXICENIN-DIHYDR0D1C0XIN EFIDIGOX1CEN1N-DIHYDRODIGOXIGENIN •COfUDCATTON PRODUCTS F i g u r e 2: METABOLIC PATHWAY OF DIGOXIN *- CARDIOACTIVE Of the 100 patients receiving digoxin (56), seven subjects excreted more than 35% of digoxin in the urine as dihydrodigcxin and other dihydro derivatives. Approxinetely 10% of the drug was excreted in the urine as water-soluble metabolites. A mean of 15% was excreted in the stool, primarily as active as well as inactive m e t a b o l i t e s . In other studies, large amounts of the digoxin reduction 10 products or DRP were found i n one subject (DRP greater than 35-40% of total urinary digoxin and i t s metabolites) (56,57,58). Since antibiotic therapy hinders formation of the DRP, they appear to be made exclusively by bacteria in the gastrointestinal tract (59). The generation, by intestinal bacteria, of large amounts of cardioinactive metabolites of digoxin with a reduced lactone ring (DRP) i n some patients was associated with resistance to therapy and inappropriately high digoxin requirements (55). 7A recent report (60) has expanded the determination of digoxin metabolism using t r i t i a t e d digoxin i n humans. Separation of the drug and i t s metabolites with subsequent RIA analysis of the collected fractions indicated that digoxin was excreted i n urine as unchanged drug (55%) , digoxigenin bis-digitoxoside (2%) , digoxigenin mono-digitoxoside (0.8%), digoxigenin (0.25%) and dihydrodigoxin (0.3%). Glomerular f i l t r a t i o n i s by far the major mechanism of excretion, although a small amount i s subjected to both tubular secretion and reabsorption. The h a l f - l i f e of elimination i s about 1.5 days and in renal failure this value may increase up to 5 days (22). The variation in the reported amounts of the metabolites had led the investigators in the f i e l d to also suggest day-to-day fluctuations of the patients' digoxin levels (61); i n addition the d i f f i c u l t y of separating toxic from nontoxic states i n the usual setting (62) are not convincingly explained and required further 11 investigation of the analytical picture as compared to the c l i n i c a l response. 1.2 Plasma digoxin levels: Digoxin i s given orally whenever possible; this i s the most convenient and the safest route. The intravenous route i s preferred only when minutes may mean the difference between l i f e and death, as in patients with pulmonary edema from acute l e f t ventricular failure. The intravenous route, however, i s by far the most dangerous and i t s use may be rapidly f a t a l . In several studies i t has been demonstrated that very l i t t l e of the v a r i a b i l i t y in plasma digoxin concentration can be accounted for by consideration of factors such as dose, age, weight, etc. Thus, Wagner et a l . (43) found that only 34% of the v a r i a b i l i t y could be accounted for, by considering age, height, dose, body weight and renal function i n 25 patients. I t i s clear that other variables need to be considered eg., patient compliance, variations in both absorption, metabolism, and perhaps the analytical procedure used. Some workers have devised normograms or equations based on knowledge of the pharmacokinetic behavior of digoxin to determine the dose which w i l l result i n a given plasma digoxin concentration. These may, however, be unreliable and should be used with great circumspection and regard to the condition of the individual patient, rather than for the average values of pharmacokinetic variables of a population (63,64). 12 Digoxin plasma levels observed during chronic therapy with maintenance doses of 0.1-0.5 mg/day may range between 0.5 and 8.2 ng/mL (65,66). Therapeutic levels are considered to range between 1 and 2.5 ng/mL in normokalemic patients. Toxic symptoms such as gastric disturbances, headache, anorexia, loss of visual acuity, weakness, fatigue and mental confusion as well as cardiac symptoms are usually associated with levels greater than 3.0 ng/mL (67). Serum digoxin levels are helpful i n patient management since digoxin has been reported to have a low theraoeutic index (1,68). The determination of the optimal dosage in c l i n i c a l practice i s d i f f i c u l t to quantitate. It i s also often d i f f i c u l t to distinguish the endpoints of daily digoxin therapy both for evaluation of the desired therapeutic effect and for the possible sign of toxicity. Variations i n absorption, distribution and renal and non-renal elimination of digoxin preclude accurate prediction of serum levels from a given dose. For these reasons, determination of serum digoxin levels can be valuable i n t i t r a t i n g dosages. However, i t i s important to recognize that the serum digoxin level i s only one of the c r i t e r i a which a c l i n i c i a n should consider when making c l i n i c a l ,j decisions. A patient's response to a drug as reflected by a serum level depends on a number of factors such as: age, acid-base balance, electrolyte balance, thyroid disease, renal impairment and the presence of other drugs. These may alter the sensitivity to digoxin and the level at which toxicity occurs. 13 Elderly patients are particularly prone to toxicity because of change i n body composition. Lean body mass and creatinine clearance are reduced i n elderly patients, thus allowing for a small volume of distribution and hence reduced excretion rate of the drug (69). A recent study revealed that the majority of c l i n i c a l l y intoxicated geriatric patients had plasma levels within the therapeutic range. Because they might be unusually sensitive to the drug, the investigator concluded that the plasma digoxin level alone seems to be of limited value in screening elderly patients for intoxication and i t has to be considered together with the entire c l i n i c a l context (70). Electrolyte imbalances, particularly those i n hypokalemia are important sensitizing agents to the myocardium. Hence, digoxin intoxication may appear at normal serum glycoside levels i n patients with hypokalemia. Thyroid disease may also affect the response to digoxin (71). The lower serum digoxin concentrations i n hyperthyroid patients and higher serum concentrations in hypothyroid patients have been reported to be due to altered distribution volume (72-74), changes i n renal excretion due to alterations i n glomerular f i l t r a t i o n rate (71), malabsorption (75,76), enhanced b i l i a r y excretion (75) and increased tissue uptake (77) of digoxin. Since a definite therapeutic range for digoxin i s not known for patients with thyroid dysfunction (78), frequent monitoring of plasma levels may avoid the over-dosage in hypothyroid patients and obvious subtherapeutic dosing i n hyperthyroid patients. The association of renal impairment, elevated serum digoxin 14 concentrations and evidence of toxicity i s well documented. However, the relationship of the dysfunction and pharmacokinetic parameters that accompany renal impairment and alter digoxin disposition are not well delineated. Several studies on variables to predict serum concentrations based on renal function or creatinine clearance have riot been encouraging. When adjustment i s made for the altered distribution volume associated with renal impairment and indices of renal function are measured, a better predictability of serum digoxin concentration i s observed (37,79,80). For this reason, the volume of distribution acts as an i n i t i a l guide to the extent of the expected reduction i n dose, and the dose subsequently can be readjusted according to the patient's c l i n i c a l response using the plasma digoxin concentration as guide. It must be remembered, however, that the changes i n volume of distribution i n renal failure patients may also be the consequence of the insensitivity of assay methodology to true digoxin. In fact, Gibson and Nelson (81) found that digoxin metabolites in renal failure patients may result i n the over-estimation of digoxin serum concentration by 5%-42% when the radioimmunoassay technique was u t i l i z e d . Certain agents have also been shown to modify the level of plasma digoxin. Digoxin disposition may be altered by changes in gastrointestinal absorption, tissue binding and body clearance by some drugs. For instance, renal clearance of digoxin has been shown to be diminished by spironolactone (39,83). However, spironolactone also interferes with serum drug immunoassay measurements by cross-reacting with the digoxin antibody. The nonspecificity of the 15 RIA ki t s for spironolactone or i t s metabolites may result i n reporting of falsely elevated serum digoxin concentration (84,85) . The interaction of quinidine and digoxin, on the other hand, appears complex. Quinidine has been reported (150) to increase average digoxin concentrations from 0.9 ng/mL to 1.6 ng/mL. This interaction has since been confirmed by others, but i t s mechanism(s) i s s t i l l not f u l l y understood. It may be due to diminished distribution of digoxin to the tissue (151,152) as well as reduced renal clearance (151) resulting from inhibition of tubular secretion (153). The non-renal elimination of digoxin i s also reduced by quinidine (152). The c l i n i c a l significance of the interaction between digoxin and quinidine includes a greater risk of digoxin toxicity (86,87). Because of the inter-patient variation to quinidine's effect, frequent digoxin monitoring i s advised. Several other drugs which are co-administered with digoxin may also increase i t s serum concentrations. They include verapamil (154,155) (which increase the concentrations by 70-80%), nifedipine (+45%), amiodarone (+70%) (156) and disopyramide (15%) (157). How these interactions are produced i s not clear. In one recent study (59), a five day course of erythromycin or tetracycline was found to raise the serum drug concentration by 43-116% in 3 volunteers who were known to reduce the drug to dihydrodigoxin. Though the mechanisms of these interactions with digoxin remain to be delineated, appropriate dosage adjustments can be made from the measurement of the plasma digoxin concentrations. 16 1-3 Method of analysis: 1.3.1 Chemical methods: Direct measurement by ultraviolet spectroscopy i s d i f f i c u l t for digoxin because the absorption maximum i s at 217 nm ( 16,595 - molar extinction coefficient) (88). For quantitative analytical determination i t must therefore be converted into more intensely absorbing derivatives. _ BAUIT (picric acid) A KIDDC (dinitrobenzoic acid) 2-25 ua f RAVITZSCM (tetrarutrobiphenyl) \ Fluonmetry / (hydrochlonc acid, < 20 nj/ml trichloracetic acid) HO) ' OH 1 KELIM-KIUANI reaction <5Mg'IOm! Xanthydrol reaction F i g . 3 Chemical procedures for the photometric and fluorimetric determination of car-denolides However, a l l chemical methods are liable to f a i l i n the presence of large quantities of other substances and furthermore, because of their low sensitivity, they are seldom suitable for the estimation o f cardioactive steroids in biologic samples. The chemical 17 procedures for the photometric and fluorometric determination of cardenolides are summarized i n Figure 3. The reactions between the lactone side chain and the polynitroaromatic derivatives i n alkaline solution - picric acid (89), 1,3-dinitrobenzene (90), 3,5-dinitrobenzoic acid (91) and tetranitrodiphenyl (92) - are based on the fact that C-C coupling of the unsaturated lactone ring with nitrated aromatic derivatives produces dye complexes which can be measured photometrically. The specificity of these reagents i s low because chemical groups such as ketones also give an intense color. Both the Keller-Kiliani (93,94) and the Xanthydrol reactions (95) convert 2-desoxy sugars into characteristic colored derivatives. In this way a l l digitoxose-containing glycosides can be quantitatively determined. Owing to i t s high specificity for the sugar component of the glycoside, the reaction i s scarcely influenced by the nature of the genin. The reaction between digoxin and strong acids gives a lower limit of detection in the nanogram range using fluorescence spectroscopy. It was suggested by J e l l i f f e (96) that when digoxin i s exposed to phosphoric acid, concentrated hydrochloric acid or trichloroacetic acid, the glycoside i s f i r s t hydrolyzed to the genin and this w i l l subsequently form the 14-anhydrogenin. The glycoside i s further dehydrated to form the 14,16-dianhydrogenin. The chemical methods so far described are nonspecific and glycosides accompanied by other substances or glycoside mixtures 18 have to be submitted to preliminary chromatographic separation or purification. Tschesche et a l . (97) studied the sensitivity of various procedures. The test with polynitroaromatic derivatives i s of much the same sensitivity as the Keller-Kiliani reactions, while the Baijet test i s five times more sensitive, and the Xanthydrol reaction seven times more sensitive. The fluorogenic reagent (HC1/H202 or trichloroacetic acid/chloramine or hydrochloric acid (HCl)), i s suitable for direct use as a spray reagent for fluorometric detection of digoxin on chromatographic plates (96,98). The sensitivity limit i s 10-20 ng per application. The f i r s t c l i n i c a l l y applicable technique for chemical measurement of digoxin i n urine was presented by J e l l i f f e (99). As digoxin i s present i n blood levels of approximately 0.5-2 ng/mL, direct determination by this technique i s not feasible and the method could only be used after elaborate concentration of correspondingly large serum volumes. 1.3.2 Radioimmunoassay (RIA): The original digoxin RIA method (100) employed 3H-digoxin. Immunologically this represented the ideal tracer due to i t s chemical identity with the unlabeled digoxin. However, due to low specific activity, long counting times were required. Since this time, many RIA kits using 1 2 5 I labeled digoxin have become commercially available. Because of the speed and ease of performing 19 the RIA technique, combined with i t s s e n s i t i v i t y , the RIA method i s the p r i n c i p a l method employed i n current references and h o s p i t a l l a b o r a t o r i e s . In the RIA assay, a small amount of the patient's serum i s mixed i n a buffer with r a d i o a c t i v e digoxin (tracer) (Figure 4) and d i g o x i n - s p e c i f i c antibody. The antibody binds the tra c e r i n inverse proportion to the concentration of unlabeled digoxin present i n the patient's serum. The tra c e r and the unlabeled digoxin are assumed to have equal a f f i n i t y f o r the antibody binding s i t e s . The unbound ra d i o a c t i v e digoxin i s then separated and the r a d i o a c t i v i t y of the bound t r a c e r i s counted i n a s c i n t i l l a t i o n instrument. Percent digoxin bound i s converted to digoxin concentrations by reference to a standard curve prepared at the same time. DIGOXIN (PATIENTS) DIGOXIN (RADIOACTIVE) + ANTIBODY DIGOXIN-ANTIBODY COMPLEX DIGOXIN-ANTIBODY (RADIOACTIVE COMPLEX) Figure 4 : THE PRINCIPLE OF THE RADIOIMMUNOASSAY METHOD Several studies have focused on the considerable discrepancies found i n the values measured by the k i t s from d i f f e r e n t manufacturers (101-103). One study (38) showed a d i f f e r e n c e i n accuracy between four digoxin RIA k i t s of as much as 30% at the proposed t o x i c concentration of 2ng/mL. The observed d i f f e r e n c e s 20 might be due in part to variations i n digoxin content of the standards supplied (102,105,106). Additionally, samples with subnormal serum albumin concentrations influence the accuracy of digoxin recovery (107,108). Recently, another study reported differences in digoxin concentrations i n samples from patients with acute renal failure when the samples were measured by different immunoassays (109). Furthermore, another contribution to differences in the measurement by the use of RIA kits i s poor specificity of the antibody. Cross-reactivity between a digoxin antibody and the primary metabolites, digoxigenin bis-digitoxoside and digoxigenin mono-digitoxoside was demonstrated by Stoll et a l . (110). If these metabolites are eliminated in the urine i t might be expected that, they could accumulate i n the plasma of patients with renal failure, hence the apparent digoxin concentration as measured by standard I RIA would be higher than expected. The extent of cross reactivity of the digoxin antibody with dihydrodigoxin depends on the manufacturing procedure of the radioimmunoassay k i t ranging from insignificant cross-reactivity (111,112) to as great as 30% (60,113-115). Sensitivity of the RIA assay i s generally between 0.2 and 0.4 ng/mL. A serum sample from a patient not receiving digoxin may be reported as containing 0.4 ng/mL of digoxin (84). The levels found in this group could not be explained and are probably caused by the discrepancies in the RIA antibodies. Spironolactone, a potassium-sparing diuretic with a steroid-like structure i s commonly co-administered with digoxin. 21 Interference by spironolactone with digoxin radioimmunoassay has been well documented, however, conflicting results regarding the interference were found by these workers (82,84). Silber et a l . (82) suggested that neither spironolactone nor a metabolite, canrenone, should be used to determine assay interference, since unidentified metabolites, and not the parent compound or canrenone, are responsible"for the falsely high value's. /Anomalous serum digoxin concentrations using radioimmunoassay were also reported i n patients with renal impairment (109,117,118). Graves et a l . (118) suggested that a substance with digoxin-like immunoactivity i s produced endogenously i n the serum of digcxin-free patients with renal insufficiency. They also found inter-assay v a r i a b i l i t y i n this population and concluded that digoxin concentration measurements by current methods should be considered as questionable. False-positive digoxin values were also found i n newborn infants (119-121). Several commonly used commercial radioimmunoassays were found to measure some "apparent digoxin" values in a number of normal infants not receiving digoxin. Values f a l l i n g well into the therapeutic range have been measured (120,121). These results cast doubts on the true, chemically useful serum concentrations as determined by the current RIA method. An homogenous enzyme immunoassay (EMIT, or enzyme multiplied immunoassay technique) method has been employed for digoxin assay (122,123). The procedure shows sufficient sensitivity of 0.5 ng 22 digoxin /mL serum. The EMIT assay does not require separation steps, makes use of UV detection and may be readily automated. However, the incomplete inhibition of the G6-PD digoxin, upon binding, represents a problem, as do potential interferences with the enzymatic reaction by serum components. The cross-reactivity of" the EMIT digoxin antibody with digoxin metabolites has been reported recently (124). These investigators found no cross-reactivity between dihydrodigoxin from two different lots, but extensive interference with the hydrolysis metabolites: digoxigenin, digoxigenin rnono-digitoxoside and digoxigenin bis-digitoxoside. They suggested the EMIT method be used only for the accurate measurements of urinary digoxin. Another immunoassay method u t i l i z e s heterogenous enzyme-linked immunosorbent or ELISA (125). The advantage of this procedure over EMIT lies i n the complete differentiation of bound and free enzyme activity, however, the required separation step i s a disadvantage of ELISA. Although the sensitivity i s 0.3 ng/mL serum and intra-assay precision i s good (8-14%), steroid-like structures such as spironolactone interfere with the assay method. 1.3.3 Biological assays: The existence of a Na-K activated ATPase (Mg+2-dependent, Na-K activated ATP-phosphohydrolase) was discovered by Skou in 1957. Soon afterwards he also found that cardiac glycosides are able to inhibit ATPase activity by binding to the enzyme (126). Since some metabolites of digoxin are able to inhibit ATPase, this method lacks specificity. However, with respect to the minor extent to which the 2 3 metabolites often occur i n plasma (127), the result obtained was said to be comparable to the RIA method. Other disadvantages of this assay are that the sensitivity strongly depends on the respective enzyme preparation, and that the preparation of the enzyme i s a crucial step. Belz et a l . (128) claimed that the discontinous displacement of 3H-Ouabain and the: steep increase of unbound 'H-ouabain in the therapeutic range make i t d i f f i c u l t to use the displacement assay for routine laboratory determination. Rubidium-86 uptake by red blocd cells i s dependent on the inhibition by cardiac glycoside of the 86Rb flux across the human red c e l l in vit r o . The dichloromethane extraction i s a crucial step i n the method since only cardiac glycosides and their metabolites are extracted, and not highly polar compounds such as glucuronides. If these substances are s t i l l cardioactive, and there are arguments that i t could be so (129), the method would lead to a marked under-estimation of the cardioactive fraction in plasma. In addition, the performance of the complete ^ Rb analysis for plasma i s much more time-consuming than RIA methods (130). 1.3.4 Chromatographic methods: Watson and Kalman (131) developed a gas chromatographic (GC) assay for digoxin i n human plasma based on the work by J e l l i f f e and Blankenhorn (132). This method involved the formation of the heptafluorobutyrate (HFBA) derivative of digoxin and detected by electron capture (ECD). However, they required 10 mL plasma and had to carry out elaborate processing i n order to obtain the derivative 24 i n a state suitable for injection. The derivatization step converts digoxin and i t s known metabolites to digoxigenin which would then not be distinguishable from each other. In a subsequent paper Watson et a l . (133) combined a thin layer chromatographic (TLC) isolation of digoxin from i t s metabolites with subsequent detection of the HFBA derivative by GC-ECD. Unfortunately the method, although sensitive, involves a prior elaborate TLC preparative step. Eichorst and Hinderling (134) developed a TLC method for the measurement of radiolabeled digoxin and i t s known apolar metabolites in plasma, urine and saliva after single-dose administration of labeled digoxin. However, this method i s less than ideal for routine work because i t i s time consuming and requires administration of labeled digoxin. Combined GLC-Mass spectrometry (GLC-MS) (54) has also been reported. This method, as the previous GLC method, would not be suitable for routine plasma sample assay because i t i s only applicable to the aglycone portion of the molecule. Glycoside separations by high performance liquid chromatography (HPLC) are carried out on various types of column phases including s i l i c a gel (2, 135) ion exchange (136) and reversed-phase (RP) columns (eg. 135-138). Reversed-phase columns are gaining wide acceptance in the separation of cardiac glycosides because of their almost universal applicability and their insensitivity towards polar contaminants. The separation i s usually followed by UV detection at 220 nm and yields a lower limit of detection of 10 ng/injection (137). 25 Various attempts have been made to increase the detection sensitivity. For example, Natchmann et a l . (2) proposed the separation and quantitation of d i g i t a l i s glycosides by HPLC following pre-derivatization with 4-nitrobenzoyl chloride. UV detection at 254 nm permitted the detection of 5 ng/mL digoxin with 3:1 signal to noise ratio. Recently, F u j i i et a l . (3) described the separation and quantitation of 3,5-dinitrobenzoyl derivatives of cardiac glycosides and their metabolites for enhancing ultraviolet (UV) detectability by micro high-performance liquid chromatography (MHPLC). A detection wavelength of 230 nm was used to enhance the sensitivity and the limit of quantitative measurement of digoxin was less than 0.6 ng (signal to noise ratio 3/1). The application of this potential method i n digitalized patients was not examined. 1.3.5 HPLC-RIA: To overcome the problem of detector sensitivity Loo et a l . (4) combined the resolving power of adsorption HPLC with the sensitivity of radioimmunoassay and applied the resulting technique (HPLC-RIA) to determine serum levels of digoxigenin i n human subjects. This method, although sensitive, involves the administration of t r i t i a t e d digoxin to the subjects, and therefore i s not readily adaptable for routine monitoring of serum samples from digitalized patients. An even more ambitious HPLC-RIA method was employed by Nelson et a l . (5) to quantify digoxin and i t s metabolites in plasma. The method employed "spiking" with 3H-digoxin for recovery monitoring, dichloromethane extraction, HPLC separation of the residue on a 26 reversed-phase column using a gradient solvent system, and quantitation of the collected fraction by RIA. Gault et a l . (6) on the other hand, administered t r i t i a t e d digoxin to volunteers, extracted radioactive digoxin and i t s metabolites from urine, separated them by RP-HPLC and used UV to detect the effluent prior to quantitating them by s c i n t i l l a t i o n counting. The method described was able to detect three new extractable metabolites. Another HPLC-RIA also involving administration of t r i t i a t e d digoxin but with higher extraction recovery was also reported (140). Because these methods involved the administration of t r i t i a t e d digoxin they w i l l therefore have the same disadvantage as the method by Loo (4). Recently, Loo et a l . (7) modified the method of Nelson et a l . (5), by simplifying the extraction steps using a trace enrichment technique developed by Schauecker et a l . (139). Another HPLC-RIA method (8) employed a combined internal standard and marker to follow the elution of digoxin and i t s metabolites. The latter method does not use labeled digoxin i n the c l i n i c a l samples u n t i l the usual quantitation by 3H-RIA. 1.3.6 Post-column derivatization: Post-column derivatization i n HPLC offers a rather different approach to analysis than pre-chromatographic derivatization. F i r s t l y , the compounds of interest are chromatographically separated before derivatization takes place. The advantages are that artifact formation i s not c r i t i c a l and that the reaction does not have to go to completion or give well-defined derivatives— provided that i t i s 27 reproducible. The kinetics of the reaction are very important and determine the type of reaction detector to use. There are three different principles for the construction of such reactors: (i) tubular or capillary reactors, (ii) bed reactor and ( i i i ) a i r segmentation streams (104). The tubular reactors consist of a narrow tube usually 0.2mm-2.0mm ID through which the effluent-reagent mixture flows i n the required reaction time. I t i s the simplest reactor and i s recommended for fast reactions taking less than 30 sec.(116). Packed bed reactors consist of a tube f i l l e d with granular inert material. This type of reactor i s used with intermediate kinetics requiring actual reaction times in the reactor from one-half to several minutes (116). It can be looked at as a chromatographic column under non-retention conditions. Segmented flow reactors are based on the segmentation of flowing liquid streams with a i r bubbles introduced into the stream at certain time intervals. The goal i s to minimize band broadening. It requires a debubbling device between the reactor and detector. This approach i s very efficient and has become of more interest for post-column derivatization with relatively slow reactions (more than 5 minutes). In recent years, this analytical technique has attracted enhanced attention. One of the reasons for this attention i s the current lack of suitable detectors for trace analysis i n biological samples. The use of post-column derivatization can often improve the 28 desired detection sensitivity and the specificity of an HPLC method for specific trace analysis. Gfeller et a l . (9) have reported the merits of post-column detection of cardiac glycosides with hydrochloric acid. The fluorogenic reaction was carried out in a modified Technicon Autoanalyzer system which employed the ai r segmentation principle and was based on the reaction of HC1 with the steroid portion of the cardiac glycoside. Fluorescence of the derivatives was further enhanced by the addition of a hydrogen peroxide/ascorbic acid mixture. It has been found possible to detect amounts as small as 500 pg/2 u l injection of desacetyllanatoside C. However, Gfeller and his co-workers (9) did not report a minimum detectable quantity for digoxin. 29 2. EXPERIMENTAL: 2.1 Supplies 2.1.1 Chemicals Digoxin, digoxigenin bisdigitoxoside, digoxigenin monodigitoxoside, digoxigenin, dihydrodigoxin, dihydrodigoxigenin and digitoxigenin were obtained from Boehringer (Mannheim, GFR) and were used without further purification. Spironolactone, furosemide and quinidine were purchased from Sigma Chemical Company (St. Louis, Mo., USA). Disopyramide was purchased from Roussel (London, England) whilst procainamide HC1 was obtained from Squibb (Montreal, Que., Canada). Other drugs such as Bactrim DS, Capoten, Colace, Isoptin, Persantine and Rythmonorm were obtained from local suppliers. 2.1.2 Reagents A l l derivatization reagents for gas chromatography were purchased from Pierce Chemical Co., (Rockford, I I . , USA). Dansyl hydrazine was obtained from Regis Chemical Co., (Morton Grove, I I . , USA). Aluminum (Al) chloride, glycinamide, isonicotinylhydrazine (INH) and potassium ferricyanide were commercially available compounds and used as supplied. Hydrogen peroxide (H2O2) (30%) was purchased from American Scientific and Chemical (Portland, Or., USA) whilst hydrochloric acid (HCl) (37-38%) was obtained from Fisher Scientific Co.(Fair Lawn, N.J., USA). L-Ascorbic acid was BDH 30 laboratory reagent grade from BDH Chemicals (Toronto, Ont., Canada). Other common reagents were obtained from various suppliers. 2.1.3 Solvents Water, methanol, isopropanol, n-propanol and dichloromethane were Fisher Scientific Co. HPLC quality ( Fair Lawn, N.J., USA). "Glass d i s t i l l e d " quality 2,2,4-trimethylpentane (isooctane) was obtained from Burdick and Jackson Laboratories Inc., (Muskegon, Mi., USA). Acetone and absolute ethanol were reagent grade quality obtained from commercial sources and used as received. 2.1.4 Supplies for extractions The f i l t e r unit consisted of a Nylon 66 membrane (0.45um, 13 mm diameter) f i l t e r disc (Rainin Instrument Co., Inc. Woburn, Ma., USA) housed i n a Swinnex 13 Millipore f i l t e r holder (Millipore Corp. Belford Ma., USA). This unit was used on the end of a Luer-Lock 5cc M u l t i f i t B-D glass syringe (Becton Dickinson Canada, Mississauga, Ont., Canada). The solid-phase cartridges Bond-Elut (Si, diol) were purchased from Analytichem International, Inc (Harbor City, Ca., USA) . 2.1.5 Post-column derivatization supplies 2.1.5.1 Packed-bed reactor The packed-bed reactor consisted of a 15 cm X 1.5 cm I.D. glass column (part # 6112, Omnifit Ltd., Cambridge, England) dry 31 packed with 20 um glass beads (a generous g i f t from Dow Corning Corp., Midland, Mi., USA). 2.1.5.2 Air-segmented reactor The mixing and reaction spirals were essentially the same as described previously (9) except for the mixing tees and debubbler which were miniaturized using 1 mm I.D. quartz tubing. The mixing tees were constructed such that the reagent solutions entered at a 30°angle against the eluent stream. The debubbler was a miniaturized version of the Technicon C-5 debubbler. 2.1.5.3 Non-segmented reactor The manifold was made from Solvaflex tubing (part # 116-0533P04, .015" I.D. orange/green collar, Technicon Instruments Corp., Tarrytown, N.Y., USA) for the delivery of dehydroascorbic acid and acidflex tubing (part # 116-0538-09 Technicon) for the delivery of HC1. A short stainless steel tube, inserted into the Solvaflex tubing, served as a connection between tubings. Polytetrafluoroethylene (PTFE) tubing of 0.3mm I.D. was used for the reactor and the connection between column and reactor to minimize sample band broadening. A 0.3mm I.D. PTFE tubing was also used as the restriction c o i l to provide increased resistance. A 0.8mm I.D. tubing was used for the mixing c o i l and connection between delivery tubings and the mixing c o i l . A three-way PTFE connector (part # 1004 Omnifit Ltd., Cambridge, England) was likewise placed in the solvent delivery 32 lines from the reagents to the mixing c o i l . A 3-way PTFE valve (part # 1102 Omnifit Ltd., Cambridge, England) was placed between the column, the mixing c o i l and the reactor. Closing of the PTFE valve between the column, reactor and the mixing c o i l allowed for cleaning of the reactor system at the end of the day. A l l valves, connectors and tubings were manufactured by Omnifit Ltd. (Cambridge, England). 2.2 Equipment 2.2.1 High-performance liquid chromatographic (HPLC) system The HPLC system consisted of a Beckman Model 100 A solvent metering system (Beckman Instrument, Inc., Fullerton, Ca., USA) equipped with a model U6K injector (Waters Assoc., Milford, Ma., USA), or a Rheodyne Model 7125 100 uL fixed loop injector (Rheodyne, Berkeley, Ca., USA), a Waters fluorescence detector model•420 AC with a homemade (1 mm ID X 40 mm length quartz tubing) flow c e l l (section 2.2.2) and an Altex CRIA Chromatopac Data Processor (Beckman Instrument, Inc.). A Schoeffel model FS-970 LC fluorometer (Acton, Ma., USA) was also used i n this work for detector sensitivity assessment. The excitation and.emission conditions were chosen as optimum for the determination of digoxin. A noise f i l t e r 33 constructed i n the laboratory (Figure 5) was coupled between the detector and integrator. To study band broadening, a variable wavelength ultraviolet (UV) detector (Beckman model 155 detector, Beckman Instrument Inc.) was placed between the post-column detection system and the analytical column. Band broadening of the post-column reactor was measured as the increase i n peak width i n minutes relative to the UV signal. 2.2.2 Flow c e l l The Schoeffel detector titanium c e l l was manufactured to the exact specifications as the standard stainless steel c e l l for the FS 970 fluorometer. The Waters detector flow c e l l block was modified to eliminate a l l stainless steel f i t t i n g s . An all-quartz flow c e l l of 40 mm in length and 1 mm ID was f i t t e d with acidflex tubing on both ends and this c e l l was held in place by epoxy glue on the c e l l block. 2.2.3 Post-column reaction system A schematic diagram of the instrumental arrangement of the segmented-flow system i s given i n Figure 6. Air bubbles and the chemical reagents were introduced via a Technicon proportioning pump (Technicon Instruments Corp.). A Manostat Casette pump (Fisher 34 1 0 0 , 0 0 0 n A A A / W 100,000 n A A A A / V + T 2700 £7 w - i w v 2700 V O U t - I V X 202,700 z i o m V igure 5. Schemat ic of the Noise Filter 35 m a t e r i a l / c o l o r c o d e H P L C c o l u m n I 0.4 m l / m i n 20x 120 * 0.1 c m 20 x 2 4 0 x 0.1 c m w a s t e -20* '•• - [ D 2 02/ A / w h i t e D2V V / o r a n g e - g r e e n v / o r a n g e - b l u e r e a c t i o n co i l 4 0 x 1 0 0 0 x 0.1 c m 4 5 " C 30 x 180 x 0.1 c m w a s t e A / w h i t e I E x 3 5 0 n m E m . cu t off 415 n m * pump K2> f l o w r a t e ( m l / m i n ) 0 5 3 0.10 0 . 0 5 h y d r o c h l o r i c a c i d a i r " d e h y d r o a s c o r b i c a c i d * T E C H N I C O N - p u m p w i t h a i r lock. 0 5 3 A = A c i f l e x V = V i n y l r e c o r d e r / i n t e g r a t o r f l u o r i m e t e r Figure6. Schematic diagram of the reaction detector unit for fluorescence detection9. Scientific Ltd.) with variable speed was used to supply the reagent solution for the packed bed reactor and non-segmented reactor. For a l l three reactors, the reaction bath was thermostated using a Haake model E51 constant temperature circulator supplied by Fisher Scientific Co.(Fair Lawn, N.J., USA). Two nitrogen pressurized bottles (1 l i t e r glass reservoir, Omnifit Ltd.) were also used to deliver the reagents (HC1 and dehydroascorbic acid) to the reactor as an alternative to the peristaltic pump. 2.2.4 Extraction A Vortex-Genie purchased from Fisher Scientific Co. was used to aid i n the precipitation of proteins i n plasma following addition of acetone, as well as to wash the supernatant with isooctane. A Fisher Roto-Rack model 343 was used i n the extraction step and f i n a l l y an IEC HN-SII Centrifuge, (Damon/IEC Division, Western Scientific, Vancouver, B.C., Canada) was used to separate the immiscible phases. 2.2.5 Gas chromatography A Hewlett-Packard model 5880 A Series reporting gas chromatograph equipped with a Nickel-63 electron-capture detector (ECD) (Avondale, Pa., USA), and an on-column injector was used for GC analyses. The injection i s made using a standard microliter 37 syringe with a fused s i l i c a needle (Hamilton Co., Reno, Nev., USA). A carbowax 20 M capillary column (20m X 0.2 mm ID) was used for the separation. The oven temperature was raised from 50°JC to 250°C at a rate of 30°C/min while maintaining helium flow of 1 mL/min through the column. I n i t i a l temperature was maintained for 2 minutes before the temperature program. Other conditions were similar to the procedure by Watson et a l . (133). 2.2.6 HPLC-Mass Spectrometry An unmodified Hewlett-Packard model 1082 B liquid chromatograph was coupled to a model 5987 A GC/MS instrument equipped with direct liquid introduction (DLI) interface (Hewlett Packard Co., Palo Alto, Ca., USA) for LC/MS operation. The normal chemical ionization (CI) operating parameters of the mass spectometer for the LC/MS experiments were as follows: electron voltage 230 V emission current 300 uA source temperature 250°C -4 ion source pressure 1.2 X10- torr as measured at the CI GC/MS interface thermocouple repeller 3 volt electron multiplier 2200 volts The HPLC eluent/CI reagent gas i n this work was 75/25 (by volume) methanol:water at a flow rate of 0.8 mL/min. 38 2.3 Stationary phases A3 cmX 2.1 mm ID direct connect guard column (Mandel Scientific Co. Ltd., Rockwood , Ont., Canada) containing dry-packed 25-37 urn CO:PELL ODS (octadecylsilane) pellicular media (Whatman Ltd., Clifton, N.J., USA) was placed prior to and i n series with the analytical column. The commercial columns used i n these experiments were: (a) Ultrasphere ODS (5u) 4.6mm X 25 cm (Beckman Instrument Inc.) (b) Microsorb Short-One HPLC column C-18 (3u) 4.6mm X 10cm (Rainin Instrument Co., Inc.) (c) Spherisorb ODS II (3u) 4.6mm X 15cm (Alltech Associates, Deerfield I I . , USA) Laboratory packed columns were as follows (on loan from Dr. Richard Wall, University of Br i t i s h Columbia, Faculty of Medicine, Dept. of Pathology, Vancouver, B.C., Canada): (a) Hypersil ODS (5u) 4.6mm X125mm (b) Lichrosorb RP 8 (lOu) 4.6mm X 125mm (c) Hypersil ODS (3u) 4.6mm X 60 mm 2.4 Derivatization procedures 2.4.1 For Gas Chromatography and Electron-Capture detection 39 2.4.1.1 Reagent: N-Methyl-bis-trifluoroacetamide (MBTFA) (141): About 0.5mg of digoxin or digoxigenin was reacted with 0.1 mL of MBTFA in dry pyridine for one hour at 65°C. A volume of 2 uL of organic phase was injected. 2.4.1.2 Reagent: T r i - S i l Z (a mixture of trimethylsilylimidazole in dry pyridine 1.5 mEq/mL) (142): A 1.0 mL sample was placed in a Reacti-vial (Pierce Chemical Co., Rockford, I I . , USA) and was mixed with 1 mL of T r i - S i l Z. The mixture was shaken occasionally u n t i l the sample dissolved. This was then heated at 60°C for 10 minutes. The mixture was injected directly into the gas chromatograph. 2.4.1.3 Reagent: Heptafluorobutyrylimidazole (HFBI) (143): A 20 uL aliquot of HFBI was added to 5 mg of digoxin. The mixture was capped and heated at 85°C for 1 hour. This was then shaken for 2 minutes with 2 mL of toluene and 0.5 mL of d i s t i l l e d water. The toluene layer was removed and injected into the gas chromatograph. 2.4.2 In-vitro fluorometric assay method Fluorescence spectra and intensity measurements were taken on an Aminco-Bowman Spectrophotometer (American Instrument Co., Inc. Silver Spring, Ma., USA) whose monochromators were calibrated 40 against the Xenon line emission spectrum and whose output was corrected for instrumental response by means of a quinine sulfate standard solution. 2.4.2.1 Sulfuric acid (145): An aliquot of 0.1 mg of digoxin i n 1 mL of chloroform was treated with 2 mL concentrated sulfuric acid (96%)and heated at 5zP C for 7 minutes. The fluorescence intensity was measured at room temperature at 420 nm emission, 390nm excitation. The background intensity from sulfuric acid and chloroform was also measured, and subtracted from the intensity obtained for the drug solution. 2.4.2.2 Acetic anhydride/ acetyl chloride/ trifluoroacetic acid (146): Solution A: An aliquot of 10 mL each of acetic anhydride and acetyl chloride were mixed i n a glass-stoppered bottle and kept in the refrigerator. Solution B (reagent): A 4 mL aliquot of solution A was pipetted into a 10 mL volumetric flask and the volume was made up to 10 mL with trifluoroacetic acid. Procedure: An aliquot of 0.1 mg of digoxin i n 1 mL chloroform was mixed with 2 mL of solution B. The solution was placed i n a water bath at 55°C for 7 minutes and was allowed to stand at room temperature before reading against the reagent blank. Excitation and emission intensity were set at 345/435 nm respectively. 41 2.4.2.3 Dansyl hydrazine: One mg of d i g i t a l i s glycoside sample was dissolved i n 1 mL of methanol to which a 2 fold molar excess (1.36 mg) of dansyl hydrazine and 2 drops of glacial acetic acid had been added. The solution was heated at 70°C for 15 minutes and was then evaporated to dryness under vacuum. The residue was dissolved i n 0.5 mL of HPLC mobile phase. 2.4.2.4 Hydrogen peroxide/ ascorbic acid/ hydrochloric acid (147): The assay was done i n accordance to the United States Pharmacopeial method for the digoxin tablet dissolution test. In this procedure 5 mg/mL digoxin solution was transferred to a glass-stoppered flask and 1.0 mL methanol was transferred to another flask to provide a blank. To this sample was added 1 mL of 2mg/mL ascorbic acid solution and the solution was then mixed. Concentrated HC1 (5 mL) was added immediately to this mixture, followed with 1 mL of .012%v/v hydrogen peroxide solution. The reaction medium was stoppered and stored at room temperature for 2 hours before the fluorescence intensity was read. 2.5 Liquid chromatography and ultraviolet detection (138) The glycosides were dissolved i n the mobile phase and injected into the liquid chromatograph. The retention time of each compound was determined by separate injections of individual 42 solutions of each glycoside. The mobile phase was prepared i n sufficient quantities for daily use and degassing was not found to be necessary. The mobile phase used i n these experiments was composed of water/ methanol/ isopropanol/ dichloromethane (47:40:9:4). The flow rate of the HPLC was maintained at 1.2 mL/min. 2.6 Preparation of reagents 2.6.1 Reagents 2.6.1.1 Glycinamide/ Potassium Ferricyanide These reagents were prepared according to the method reported by Seki and Yamaguchi(148) where 0.5 g of glycinamide and 30 mg of potassium ferricyanide are dissolved i n 0.3M borate buffer solution (pH 9.8). 2.6.1.2 Isonicotinylhydrazine (INH)- aluminum (Al) chloride solution These reagents were prepared according to the protocol reported by Horikawa et a l . (149). A 16 mM INH solution and an 80 mM aluminum chloride solution were prepared by dissolving the powders in methanol. The solutions were kept i n an aluminum f o i l - wrapped container and stored i n a dark place un t i l required for use. 2.6.1.3 Dehydroascorbic acid .. A stock solution of dehydroascorbic acid solution was prepared weekly and kept i n a refrigerator u n t i l required for use. 43 Solution A: A solution containing 250 mg of ascorbic acid was made up to 500 mL with d i s t i l l e d water i n a volumetric flask. Solution B: A 1 mL aliquot of hydrogen peroxide (30%) was transferred to a 200 mL volumetric flask and made up to volume with d i s t i l l e d water. The day prior to analysis, 100 mL of solution A and 2.5 mL of solution B were added dropwise to an Erlenmeyer flask, followed by s t i r r i n g for 2 hours. This solution was kept i n a refrigerator u n t i l needed. 2.6.2 Standard solutions 2.6.2.1 Digoxin standards About 1 mg of digoxin reference standard was accurately weighed using an electrobalance model G (Cahn Instrument Co., Paramount, Ca., USA) into a 100 mL volumetric flask and made up to volume with methanol. Aliquots of this solution were diluted with methanol to yield working standards i n the 1.5-10 ng/10 uL range. 2.6.2.2 Internal standard solutions Standard solutions weighing approximately 4 mg/mL of digitoxigenin and 1 mg/mL each of 17- methyltestosterone, ethinylestradiol, prednisone, 17-estradiol, norethindrone, ouabain, cymarin, triamcinolone, diethylstilbestrol, mestranol and testosterone were individually prepared and injected into the liquid chromatograph to determine the retention times of each. An internal standard that elutes near the digoxin peak but which does not co-elute with any of the metabolites and endogenous material was considered to be ideal for this study. For this purpose 44 digitoxigenin, which eluted 10 minutes after digoxin, was used as the internal standard. A stock solution of digitoxigenin was prepared i n methanol at a concentration of 4 mg/100 mL. To each of the digoxin working standards were added 2 mL of this internal standard solution. These standards were made up to 10 mL with methanol. These working standards with the internal standard were used for the calibration curve. Another 2 mL of the internal standard solution was also transferred to a 10 mL volumetric flask and made up to volume with methanol. Plasma samples were spiked with 10 uL of this diluted internal standard. 2.6.2.3 Digoxin metabolites One milligram of each of the metabolites of digoxin, namely digoxigenin, dihydrodigoxigenin, digoxigenin mono-digitoxoside, digoxigenin bis-digitoxoside and dihydrodigoxin, were accurately weighed and separately made up to 10 mL with methanol. An aliquot of 500 uL of each solution was transferred to a 10 mL volumetric flask and 10 uL of these were used for the specificity test of the HPLC post-column reaction. 2.6.2.4 Co-administered drugs: Aliquots of 3 mg of furosemide, spironolactone and quinidine were weighed separately and dissolved i n 10 mL methanol. An aliquot of 3 mg of procainamide HC1 was dissolved i n water, basified to pH 8 with sodium hydroxide solution and extracted twice with 10 mL of dichloromethane. The combined organic phases were evaporated to dryness. The residue was dissolved i n 10 mL of 45 methanol. An aliquot of 1 mg of disopyramide (base) was made up to 10 mL with methanol. Other drugs evaluated for fluorescent response using the post-column derivatization method included: Trimethoprim-sulfamethoxazole (Bactrim DS, Roche) Captopril (Capoten 100 mg tablets, Squibb) Docusate Sodium (Colace capsules, Bristol) Verapamil Hydrochloride (Isoptin Injectable 2.5 mg/mL, Searle) Dipyridamole (Persantine 75 mg tablets, Boehringer) Propafenone (Rythmonorm 300 mg tablets, Knoll). Tablets and capsules were extracted i n the manner described in the current Pharmacopoeiae. Extracts were usually evaporated and resuspended i n the mobile phase and analyzed subsequently. The injectable preparation of verapamil was analyzed directly without further processing. Amounts of each drug analyzed were i n microgram quantities, and hence were above plasma levels. The chromatogram of each drug examined was allowed to run for up to one hour to determine i t s elution profile. In a l l such experiments, 10 uL of each drug solution i n methanol was injected using a 25 uL syringe (Waters Associates 25 microliters, SGE Scientific glass engineering Ltd. Melbourne, Australia). 2.6.2.5 Mobile phase 46 The mobile phase was prepared by mixing the solvents and then degassing the composite by rapid s t i r r i n g for 30 minutes. The mobile phases examined for this study are summarized on Table 3 of the Results and Discussion section. 2.7 HPLC post-column derivatization procedure 2.7.1 Column selection Columns of different lengths and sizes of packing materials were examined individually by attaching columns enumerated i n section 2.3 to the optimized non-segmented post-column derivatization system. Samples of digoxin mixed with i t s metabolites were injected and the resolutions of these were compared. 2.7.2 Optimized HPLC non-segmented post-column derivatization system The HPLC post-column derivatization was carried out as shown i n Figure 7. The cardiac glycoside samples were injected via the Waters U6K universal injector. The glycosides were eluted from the 15cm, 3u ODS packed column equipped with an ODS guard column using an isocratic mobile phase composition; ethanol, methanol, isopropanol, water (3/52/1/45) at a flow rate of 0.3 mL/min. At the exit end of the column, the effluent was joined by the reagent line through a 0.8 mm ID PTFE tee. The post-column reagents (HC1 and dehydroascorbic acid solution) were delivered by a peri s t a l t i c pump at a rate of 0.23 ± .01 mL/min and premixed i n 2 m X 0.8 mm ID c o i l before joining the 47 Figure 7. Flow Diagram of tho Present Poet-Column Derivatization Method co HPLC MOBILE PHASE 1 I \ INJB ECTOR HPLC PUMP, 03mL/min PRESSURE gjj^L^TO WASTE 0.53mL/nin l OIL MIXING COIL 415 DEHYDROASCORBIC ACID •^ -CONC. HCI PERISTALTIC PUMP effluent. The combined mixture was immediately passed into the 10 m knitted reactor maintained at 79 + 1°C. The glycosides, i n the presence of HC1 and dehydroascorbic acid, reacted to form the fluorophore which was monitored by a Waters fluorescence detector equipped with a 360 nm excitation f i l t e r and a 425 nm emission f i l t e r . The total flow rate of the system was found to be 0.53 mL/min. The flow rates in the HPLC and reagent line were measured by collecting the effluent i n a graduated cylinder and timing with a stopwatch. Attached to the detector output was a homemade noise f i l t e r which was placed in series with the integrator. A back pressure c o i l of 1 m X 0.3 mm ID PTFE tubing was connectd to the fluorometer outlet to prevent bubble formation i n the detector c e l l . For constant proportioning of the peristaltic pump, the delivery tubes were stretched on the platen to avoid tube snaking and hence generating proportioning of equal pulsation. To maintain reproducible pumping rates for day-to-day operation, the acidflex tubing, which was found to stretch and flatten with time, was changed daily. After each analysis day, the HPLC column was thoroughly rinsed with methanol and the post-column reactor was l e f t dry. Baseline s t a b i l i t y was obtained after the f i r s t hour of operation of the system. 2.7.3 Other alternative reagents employed for derivati zation In an attempt to eliminate the more volatile HC1, other 49 acidic reagents such as sulfuric acid (70%), perchloric acid or trichloroacetic acid were substituted for the concentrated HC1 described above. A l l other post-column and HPLC parameters were maintained as before. Although several concentrations of sulfuric acid were tested (40, 60, 70 and 100%) 70% was chosen for i t s efficiency i n derivatization and appropriate viscosity. The 100% sulfuric acid was too viscous to pump through the small internal bore PTFE reactor. Procedures for the INH /Al and glycinamide / ferricyanide post-column derivatization reagents were mainly i n accordance with the methods described i n the literature (148,149). Since they were optimized for steroid determination, parameters such as the reagent flow rate and reaction temperature were modified to give the maximum response. The optimized post-column procedure using these reagents were as follows: Digoxin solution was injected into the HPLC and eluted at a flow rate of 0.3 mL/min using the mobile phase (methanol/ ethanol/ isopropanol/ water 53/3/1/45). The effluent from the HPLC was mixed with the fluorescent reagents (INH/Al or glycinamide/ ferricyanide) at a flow rate of 0.2 mL/min and the mixed solution was heated i n a PTFE coiled reactor (10 m X 0.3 mm) maintained i n a water bath at 65°and 70°C respectively. Fluorescence was measured and the intensity of the peaks recorded. The excitation/ emission wavelength for INH/Al and glycinamide/ 50 ferricyanide were 360/ 425nm and 334/ 395nm respectively. The derivatization using these two methods was found to be less efficient than the originally proposed method (procedure described in section 2.7.2). 2.7.4 Packed-bed reaction system The glycosides were injected into the chromatograph and eluted using the isocratic mobile phase described i n 2.7.2. The column effluent was joined by the reagents through a four-way PTFE tee. The mixing c o i l was omitted and the combined mixture was delivered directly into the packed bed reactor maintained at 79°C. A l l other parameters were similar to the non-segmented reaction detector. Although this reactor would seem to be ideal for the required reaction time proposed by Gfeller et al.(9), the ins t a b i l i t y of the beads for the packed-bed defeated the theory. 2.7.5 Air-segmented reaction system A schematic diagram of the KPLC post-column derivatization used i n this study i s shown i n Figure 6. The procedure used in this study was a modification of the system developed by Gfeller et a l . (9). Digoxin standard solutions were injected onto the reversed-phase column via the Waters Universal injector. The glycosides were eluted from the column using the mobile phase given i n 2.7.2 at a flow rate of 0.4 mL/min. At the exit end of the column, 0.5 mL/min 51 of concentrated HCl, segmented with 0.5 cm a i r bubbles, was added to the column effluent through a 1 mm ID mixing tee described in 2.1.5.2. This segmented solution was passed into a 20-turn mixing c o i l after which .05 mL/min dehydroascorbic acid solution was added through a second 1 mm ID mixing tee. The combined mixture was then introduced into a 40-turn mixing c o i l . The fluorophore was formed i n the 40-turn reaction c o i l maintained at 55°C in a water bath. The mixture was cooled to room temperature i n a 30-turn glass c o i l , jacketed with running tap water. The mixture was introduced at a 90° angle against the a i r bubble stream. The combined, cooled mixture was debubbled before entering the fluorescence detector. The position of the debubbler was found to be significant for the efficient removal of ai r segments. 2.7.6 Optimization procedure for post-column derivatization Before the post-column reaction conditions were studied, the HPLC mobile phase, octadecylsilane column, fluorescence detector and the reagent mixing tees were evaluated for maximum efficiency and detection sensitivity. Different HPLC mobile phases were examined to determine compatibility with the reaction medium, hence providing a maximum response. The 3u ODS column was selected as the optimum column packing as i t gave satisfactory separation of digoxin and i t s metabolites. A Waters fluorescence detector with a 40 mm X 1 52 mm ID flow c e l l and a noise f i l t e r were satisfactory for detection of nanogram quantities of digoxin. As the source lamp output decreased with age, the detector response was kept constant by increasing the voltage on the photomultiplier tube. The mixing tees were of minimal internal diameter to minimize band spreading. The 1 m PTFE tubing connected at the exit end of the detector enabled the temperature of the reactor heating bath to be increased, thereby increasing the rate of post-column reaction. 2.7.6.1 Relative proportions of fluorogenic reagents As i t appeared that the HCl-dehydroascorbic acid fluorogenic reagents gave maximum sensitivity, the relative proportions of these were further examined to further increase detector response. The HPLC flow rate was i n i t i a l l y maintained at 0.1 mL/min for subsequent experiments. Internal diameters of 0.02, 0.015 and 0.010" for dehydroascorbic acid delivery tubings and 0.29, 0.53 and 0.63" ID acidflex tubings for HC1 delivery were used to set the relative proportions of reagents for maximum fluorescence intensity studies. The other post-column reaction parameters described i n 2.7.2 were held constant. In addition, concentrations of hydrogen peroxide i n ascorbic acid and vice versa were also studied. The optimum concentration of ascorbic acid from .01 to 1% was i n i t i a l l y determined by maintaining the concentration of hydrogen peroxide at 53 1.1 X 10 M. An optimal concentration of ascorbic acid was found to be 0.1%. Then hydrogen peroxide concentrations from 4.4 X 10~4M to _3 2.2 X 10 M were also examined with the concentration of ascorbic acid at 0.1%. An optimum concentration of 1.1 X 10~3M hydrogen peroxide was observed and used in the subsequent experiments. Further changes i n the ratios of ascorbic acid / hydrogen peroxide did not increase fluorescence intensity of the derivative. 2.7.6.2 Reaction temperatures I n i t i a l experiments were carried out using digoxin alone and an HPLC mobile phase of methanol. Temperatures of 45° 50? 55° 57°. 59°and 64°C were used to determine the effect of temperature on the fluorescence intensity of digoxin. Response increased as the temperature increased. No data were obtained above 64°C (the boiling point of methanol). However, when the mobile phase described in 2.7.2 was used, the temperature could be further increased to 80°C, beyond this temperature the response remained constant. At about 80° C the reaction of the glycoside and the reagents was apparently complete. Therefore 79+1°C was selected as the overall optimum reaction temperature. 2.7.6.3 Reaction kinetics The effect of the flow rates of the reagents and flow rate 54 of the mobile phase on the reaction time, and consequently the detector response to the fluorophore formed, was investigated. This study was undertaken by varying the speed of the peristaltic pump which delivered the HC1 and dehydroascorbic acid. Flow rates of 0.24, 0.27, 0.33, 0.37, 0.41 and 0.43 mL/min lead to average peak heights of 1.1, 2.35, 2.27, 1.93, 1.85 and 1.70 cm respectively. Subsequent to this optimization,the length of the reaction coils was investigated. Coils of 10, 15, 20 and 25 m with 0.3 mm ID provided 3,5,7 and 8.6 minutes reaction times, respectively. In addition, different reactor geometries were also examined. They were: a 20 m length of 0.3 mm I.D. tubing with an outer c o i l diameter of 50 cm, a 20 m length, 0.3 mm I.D. tubing with an outer diameter of 2.7 cm ,a 10 m length c o i l of 0.3 mm I.D. tubing of 2.7 cm outer diameter and f i n a l l y a 10 m knitted reactor. The knitted reactor was composed of fringes like those used for crochet work. The results of this investigation showed that the 10 m ]cnitted reactor was found to give the least band broadening with comparable response to the 20 m coiled reactor. As expected, the back pressure of the 10 m knitted reactor was less than the 20 m c o i l . As a result, the flow rate of the HPLC was increased further up to 0.45 mL/min. The response curve (Fig. 24 of the Results and Discussion section) showed an optimum flow rate of 0.3 mL/min for the HPLC mobile phase. 55 2.7.6.4 Detector wavelength Excitation and emission maxima of the digoxin derivative were determined by alternately changing excitation and emission f i l t e r s . An optimum excitation and emission wavelength of 360 and 425 nm respectively was found. 2.7.6.5 Control of baseline noise Two nitrogen pressurized glass reagent reservoirs were used to generate pulseless flow of the reagents. The reservoir i s of heavy-walled glass with three PTFE valves. The three individually controlled valves permit the application of gas under pressure to the bottle for delivery, venting or flushing of the reagent. One valve of the reservoir remained closed at a l l times during the operation, while the second valve was connected to a two-stage regulator attached to a nitrogen cylinder. The reagents from the two reservoirs were mixed via a three way tee. The flow of the reagents was regulated by altering the pressure of the nitrogen gas over the solution i n the reservoir. Since the flow of dehydroascorbic acid had to be less than the flow of HC1, a restriction c o i l of varying length (3 to 7 m) was attached to the valve of the dehydroascorbic acid reservoir to add back pressure to that reservoir. A length of 5 m X 0.3 mm ID PTFE tubing was found to give sufficient back pressure and allow for an adequate amount of dehydroascorbic acid delivered relative to the delivery of HC1. Using this system, a nitrogen pressure of approximately 50 psi was sufficient to deliver 0.25 56 mL/min of reagents (HC1 flow of .20 mL/min and dehydroascorbic acid flow of 0.05 mL/min) and the column flow rate was maintained at 0.15 mL/min. Higher HPLC column flow rates could not be achieved since these would necessitate higher flow rates of the fluorogenic reagents and the pressures necessary were above the maximum safe lim i t of the glass reservoirs. Flow of the reagents was easily halted by relieving the pressure i n the reservoir via the unattached valve. 2.7.6.6 @-Cyclodextrin addition The effect of £-cyclodextrin on the fluorescence intensity of the digoxin derivative was also examined. This was readily studied by introducing ^-cyclodextrin solution into the end of the reactor described i n 2.7.1, via a three-way tee. The mixture was immediately passed into the fluorometer. Concentrations of 1.76 X 10"' _3 mM to 3 X 10 mM were studied. No change i n sensitivity was noted. The wavelengths for excitation and emission were also considered; however, excitation/ emission wavelengths of 360/ 425nm were found to be ideal. In addition the reaction time of the digoxin derivative i n the presence of §-cyclodextrin was increased by inserting a i m c o i l between the tee and the detector. This f i n a l modification led to a decreased noise level; however, the sensitivity of detection was unchanged. While &-cyclodextrin did reduce the baseline noise by a factor of 2, the expense of this reagent precluded i t s further use. 57 2.8 Reproducibility and linearity of the assay method Day-to-day precision of the detection of digoxin standards was assessed by comparing the fluorescence response of digoxin at levels from 1.5 to 10 ng. A total of 10 analyses per day for 5 days were used in this study. Standard deviations of the peak height ratios were determined for each concentration and the mean of each concentration point was plotted against the weight ratios of digoxin/ internal standard. The coefficient of variation at the 3 ng level for a single day determination was 2%. 2.9 Plasma extraction procedure 2.9.1 Extraction procedure i n water 2.9.1.1 Solvent-solvent extraction of digoxin i n water Water (5 mL) containing 1 ng/mL of digoxin was added to a screw capped tube. After the addition of an organic phase (see Table 8 of Results and Discussion section) the capped tube was placed on a rotating disk for 10 minutes at low speed to effect partition of digoxin into the organic phase. More rapid agitation led to emulsion formation. After centrifugation for 5 minutes at 2,500 rpm the lower layer was removed to a second tube and evaporated to dryness under a stream of nitrogen. The residue was reconstituted with 100 uL of the HPLC grade solvent used for chromatographic elution. 58 2.9.1.2 Solid phase cartridge extraction of digoxin in water Cartridge activition step: In a l l cases of solid phase extractions, the cartridge was positioned in a Luer-lock needle mounted on a rubber stopper on a test tube equipped with a side arm connected to a vacuum line. Vacuum pressure was adjusted to 15 i n . of mercury. Each cartridge was prewashed repeatedly with 5 mL methanol (3X) and then with water through opening and closing of the vacuum pressure line. This process activated the packing surface. Extraction step: With the vacuum i n i t i a l l y turned off, the "spiked" water containing known additions of digoxin was added. The sample was drawn through the column by vacuum and the drug was adsorbed on the column matrix. With the vacuum off, 10 mL of organic solvent (Table 9) was placed onto the column. The vacuum was turned on and the eluent was collected. The upper aqueous layer was discarded and the lower organic solvent was dried under a stream of nitrogen. The residue was reconstituted in 100 uL of the HPLC solvent used for chromatographic elution. The whole volume was injected into the chromatograph. 2.9.2 Extraction procedure i n "spiked" plasma 2.9.2.1 Solvent-solvent extraction of digoxin in plasma A volume of 3 mL of pooled human plasma, taken from the local blood bank, was pipetted into a 15 mL screw-capped tube. An 59 aliquot of 1 ng/mL of working digoxin standard was added and the mixture was mixed on a vortex mixer for 5 seconds. Then, 10 mL of the following organic solvents: dichloromethane, ethyl acetate and combination of either isobutanol, propanol ethanol or isopropanol and dichloromethane, as enumerated i n Table 8 of the Results and Discussion section, were added and the mixture extracted for 10 minutes as described previously. After centrifugation for 5 minutes at 1000 rpm, the organic phase was transferred into a 10 mL screw-capped tube and evaporated to dryness at 40°C under a stream of nitrogen. The residue was reconstituted in 100 uL with the HPLC grade solvent used for chromatographic elution and the whole volume was injected into the chromatograph. Because of the presence of interfering peaks, only two of the solvents mentioned above are shown on Table 10. 2.9.2.2 Solid phase cartridge extraction of digoxin in plasma To 3 mL of pooled plasma was added 1 ng/mL of digoxin in methanol. The plasma proteins were denatured by the addition of 3 mL of acetonitrile. After centrifugation for 5 minutes at 1000 rpm, the supernatant was passed through the cartridge by gravity with the vacuum in the test tube turned off. After 2 minutes, the vacuum was applied and the eluent discarded. The vacuum was then turned off and 10 mL of the following organic solvents: dichloromethane or dichloromethane/ isopropanol (Table 11) were added to the column and eluted through by gravity. After 1 minute of equilibration time, the 60 vacuum was turned on and the eluent was collected. This was then evaporated to dryness and the residue resuspended as before (2.9.1.2). The solid phase cartridge was activated as described in 2.9.1.2 " cartridge activation step". Several modifications to the method described i n Table 11 of the Results and Discusssion section were also examined. The following three methods were also summarized i n Table 12. Method A involved the solvent/ solvent extraction of 3mL of plasma, containing 1 ng/mL of digoxin with 10 mL dichloromethane/ methanol (15:1). After shaking the tube for 10 minutes, the organic phase was transferred onto the activated cartridge. The cartridge was then washed with 2 mL of water and the vacuum was turned on to force the endogenous substances through the cartridge. With the vacuum off, 15 mL of dichloromethane were added onto the cartridge and the eluent was collected. The organic solvent was then evaporated and reconstituted as before (2.9.1.2). Method B evaluated several bonded phases including d i o l , s i l i c a and ODS for their effectiveness i n retaining digoxin from the plasma. Method B, similar to method A, also involved a pre-extraction step with two aliquots of 10 mL dichloromethane/ isopropanol (15:1). The organic phase was then transferred to a 10 mL test tube and evaporated to dryness under a stream of nitrogen. The residue was reconstituted with 1 mL of solvent (such as water for reversed phase cartridges and chloroform for forward phase cartridges). This solvent was allowed to slowly pass through the extraction tube by gravity and the sample was eluted from the cartridge 61 using dichloromethane/ isopropanol (15:1) for the reversed phase cartridges and isopropanol for the s i l i c a cartridge. The vacuum was turned on after 1 minute and the eluent collected and processed as before (2.9.1.2). Method C was designed to remove the interfering peaks present i n the eluent found from the results of using method A and B. The procedure was similar to method B but with an additional solvent wash . After solvent-solvent extraction, the sample was resuspended in either water or chloroform depending on the type of cartridge used. The reconstituted sample was passed through the cartridge. The cartridge was flushed with either ethyl-ether/isopropanol, chloroform or ethyl acetate depending on the cartridge used. This washing step was found to remove the interfering endogenous substances. Digoxin was then eluted with solvent 4 which was either dichloromethane/ isopropanol or isopropanol (see Table 12). 2.9.3 Actual extraction procedure of digoxin in plasma Just prior to analysis, the frozen plasma samples were allowed to thaw at room temperature. In this procedure, from 1.5 to 10 ng/ 10 uL of digoxin solutions and 80 ng/ 10 uL internal standard solution was added to 3 mL of plasma. I n i t i a l l y this was deproteinized using 3 mL of acetone. The protein-free plasma was then washed with 2 mL of isooctane to remove any non-polar endogenous materials. The isooctane layer was then separated and the acetone layer was partially evaporated under a stream of nitrogen 62 for 20 minutes to a volume of approximately 3 mL. The remaining supernatant was then extracted with 2 aliquots of 10 mL of 2% n-propanol in dichloromethane. The combined organic phases were filter e d and evaporated to dryness under nitrogen. The residue was resuspended i n 100 uL of 50/50 methanol/ water. Plasma standards were prepared fresh for each assay run to establish linearity of the method. Injections of the resuspended residues into the liquid chromatograph were performed with the aid of a 1 rnL syringe (Waters Assoc.). The schematic outline of the sample preparation i s given in Figure 8. Plasma used i n this study was obtained from The Red Cross Blood Bank and U3C Health Science Acute Care Hospital blood bank. 2.10 Recovery and precision of extraction Recovery of digoxin from plasma following 'the method previously outlined (2.9.3) was checked by comparing the peak height of digoxin in plasma samples to those of the solutions of the drug in methanol of equivalent known concentration. Repeated extractions of the same concentrations in plasma were also compared from day-to-day to determine their inter- and intra-assay v a r i a b i l i t y . 2.11 Calibration curve and sensitivity of extraction Concentrations of 0.5 to 3.3 ng/mL were prepared and analyzed. The relationship of peak height ratios to their respective concentrations was then determined. The sensitivity of extraction was assessed by adding 0.2 , 63 3 mL p lasma s p i k e d w i t h 10 uL d i g o x i n (STD) add 10 uL i n t e r n a l s t a n d a r d 3 mL a c e t one V o r t e x C e n t r i f u g e 5min a t 2 ,500 Wash w i t h 2 mL i s o - o c t a n e V o r t e x C e n t r i f u g e and s e p a r a t e I s o - o c t a n e l a y e r ( d i s c a r d ) A ce tone /Aqueous l a y e r e v a p o r a t e f o r 20 mi n s . Ix add 10 mL e x t r a c t a n t [2% p r opano l - i n d i c h l o r o m e t h a n e ) Aqueous l a y e r mix f o r 10 mins c e n t r i f u g e 5 mins a t 2 ,500 RPM O r g a n i c l a y e r F i l t e r E v a p o r a t e Res i due add 100 uL 50 /50 MeoH/H20 F igure 8. E x t r a c t i o n S c heme of D igox i n f rom P l a sma 64 0.3 and 0.5 ng/niL of digoxin to plasma samples and extracting these as outlined i n 2.9.3. A fluorescence response at the detector, with a signal to noise ratio of at least 2:1 was considered to be the minimum detection limit. 2.12 Specificity Using drug-free plasma samples taken from the Red Cross Blood Bank (Vancouver, Canada) and UBC Health Science Acute Care Hospital (ACU) (Vancouver, Canada), the extraction procedure yielded blanks which were consistently free from interfering peaks in the retention area of digoxin (Fig.28 of Results and Discussion section). Plasma pools, to which commonly co-administered drugs such as furoseraide, quinidine, spironolactone, disopyrainide, procainamide, persantine, propafenone, trimethoprim, sulfamethoxazole, captopril, docusate sodium and verapamil were tested for potential interference in the assay. Furosemide and spironolactone were among those drugs that were detected but did not elute i n the retention area of digoxin. The other drugs did not show any peaks using this assay method. The parent drug and i t s respective metabolites were also well separated from digoxin in accordance with expected performance. 2.13 Quality control procedure Lyphochek Radioassay control serum (human) levels I, II 65 a n d I I I s u p p l i e d b y E n v i r o n m e n t a l C h e m i c a l S p e c i a l t i e s ( A n a h e i m , C a . , U S A ) w e r e u s e d a s t h e q u a l i t y c o n t r o l c h e c k f o r t h e c o r r e l a t i o n o f t h e h o s p i t a l R I A m e t h o d a n d t h e H P L C p o s t - c o l u m n m e t h o d d e v e l o p e d . L e v e l I l o t ( # 1 9 1 8 1 ) , l e v e l I I l o t ( # 1 9 1 8 2 ) , a n d l e v e l I I I l o t ( # 1 9 1 8 3 ) , w e r e u s e d i n t h i s s t u d y . K n o w n a s s a y v a l u e s f o r l e v e l I w e r e i n t h e l o w r a n g e , l e v e l I I w e r e i n t h e m i d - r a n g e a n d l e v e l I I I w e r e e l e v a t e d . E a c h L y p h o c h e k R a d i o a s s a y c o n t r o l s e r u m w a s r e c o n s t i t u t e d t o 5 m L a n d e x t r a c t e d i n t h e m a n n e r d e s c r i b e d i n t h e p l a s m a e x t r a c t i o n p r o c e d u r e . T h e m e a n v a l u e s d e t e r m i n e d f o r t h e H P L C p o s t - c o l u m n m e t h o d w e r e d e r i v e d f r o m r e p l i c a t e a n a l y s e s . T h e U B C H e a l t h S c i e n c e A c u t e C a r e H o s p i t a l R I A k i t s u s e d t o a s s a y t h e s e s a m p l e s w e r e s u p p l i e d b y N u c l e a r M e d i c a l L a b s . I n c . ( D a l l a s , T x . , U S A ) . 2 . 1 4 A n a l y s i s o f c a r d i a c p a t i e n t p l a s m a s a m p l e s S a m p l e s w e r e o b t a i n e d f r o m U B C H e a l t h S c i e n c e A c u t e C a r e H o s p i t a l a n d V a n c o u v e r G e n e r a l H o s p i t a l ( V G H ) ( V a n c o u v e r , C a n a d a ) . A p o r t i o n o f t h e p l a s m a w a s a n a l y z e d b y t h e h o s p i t a l R I A m e t h o d a n d t h e r e m a i n i n g p l a s m a w a s s t o r e d i n a r e f r i g e r a t o r u n t i l r e q u i r e d f o r H P L C a n a l y s i s . W a t e r w a s s o m e t i m e s a d d e d t o t h e p l a s m a s a m p l e t o m a k e u p t h e v o l u m e t o 3 m L t o a v o i d c h a n g e s i n t h e p a r t i t i o n c o e f f i c i e n t s i n t h e e x t r a c t i o n s t e p . T h e s u p e r n a t a n t w a s p r o c e s s e d a c c o r d i n g t o t h e p r o c e d u r e o u t l i n e d a b o v e . 2 . 1 5 C a l c u l a t i o n s 66 Linearity of the assay procedure was established i n the range 0.5 ng/mL to 3.3 ng/mL of plasma by analyzing spiked human plasma samples covering the range. A calibration curve, consisting of 6 different concentrations within the expected range of the samples to be analyzed was generated by least-square regression of the peak height ratios (drug/ internal standard) against the concentration ratios. The slope and intercept so obtained were used to calculate the concentrations of the patient samples. The calculations of the correlation coefficient, slope and intercept of the standard curve of the assay method were dealt with similarly as above. 67 3. RESULTS AND DISCUSSION 3.1 Preliminary studies for the development of an analytical system for digoxin 3.1.1 On-column injection for capillary gas chromatography(GC) The chemical i n s t a b i l i t y of digoxin i n the presence of HFBA has been shown to be the complicating factor i n the analysis using GC-ECD (133). The close resemblance of digoxin to i t s hydrolysis products hampers the detection of degradation products or metabolites of digoxin. The extensive pre-purification step added to make the method specific, represents the major problem in this assay. In the recent literature, on-column injection with capillary GC has been consistently proven to be useful when thermally labile compounds or high molecular weight compounds are to be analyzed (158) . Because the sample i s placed directly on the column without a flash vaporization step, those substances such as digoxin, which may degrade during a conventional syringe flash vaporization injection may be unaffected. In order to investigate the usefulness of on-column injection in GLC procedures, several experiments using mild reagents, such as T r i - S i l Z, MBTFA and HFBI were undertaken. The f i r s t two reagents did not lead to any peaks in the chromatogram that could be identified as the digoxin derivative. 68 HFBI, on the other hand, was described by the manufacturer as a mild reagent that does not release acid by-products that could lead to hydrolysis of labile compounds. Unfortunately, this reagent did lead to hydrolysis of digoxin to i t s aglycone and i t was assumed to be due to the temperature required to effect derivatization (Figure 9 ) . No other derivatizing reagent i s commercially available that would offer any advantage i n terms of sensitivity and specificity for the measurement of digoxin to those already tried. 3.1.2 Liquid chromatography- mass spectrometry The development of combined high-performance liquid chromatography (HPLC) and mass spectrometry (MS) i s arousing much interest as the technique shows a growing capability for handling non-volatile organic substances. Mass spectrometry i s the most powerful technique available for structural characterization and i s often the method of choice for the identification of compounds. By u t i l i z i n g the high specificity of the mass spectrometer combined with the resolving power of the HPLC, the individual metabolites of digoxin could feasibly be assayed independently and therefore lead to accurate quantitation of digoxin i t s e l f . A conventional reversed-phase octadecylsilane (ODS) 25 cm column was coupled to the LC/MS system using a spray interface. Several experiments were attempted to determine the chromatographic parameters for the digoxin standard samples. It appeared from these experiments that the concentrations of the samples and the amount of 69 F igure 9. Gas Ch roma tog ram of HFB -D i go x i n Us ing On -Co lumn In jec t i on Ch r oma t og r aph i c c o n d i t i o n s : Co lumn: Carbowax 20M C a p i l l a r y co lumn (20 m X 0 .2 mm I D ) ; I n i t i a l t e m p e r a t u r e : 50°C ; I n i t i a l t i m e : 2 m i n . ; Tempera tu re program r a t e : 30 / m i n . ; F i n a l t e m p e r a t u r e v a l u e : 250 C; C a r r i e r gas (He l i um) f l o w : 1 mL /m in ; Sample s i z e : 5ug/2uL 70 water i n the effluent affected the performance of the mass spectrometer. The mass spectra were found to vary with the concentrations introduced into the interface. Several mobile phases of different water composition were examined and each time a different fragmentation pattern was obtained. The present LC-MS system equipped with a direct liquid introduction (DLI) interface requires solvent s p l i t t i n g . Only lOuL/min can be accepted by the mass spectrometer; therefore only 1% of the solute injected onto the conventional column can be introduced into the mass spectrometer at nominal flow rates of 1 mL/min. Consequently, the detection limit of this system i s relatively unfavorable for digoxin therapeutic monitoring. The use of microbore packed columns which have become commercially available can be considered as an attractive and efficient solution to the problem because with such columns a l l the mobile phase can be continously delivered into the ion-source. However, the major disadvantage i s that the entire LC system has to be converted to microbore capability to accomodate such columns. In addition, in order to use the LC-MS available for digoxin quantitation, modifications on the LC-MS interface would be needed. Therefore, the lack of reproducible ion spectra and the lack of a sufficiently efficient interface prevented further studies of this mode of analysis. 3.1.3 Liquid chromatography and UV detection 71 The assessment of plasma levels of digoxin and i t s metabolites would definitely require methods that could resolve a l l the metabolites from digoxin. We have reported (138) the usefulness of adapting a reversed-phase system to the HPLC analysis of this cardenolide as determined by spectrophotometric assay, based on the UV absorbance (220 nm) of the lactone ring. Unfortunately, these compounds exhibited relatively weak absorption maxima at this wavelength which was not sufficient for-detection of nanogram amounts of digoxin i n a biological matrix. 3.1.4 Fluorometric determination of digoxin An i n i t i a l attempt to develop a fluorescence assay for digoxin was accomplished by using a fluorotag, dansyl hydrazine. This tag would derivatize carbonyl functions such as that present i n the lactone ring of digoxin to give a dansyl hydrazone derivative. However, sensitivity was poor and also yielded several derivatives as depicted by the chromatogram (Figure 10). Analytical studies on cardiac glycosides so far have shown that specific reactions may only be expected from a reaction taking place somewhere i n the steroid ring system. Methods based on such reactions were for the most part fluorometric. We therefore compared the fluorescence intensity generated by different existing fluorometric methods (Table 1). It i s evident from the results that the United States Pharmacopeial (USP) method i s the most sensitive of the procedures used. Consequently, attention was focused on this method, which depends on the action of hydrochloric acid in the 72 F igure 10. HPLC Ch r oma tog r am of Dansy l Hyd ra zone of D igox in M o b i l e pha se : M e t h a n o l / w a t e r ( 7 0 : 3 0 ) ; f l o w r a t e : 1.0 mL/m in ; f l u o r e s c e n c e d e t e c t i o n w a v e l e n g t h : e x c i t a t i o n / e m i s s i o n 350/520 nm; samp le s i z e : 40ug/20uL 73 TABLE 1. Results o f Fluorometric Methods METHOD WAVELENGTH (nm) RFI n=3 Naik (a) 390/420 3.0 + 0.2 J a k o v l j e v i c (b) 345/435 0.8 + .01 USP method (c) 375/420 7.1 + 0.2 (a) s u l f u r i c a c i d (145) (b) a c e t i c anhydride, a c e t y l c h l o r i d e and t r i f l u o r o a c e t i c a c i d (146) (c) a s c o r b i c a c i d , hydrogen peroxide and concentrated h y d r o c h l o r i c a c i d (147). 74 presence of hydrogen peroxide and ascorbic acid in methanol. With this procedure, the excitation and emission spectra were determined for digoxin using an Aminco Bowman spectrofluorometer (Figure 11). The development and decay of digoxin fluorescence, using the USP method, from 1 to 7 hours were studied. It was found that by substituting water for methanol (as required in the USP) there was no change i n the intensity of the fluorescence and that the fluorescence was stable throughout the period examined. This procedure was therefore adapted and modified for the post-column derivatization technique. Since water i s preferred in any reversed-phase HPLC mobile phase for chromatographic resolution, this substitution was employed. 3.2 Optimization of the HPLC post-column derivatization method 3.2.1 Other reagents for derivatization In order to minimize the deleterious effects of HC1 fumes on the instrumentation, several attempts were made to substitute HC1 with other less fuming acids such as sulfuric acid (70%) , perchloric acid and trichloroacetic acid. Unfortunately, these acids produced less fluorescence intensity than did HC1 (Table 2). Other investigations of additional fluorogenic reagents were chosen from those that involved the functional groups present in the steroid ring structure of digoxin. Horikawa et a l . (149) 75 Figure 11. Excitation & Emission Spectra of Digoxin Wavelength TABLE 2. Effect of Different Acids on the Fluorescence Intensity of Digoxin Using Post-column Derivatization Acid Peak Height (cm) Hydrochloric acid 5.4 Sulfuric Acid (70%) 1.1 Perchloric Acid 3.0 Trichloroacetic Acid 1.0 77 developed a fluorometric method for the determination of 4 A-3-ketosteroids i n which the steroids react with isonicotinylhydrazine (INH) in a methanolic aluminum chloride solution to form hydrazones. These hydrazones fluoresce owing to complex formation with aluminum ions. Another method was that of Seki et a l . (148) who used glycinamide/hexacyanoferrate in a weakly alkaline media of borate solution to detect urinary 17-hydroxyl-corticosteroids. Several parameters were changed to optimize the reactions but these proved to be unsuccessful since the derivative formed exhibited a signal less than 10% of the HC1 induced fluorescent derivative. 3.2.2 Flow c e l l Our progress had been hampered, i n great measure, due to detector problems. Most of the detector flow cells available on the market are made of stainless steel, and a major drawback was encountered when i t was realized that the rate at which hydrochloric acid was etching the c e l l surface of the Schoeffel 970 fluorescence detector, rendering i t unserviceable in a few weeks. In order to minimize this etching and prevent damage to the detector, i t was necessary to substitute the stainless steel flow c e l l . A titanium c e l l was f i n a l l y suggested (159) and manufactured to the exact specifications as the standard stainless steel c e l l . When the titanium c e l l was installed and tested i t was found to be more resistant to hydrochloric acid but s t i l l was etched. This etching, however slight, led to leakage of the HC1 gas from the solution and 78 was suspected to damage the electronics of the instrument. These problems with the Schoeffel detector required us to return to using the less sensitive Waters 420AC fluorometer. An a l l quartz flow c e l l of 40 mm i n length and 1 mm ID, connected by acidflex tubing and held i n place by epoxy glue, was used. Although the Waters fluorometer was not as sensitive as the Schoeffel instrument, the advantages of a non-corrosive c e l l were considered significant and i t was therefore used i n subsequent experiments. 3.2.3 Choice of chromatographic columns An i n i t i a l study to optimize the HPLC post-column fluorometric technique was focused on the development of a rapid method for the separation and accurate quantitation of digoxin. Dihydrodigoxin i s quantitatively of major importance in some patients and therefore should be separated from digoxin i f accurate measurements of digoxin are to be achieved. Thus far, the separation of dihydrodigoxin from digoxin has never been reported in the literature. As a consequence, i n approaching this goal, we have included the dihydro-metabolites dihydrodigoxin and i t s aglycone, in the separation of digoxin. It i s relevant to emphasize the importance of obtaining a short chromatographic time. Since LC bands widen as retention time increases, later-eluting bands show a corresponding reduction i n peak height and eventually disappear into the baseline. Several reversed-phase columns were investigated as to their a b i l i t y to separate digoxin and i t s metabolites using the 79 combination alcohol/water mobile phase. To compare the efficiency of these columns, i t i s necessary to have some quantitative measure of the relative resolution achieved (see Fig. 12 to 15). The resolution (Rs) of digoxin (I) and dihydrodigoxin (II) i s defined as the distance between the two band centers, divided by the average band width (160): R s = " V equat ion 1 ih)(*l + w n) t j and t j j r e f e r to the r e t en t i on time o f d ihyd rod igox in and d igox in r e s p e c t i v e l y . Wj and W J J r e f e r to the bandwidth o f the two bands. The chromatograms in Figure 13 show that the separation and resolution could be optimized by choosing the appropriate column length and the size of the packing materials used. It was evident that the shorter columns reduce analysis time, albeit at the cost of resolution. Some of the selectivity can be restored, however, by using smaller size particles for packing the column. The 10 cm column showed an overall shorter chromatogram. A more dramatic difference between the 10 cm and 25 cm column i s depicted i n Figure 12. Additional separation selectivity was obtained by using a C-8 column (Figure 14) but at the expense of increased elution time. Finally, an acceptable total chromatographic time and resolution was achieved by using a 3u, 15cm length column (Figure 15). Although the separation of dihydrodigoxin from digoxin was not completely resolved (Rs= .909) , measurement of the quantity of digoxin would be accurate i f peak height measurements were used. 80 Figure 12. Separation of Digoxin and Its Metabolites Using 10 and 25 cm • 01 oira i 1 1 — i , — — r 0 8 14 0 20 40min 10cm ODS 3um 25cm ODS 5 urn C h r o m a t o g r a p h i c c o n d i t i o n s = M o b i l e phase = Me thano l - Wa te r ( 7 0 - 3 0 ) ; f l o w r a t e = 0 . 1 mL /m in ; UV d e t e c t i o n a t 210 nm, f o r i d e n t i f i c a t i o n o f t h e peaks see F i g u r e 13 < F igure 13 . Op t im i za t i on of Ch r oma tog r aph i c Reso l u t i on CN + CO R s = . 3 4 2 CN + 0 15 6cm ODS R s = 0 CO NJ1 (O + CM + Rs=-643 CO VJ I 0 2 0 0 2 5 m i n 12.5 cm ODS 10 cm ODS 3 urn hype r s i l 5 um 3 urn 82 Legends 1. d i h y d r o d i g o x i g e n i n 2 . d i g o x i g e n i n 3 . d i g o x i g e n i n m o n o d i g i t o x o s i d e 4 . d i g o x i g e n i n b i s d i g i t o x o s i d e 5 . d i h y d r o d i g o x i n 6 . d i g o x i n Ch r oma t og r aph i c c o n d i t i o n s = M o b i l e phase = Me thano l - e t h a n o l - i s o p r o p a n o l - w a t e r ( 5 2 - 3 - 1 - 4 5 ) ; f l o w r a t e = 0 . 3 mL /m in ; p o s t - c o l u m n f l u o r e s e n c e d e t e c t i o n u s i n g non-segmented r e a c t i o n s y s t e m as d e s c r i b e d i n F i g u r e 7 . 83 F igure 14 . S epa r a t i o n of D igox in and i ts Me t abo l i t e s Us ing C - 8Co l umn CM + Rs=-375 i 1 — 0 4 0 m i n C o n d i t i o n s and i d e n t i f i c a t i o n o f peaks same as i n F i g u r e 1 3 . 8 4 F igure 15 . Op t im i z ed Sepa r a t i o n of D igox in and i ts Me tabo l i t e s CM + R s = . 9 0 9 i : 1— 0 4 0 m i n 15 cm ODS 3 urn C o n d i t i o n s and i d e n t i f i c a t i o n o f peaks same as i n F i g u r e 1 3 . 85 3.2.4 E f f e c t o f mobile phase In the process o f developing a s p e c i f i c assay, a change i n sep a r a t i o n s e l e c t i v i t y was i n d i c a t e d . This was most r e a d i l y achieved by a change o f mobile phase s e l e c t i v i t y . A systematic approach t o s e l e c t i v i t y o p t i m i z a t i o n was undertaken t o f i n d the c o r r e c t s o l v e n t s t r e n g t h o f the mobile phase. Solvent mixtures o f three o r f o u r water m i s c i b l e organic m o d i f i e r s were a l s o used i n f i n e - t u n i n g s o l v e n t s e l e c t i v i t y t o g i v e the best s e p a r a t i o n . Solvent s t r e n g t h can be a c c u r a t e l y measured by p o l a r i t y ( P 1 ) . P 1 of a s o l v e n t mixture i s the a r i t h m e t i c average o f the P 1 values o f the pure s o l v e n t s i n the mixture, weighted according t o the volume f r a c t i o n o f each s o l v e n t (161):  n P' = I 0 . P . e q u a t i o n 2 i = l 1 1 n = the number o f pure s o l v e n t i n t h e m i x t u r e 0. = volume f r a c t i o n o f component o f s o l v e n t s y s t em = s o l v e n t p o l a r i t y p a r ame t e r Table 3 shows the s o l v e n t strengths o f the mobile phases used. Figure 16 i l l u s t r a t e s the l i n e a r r e l a t i o n s h i p o f the r e s o l u t i o n o f d i g o x i n and d i h y d r o d i g o x i n w i t h the p o l a r i t i e s o f the s o l v e n t s . The i d e a of combining l i q u i d chromatography w i t h post-column r e a c t i o n s f o r d e t e c t i o n purposes has some drawbacks. The major l i m i t a t i o n l i e s i n the r e s t r i c t i o n p l a c e d on the choice o f the chromatographic s o l v e n t s s i n c e t h i s w i l l i n f l u e n c e the r e a c t i o n . Wells e t a l . (177) s t u d i e d the e f f e c t o f methanol on the fluores c e n c e of c a r d i a c g l y c o s i d e s and found t h a t methanol i s an 86 TABLE 3 . M o b i l e Phase P o l a r i t y and R e s o l u t i o n o f D i g o x i n f rom D i h y d r o d i g o x i n n MOBILE ETOH PHASE MEOH :OMPOSITION WATER U ) (a) IPA DCM P ' e (b) Rs ( c ) 1 — 70 30 — — 663 0 2 — 40 47 9 4 727 .3 ? 3 5 55 40 — 710 0 . 3 4 5 50 45 — 735 .5 0 .65 5 4 . 95 54 .46 39 . 60 0 . 99 — 706.81 0 .298 6 4 . 95 49 . 50 44 . 55 0 .99 — 732 .0 0 .752 ' 7 2 .97 58 .42 37 .62 0 . 99 — 698 .3 0 8 2 .97 54 .46 41 . 58 0 .99 — 718 .5 0 .808 9 2.97 51 .49 44 . 55 0 . 99 — 733 .6 0 .909 10 2 55 42 1 — 721 . 4 0 .516 (a ) P o l a r i t y o f s o l v e n t s : ETOH ( e t h a n o l ) 4 . 3 , IPA ( i s o p r o p a n o l ) 3 . 9 , MEOH (me thano l ) 5 . 1 , DCM ( d i c h l o r o m e t h a n e ) 3 . 1 , WATER 10 .2 (161) (b ) P o l a r i t y c a l c u l a t e d f rom r a t i o o f s o l v e n t s and u s i n g t he p o l a r i t y v a l u e ( e q u a t i o n 2 ) . ( c ) R e s o l u t i o n c a l c u l a t e d u s i n g e q u a t i o n 1. 87 F i gu re 16 E f f e c t of s o l v e n t p o l a r i t i e s on r e s o l u t i o n S o l v e n t P o l a r i t y important component of the reaction medium. Similar observations were found in the optimization of the mobile phase, that i s , the combined alcohols (Table 3) were preferred as organic solvents for the post-column reaction, since other solvents such as acetonitrile and dioxane were found to decrease the fluorescence yield of digoxin. The mobile phase developed by Desta et a l . (138) was not suitable for the present post-column reaction because of the quenching of the fluorescence by dichloromethane and the lowering of the boiling point of the mobile phase, therefore creating bubbles i n the detector c e l l . The combination of the different proportions of alcohols i n the mobile phase provided the ideal reaction medium. The optimal mobile phase for this study was chosen to be #9 of Table 3. 3.2.5 Choice of reaction detector The choice of a particular reaction detector type i s influenced by the kinetics of the reaction since the primary aim i s to preserve the original chromatographic resolution as much as possible. Band broadening i s another crucial' factor to consider i n the construction of the proper reaction detector. On the basis of literature data, i t could not be expected that, for a certain reaction time, a segmented system can compete with a tubular or a packed bed reactor. The three types of reactors were therefore compared. 3.2.5.1 Packed-bed reactor A packed-bed reactor was examined i n consideration to this 89 type of reactor can be favorable for reactions that take up to several minutes. The flow rates of the reagents and mobile phase were maintained at the rate optimized by Gfeller et a l . (9). In order to minimize band broadening phenomena, good packing technique of the packed bed similar to the packing of HPLC column was used. Packings were i n i t i a l l y washed with water then methanol to remove fines and organic impurities from the beads. Unfortunately, when the fluorogenic reagents (HC1 and dehydroascorbic acid) were introduced into the bed, a yellowish discoloration was observed. Consequently, the back pressure inside the column increased preventing the reagents to flow through the bed. This was assumed to be due to the disintegration of the beads caused by the HC1 solution. The ins t a b i l i t y of the glass beads under the reaction conditions precluded further experimentation with this type of reactor. 3.2.5.2 Air segmented reactor Studies of the post-column reactor for the fluorogenic analysis were also undertaken using a continuous flow a i r segmentation system (Figure 6). Several modifications to the method developed by Gfeller et a l . (9) were introduced. Miniaturization of the mixing tees, mixing c o i l s , reaction spirals and debubbler reduced the band broadening and hence increased the sensitivity for detection of digoxin. The reagent delivery geometry was also altered to conform with the findings of Frei et a l . (162,163) who noted that the construction of the mixing unit to be crucial, especially when eluent and reagent solutions are of widely differing densities. A 90 design, whereby the reagent solution enters at a 30 angle against the eluent stream, causes enough turbulence and good radial mixing to reduce band broadening by more than 30% as compared to addition at a 90°angle to the eluent flow. Therefore this was adapted as the mixing unit. It has also been noted i n the literature that the Technicon (C5) debubbler i t s e l f contributes heavily to sample carry-over in the system (164). By reducing the internal volume of this C5 debubbler to a minimum, we anticipated a notable improvement in band broadening of the peaks. The least detectable amount achieved using this modified system was 5 ng/injection at signal/ noise= 3/1 (Figure 17). From the sensitivity observed and data derived after the band broadening study (see band broadening section 3.2.10) i t appeared that the miniaturization only contributed to slight improvement i n the performance of the system. The beneficial effect of a i r segmentation in preventing carry-over has been emphasized extensively (165). However, there are technical drawbacks that are worth noting: (a) the streams have to be debubbled before they reach the flow c e l l ; (b) the size of the air bubbles has to be controlled; hence, only glass coils are used for mixing to preserve the integrity of the bubbles(166); (c) a leak-free system i s required for caustic reagents. The fluctuation in the solvent flowing pressure created problems for the acidflex tubing which held the mixing units and the reactor i n series. It was observed that an increase in flowing pressure disconnected the acidflex tubing from the glass coi l s making i t d i f f i c u l t to maintain the leak-free system. Therefore for the reasons enumerated, further 91 Figure 17. Low L e v e l De t e c t i on of D igox in Us ing Ai r Segmen ta t i on S y s t e m Ch roma tog r aph i c c o n d i t i o n s : Co lumn: 15 cm, 3 urn ODS; m o b i l e pha se : m e t h a n o l / e t h a n o l / i s o p r o p a n o l / w a t e r ( 5 2 : 3 : 1 : 4 5 ) ; f l o w r a t e : 0 .4 mL/m in ; p o s t - c o l u m n d e t e c t i o n u s i n g a i r - s e g m e n t e d r e a c t i o n s y s t em as d e s c r i b e d i n F i g u r e 6 . 92 experiments on this type of reactor were discontinued. 3.2.5.3 Tubular non-segmented reactor Further studies on the post-column derivatization for the fluorogenic analysis of digoxin were undertaken using a non-segmented reactor as described i n the experimental section. The elimination of the bubble segment i n this case greatly increased the ease of maintaining the post-column reactor flow. Several of the HPLC post-column derivatization parameters were investigated individually. 3.2.6 Relative proportions of reagents By altering the concentrations of dehydroascorbic acid, while maintaining the quantity of HC1 constant, and then reversing the relationship, an optimal balance was found (Table 4). A flow ratio of 0.1 dehydroascorbic acid to 0.5 HC1 yielded an optimal fluorogenic efficiency for these two reagents. In addition to this study, i t was found that the concentration of hydrogen peroxide in ascorbic acid (dehydroascorbic acid) was affecting the efficiency of fluorescence. In this study (Figure 18) a very sharp maximum was observed. Hence, this reagent was prepared fresh daily to make certain that the optimum _ 3 concentration was maintained (1.1X10 M in a 0.1%w/v ascorbic acid solution). 93 TABLE 4 . R e l a t i v e P r o p o r t i o n s o f Reagents I.D. DA D e l i v e r y Tubing (a) Peak Height n=5 . 02 " . 015 .010 1 .95 + . 05 cm 2 .33 + .04 1.48 + . 07 I.D. HC1 D e l i v e r y Tubing (b) Peak Height n=5 .29 " . 53 .63 1.80 + . 07 cm 2 .35 + .06 1.95 + . 07 (a) dehydroascorbic a c i d t u b i n g ( S o l v a f l e x ) i n t e r n a l diameter where HC1 t u b i n g used was 0 . 53 "ID (b) h y d r o c h l o r i c a c i d t u b i n g ( a c i d f l e x ) i n t e r n a l diameter where DA tubi n g used was . 015 "ID 94 F igure 18 EFFECT OF HYDROGEN PEROXIDE CONCENTRATION IN ASCORBIC ACID ON FLUORESCENCE 2 .4 , 3.0 X 1 0 " 3 M C O N C E N T R A T I O N ( H J O J ) 3.2.7 Optimization of reaction temperature I n i t i a l optimization of this parameter was studied with methanol as the mobile phase. In order to accelerate the reaction and simultaneously increase the fluorescence efficiency of the derivatization, the temperature of the bath was altered up to 64°C. Above 64°C, the methanol present i n the reaction medium boiled and data could not be obtained. A reaction temperature of 59°C was therefore selected as the optimum reaction temperature when the mobile phase consisted of pure methanol (Figure 19). Following the development of the mobile phase for the separation of digoxin from i t s metabolites, i t was found that the temperature of the reaction bath could be increased beyond the boiling point of methanol without bubble formation occurring i n the detector c e l l . This increase i n temperature considerably accelerated the reaction time i n the post-column reactor. A restriction c o i l (1 m teflon tubing) connected at the exit end of the detector added back pressure to the system which facilitated s t i l l higher temperatures without boiling of the reaction medium, thereby further increasing the rate of the post-column reaction. Maximum peak height was obtained at a temperature of 79 + 1°C. Figure 20 shows the effect of the increased reaction temperature on the sensitivity of the post-column detection. 3.2.8 Reaction kinetics In addition to the reaction temperature, the reaction time must also be optimized. This study was most readily accomplished by 96 Figure 19 OPTIMIZATION OF REACTION TEMPERATURE 2.5, 0.5 0 43 50 55 60 65 TEMPERATURE (°C) F igure 20 . Inf luence of Tempera tu re on the Sens i t i v i t y of De t e c t i on HPLC p o s t - c o l u m n d e t e c t i o n c o n d i t i o n s same as i n F i g u r e 1 3 . Amount o f d i g o x i n i n e j e c t e d : 1 n g . 98 varying the speed of the peristaltic pump which delivers the HC1 and dehydroascorbic acid. From the experimental data (see Figure 21) i t was determined that the total flow rate of 0.27 mL/min (HPLC pump plus peristaltic pump flow) was optimal for the reaction time. The HPLC pump was operated at 0.1 mL/min for the whole range of the peristaltic pump speed. In terms of actual reaction time, a total flow of 0.27 mL/min requires 7 minutes to traverse a 0.3mm X 20m c o i l . Reaction kinetics were also studied by controlling the length of the capillary reactor. Reaction coils of 10, 15, 20 and 25m with 0.3mm internal diameter provided reaction times of 3, 5, 7 and 8.6 minutes respectively with combined 0.1 mL/min mobile phase and 0.17 mL/min total reagent flow rates. Response curves obtained for digoxin at the four reaction times are shown in Figure 22. The decrease in peak height response for the 25 m c o i l appeared to be due i n part to peak dispersion i n the reaction tube. The 20 m c o i l enabled maximum response with minimal peak broadening. It has been established (165) that while traversing a straight reaction tube more dispersion i s obtained. This situation was suggested to improve drastically when moving through a coiled reaction tube which was believed to create turbulent flow. This phenomenon was deduced to be due to centrifugal forces acting on the flow pattern; secondary flows perpendicular to the main flow direction are produced. As a result, better radial mixing and consequently, a reduction of band broadening occurs (see Figure 23). With this knowledge i n mind, two different reactors were studied which were wound to give an outer diameter of either 50 cm 99 E o 2/»r 2.2r 2.0f 1.8 X < £ 1.4 r •1.2 Figure 21 REACTION KWETICS 1.0* 0.1 0.2 0.3 FLOW RATE (ml/mm) 100 Figure 22 EFFECT OF REACTION COIL LENGTH ON FLUORESCENCE T 0 1 2 3 4 5 6 7 8 9 REACTION TIME (min) FIGURE 23. S e c o n d a r y f l o w p a t t e r n i n t h e c r o s s s e c t i o n o f a c o i l e d t u b e . 1 6 4 102 or 2.7 cm respectively. The peak height, which reduces as band broadening increases, was 42% greater for the narrow (2.7 cm) c o i l . Hofmann and Halasz (167) proposed that geometrical deformation of a tube also leads to reduced band dispersion. In order to optimize the arrangement of the reaction tube, several types of reactor configurations were compared with respect to peak broadening. The results obtained for the different reactor geometries examined are as follows: (a) 20 m coils of 2.7 cm diameter post-column reactor had a basewidth of 3 minutes. (b) 10 m c o i l of 2.7 cm diameter post-column reactor had a basewidth of 2.5 minutes. Studies by Uihlein and Schwab (168) have shown that knitted ("fringes" like those used for crochet work) PTFE capillaries lead to further improvement with respect to peak broadening. This type of reactor geometry was therefore adapted to the fluorogenic system under investigation. A 10 m capillary tube was knitted into fringes and used as the reaction unit. A basewidth of 2.2 minutes was found. The asymmetric factor for the knitted configuration was 2.5 while the coiled reactor had an asymmetric factor of 3. The knitted capillary reactor had a slight advantage over the tightly coiled tube. This was similar to the observations made by Deelder et a l . (169). I n i t i a l l y , using the 2.7 cm diameter PTFE reactor (20 m) the flow rate of the HPLC pump was varied to optimize resolution 103 and speed of analysis. A flow rate of 0.15 mL/min was found to give the best result without contributing excessive back pressure to the reagent delivery path. The decrease i n back pressure of a 10 m reactor, as compared to the 20 m reactor, allowed us to increase the HPLC flow rate from 0.15 mL/min to 0.45 mL/min. Figure 24 shows the results from this study. The consequent increase i n the flow rate of the HPLC pump shortened the analysis time by 15 minutes. A maximum flow rate of 0.3 mL/min from the HPLC pump with the peristaltic pump flow fine tuned at a flow rate of 0.23 mL/min was used i n subsequent experiments. 3.2.9 Detector wavelength Excitation and emission maxima of the derivatized digoxin were obtained by changing the excitation and emission f i l t e r combinations and are listed on Table 5. The emission maximum was found to be at 425 nm, whilst the excitation maximum occurred at 360 nm. Use of these f i l t e r s gave the maximum sensitivity for the detection of digoxin. 3.2.10 Band broadening In order to preserve the resolution of the column, the band broadening contribution of the reactor obviously should be as small as possible. Therefore, a comparison between the different reactors (Table 6) was made. An ultra-violet(UV) detector was inserted between the HPLC column and the post-column reactor system to measure relative band dispersion due to the reactor. Band 104 F i gu re24 Op t im i z a t i o n of HPLC f l ow rate TABLE 5. Filter Combinations for Optimum Detector Wavelength FILTERS nm (a) PEAK HEIGHT(cm) n=3 360/470 4.50 +.03 360/425 6.25 +.04 360/440 5.50 +.03 360/450 0 340/450 0 360/460 1.50 +.02 340/470 .0 340/440 0.50 +.01 340/425 0.60 +.02 (a) excitation/emission filters for Waters Fluorometer 106 TABLE 6 . BAND BROADENING OF REACTION SYSTEM DIMENSION TOTAL FLOW PEAK PEAK RESIDENCE' CONFIGURATION RATE (mL/min) HEIGHT (cm) WIDTH (min) TIME (min) I . AIR SEGMENTATION: I I . 3 m(L) X 9cm d i ame t e r 1 mm ID C o i l modi f i e d unmodi f i e d 0 .53 0 .53 3 .0 2 .75 2 .5 2 .8 15 15 I I . NON SEGMENTED REACTOR: 20 m(L) X 2.7cm d i ame t e r 0 . 3 mm ID c o i l 10 m(L) X 2.7cm d i ame t e r 0 .3 mm ID c o i l 10 m(L) X 0 . 3 mm ID k n i t t e d 0 .34 0 .53 (a ) 0 .40 (b) 0 .53 (a ) 0 .40 (b) 6 . 1 5 .5 5 .0 6 .0 5.3 3 .0 2 .0 2 .5 2 . 0 2 .2 I I I . UV d e t e c t i o n 0 .30 ( c ) 1.8 (a ) 0 .3 mL/min HPLC f l ow r a t e (b) 0 .15 mL/min HPLC f l ow r a t e ( c ) r e t e n t i o n t ime f o r d i g o x i n peak by UV d e t e c t i o n : 30 m i nu t e s broadening as determined from the peak width at base between UV and fluorescence detection of the different reactors investigated i s shown i n Table 6. The data i n Table 6 showed that the 10 m knitted configuration contributed the least to peak dispersion. This reactor was chosen as a compromise between the time needed for the reaction and the peak broadening due to the length of this c o i l , which might even destroy a separation already obtained in the column. Since other components of the LC system may also contribute to peak dispersion, we examined the effect of using a Rheodyne loop injector and the Waters U6K Universal injector. It was found that the latter gave less dispersion with better reproducibility between injections. The design of this injection port has been reported to allow the introduction of the sample as a narrow band, thus reducing the occurrence of asymmetrical t a i l i n g peaks. 3.2.11 Baseline noise Proper proportioning of the reagents i s important since the ratio of the reagent and effluent may vary with time, leading to noisy and/or drift i n g baseline and ragged looking bands. A smooth, constant delivery of the reagent i s essential, and this has been proposed to be most economically obtained with a nitrogen pressurized glass reagent reservoir (170). Unfortunately, the glass reservoirs available that are acid resistant are only rated to 25 108 psi, therefore making i t d i f f i c u l t to propel the reagents ef f i c i e n t l y through the 20 m reaction c o i l which has an estimated back pressure of 50 psi. In addition, the presence of HC1 in the reagent makes i t unsafe to operate. Therefore a peristaltic pump was, inspite of i t s shortcomings, used as the means of propelling the carrier streams. The main drawback of this type of pump i s that the stream i s never completely pulse-free. However, an improvement in the proportioning was achieved by frequent change of the acidflex pump tubes that delivered the HCl. The adjustment of the pressure of the acidflex tubing was also found to be important to achieve optimal s t a b i l i t y of the flow of the reagents. These minor manipulations of the delivery of the reagents also improved the day-to-day reproducibility of the post-column derivatization method. The mixing of effluent and reagents must occur before the completion of the reaction. Mixing i n non-segmented streams i s often slow. This problem has found to be alleviated by the use of a premixing PTFE c o i l before the addition of effluent and by the use of the knitted reactor (171). Poor mixing led to symptoms analogous to improper proportioning, mainly noisy baseline. An alternate approach to suppress excessive baseline noise i s the use of a noise f i l t e r . Thus, a noise f i l t e r (as described i n the experimental section 2.2.1) was constructed and f i t t e d between the detector and data system. The chromatograms i n Figure 25 ill u s t r a t e the difference obtained following incorporation of a noise f i l t e r . 109 F i gu r e25 . E f f e c t of the No i se F i l te r on the S igna l /No i se Rat io of D igox in A : W I T H O U T N O I S E F I L T E R B: W I T H N O I S E F I L T E R 110 3.2.12 Sensitivity The incorporation of^-cyclodextrin, a fluorescence enhancer, into the manifold was i n i t i a l l y proposed to increase the sensitivity of detection. Cyclodextrin, a monocyclic polymer i s known to form an inclusion complex with digoxin and this has been found to increase the fluorescence intensity of a variety of organic compounds (172,173). Several modifications such as,6-cyclodextrin concentration, detector wavelength and reaction times were considered. However, no apparent increase in fluorescence intensity was found when cyclodextrin was added into the reaction medium. This suggested that either there was no interaction of the ^-cyclodextrin with digoxin under this time condition, or that the polymer was being degraded i n the acidic environment as shown by Saenger (174). 3.2.13 Reproducibility and Calibration Curve The coefficient of variation obtained for repeated injections of standards of digoxin at the 3 ng level was 2% (n=10), indicating that the present method i s satisfactorily reproducible. A linear correlation between peak height ratios and concentration over the range of 1.5 ng to 10 ng per injection was observed (Table 7, Figure 26) which was adequate for current c l i n i c a l needs. Good reproducibility was also obtained and assessed by repeating the calibration procedure on 5 different days. An overall average of 8% coefficient of variation was found. I l l TABLE 7. Data for the Standard Curve Weight Ratio (a) Peak Height Ratio (b) + S.D. .022 .210 .01 .029 .300 .02 .044 .450 .04 .073 .730 .07 .102 1.02 .06 .147 1.47 .10 (a) digoxin weight/internal standard weight (b) digoxin peak height/internal standard peak height (mean of 5 days, repeated each day) 112 Figure 26 STANDARD CURVE OF DIGOXIN Weight Ratio 113 3.2.14 Choice of internal standard An internal standard i s a substance added to the sample at the earliest possible point in the analytical scheme to compensate for sample loss occurring during sample extraction, clean-up and f i n a l chromatographic analysis. The desirable characteristics of an internal standard are that i t w i l l co-extract with similar partition efficiency to that of the drug; that i t has similar detection characteristics to that of the analyte; that i t does not co-elute with the drug, metabolites or any biological endogenous substances and las t l y , that i t i s readily available. The choice of an internal standard i n the present method was further complicated by the fact that i t must, after passing through the post-column reactor, fluoresce at the chosen wavelengths (360/425 nm) similar to the drug being analyzed. In addition, the drug substance should not be a "co-administered" drug so that i t would not be expected i n the plasma of a significant number of patients. Within these limitations the following compounds were examined as possible internal standards: 17c<-methyl-testosterone (93 mins), ethinylestradiol (greater than 100 mins), prednisone (60 min), 17- estradiol (68 min), norethindrone (60 min), ouabain (18.4 min), cymarin (20 min), and digitoxigenin (40 min). Triamcinolone, ... r > diethylstilbestrol, mestranol and testosterone did not give any response. Most of the steroids have long retention times and were not practical as internal standards. Cymarin and ouabain, on the other hand, were too polar and co-eluted with the metabolites. 114 F igure 2 7 . S epa r a t i on of D i gox i n , i ts me t abo l i t e s and c o - adm in i s t e r ed Drugs Us ing Po s t - Co l umn De t e c t i on co + CN O) 20 40 To MINS 115 Chromatographic conditions: Mobile phase: Methanol-ethanol- isopropanol-water (52/3/1/45); Flow rate: 0.3 mL/min; post-column fluorescence detection using non-segmented reaction system as described i n Figure 7. Legends: 1. furosemide 2. dihydrodigoxigenin 3. digoxigenin 4. digoxigenin monodigitoxoside 5. digoxigenin bisdigitoxoside 6. dihydrodigoxin 7. digoxin 8. digitoxigenin (Int. Std.) 9. spironolactone 116 Digitoxigenin, however, f u l f i l l e d the c r i t e r i a for an internal standard. Figure 27 depicts the chromatogram of the separation of digoxin, i t s metabolites and digitoxigenin and co-administered drugs eluted with the quaternary solvent system. 3.3 Application of the analytical procedure to plasma samples 3.3.1 Cptimization of extraction method i n spiked water The efficiency of an extracting solvent depends primarily on the a f f i n i t y of the solute for the extracting solvent as measured by partition coefficient, the phase ratio (volume of extracting solvent/ volume of sample), and the number of extraction steps. These studies can be divided into liquid-liquid extractions, solid-phase or cartridge extractions and combination of these two methods. I n i t i a l experiments were carried out by extracting 1 ng/mL aliquots of digoxin in water (5 mL). The most appropriate methods i n terms of recovery and time were ultimately tested with plasma aliquots containing the same quantity of digoxin. A wide selection of pure solvents providing a wide range of solubility and selectivity properties was examined. Improved recovery of digoxirf'was observed by using a binary solvent.mixture. In addition to this optimum adjustment in solvent polarity, recovery could be further increased by using larger volumes of solvent for the same sample quantity, or by using multiple sequential 117 TABLE 8." SOLVENT-SOLVENT EXTRACTION OF DIGOXIN IN WATER P r o c e d u r e : 5 mL w a t e r c o n t a i n i n g 1 'ng/mL d i g o x i n E x t r a c t w i t h o r g a n i c s o l v e n t (see be low) Remove O r g a n i c s o l v e n t E v a p o r a t e O r g a n i c s o l v e n t under n i t r o g e n R e c o n s t i t u t e w i t h 100 uL HPLC m o b i l e phase O r g a n i c S o l v e n t s V o l u m e / r a t i o A s s ay Re cove r y D i c h l o r o m e t h a n e (DCM) 15 mL 31% II 2x 15,)mL 38% 3x 15 mL 47% II 4x 15 mL 69% E t h y l a c e t a t e 15 mL 36% DCM/ i s obu t ano l 15 :2 (10 mL) 34% DCM/propano l 15 :2 (10 mL) 41% DCM/ethano l 15 :2 (10 mLj ' 31% DCM/methanol 15 :2 (10 mL) 47% DCM/ i s op ropano l 15:2 (10 mL) 42% DCM/ i s op r opano l 15:1 (10 mL) 41% DCM/ i s op ropano l 15:5 (10 m l ) 42% DCM/ i s op r opano l 2x 15:1 (10 mL) 75% C h i o r o f o r m / m e t h a n o l 2x 6 :4 (10 mL) 55% C h l o r o f o r m / i s o p r o p a n o l 2X 9:1 (10 mL) 80% 118 extractions (see Table 8). There i s a limit to the phase volume of the extracting solvent and the number of extractions performed before the method becomes tedious. Results obtained from some of the organic solvents are presented i n Table 8. The data indicated that the highest recovery of the drug could be expected i n double extraction using either chloroform / isopropanol i n a ratio of 9:1 or dichloromethane / isopropanol i n a ratio of 15:1. Solid-phase cartridge or bonded phase sorbents introduced for off-line sample preparation are widely used for trace enrichment and sample clean-up. Chemically they are similar to the column packings used i n HPLC such as s i l i c a , diol (C2(OH)2) or octadecylsilane bonded to s i l i c a , except for i t s particle size (usually 40 um). The cartridge i s used to f a c i l i t a t e the sampling process and further increase the extraction yield. The sorbents are usually packaged i n disposable polypropylene columns sandwiched between two polyethylene f r i t s . The principle of the procedure involves the selective retaining of the compound of interest on the adsorbent as i t f i r s t passes through the column and then subsequently the analyte i s eluted with a specific solvent. Certain undesirable compounds which are adsorbed on the f i r s t pass can be selectively removed by washing with an intermediate solvent prior to f i n a l elution of the compound of interest (175) . The data given in Table 9 for the cartridge recovery of digoxin from water revealed an 81% yield when the elution solvent consisted of dichloromethane / isopropanol i n a ratio of 15:1. 119 TABLE § . S o l i d Phase C a r t r i d g e E x t r a c t i o n o f D i g o x i n i n Water P r o c e d u r e : E l u t e 5 mL o f w a t e r c o n t a i n i n g D i g o x i n ( Ing/mL) t h r ough 2 .8 mL c a r t r i d g e E x t r a c t c a r t r i d g e w i t h 10 mL O r g a n i c s o l v e n t ( see be low) E v a p o r a t e O r g a n i c S o l v e n t unde r n i t r o g e n R e c o n s t i t u t e R e s i d u e i n HPLC m o b i l e phase ( m e t h a n o l : Wa t e r . 50 :50 ) C a r t r i d g e O r g a n i c S o l v e n t s R a t i o o f Re cove r y So l ven t s (%) O c t a d e c y l S i l a n e C h l o r o f o r m / I s o p r o p a n o l 3 2 40 RP-18 fl ll 6 1 43 II II C h l o r o f o r m / M e t h a n o l 6 1 73 II II D i c h l o r o m e t h a n e / I s o p r o p a n o l 15 1 46 S i l i c a C h l o r o f o r m / I s o p r o p a n o l 3 2 65 n •I n 6 1 69 II C h l o r o f o r m / M e t h a n o l 6 1 69 II II 3 2 83 Di c h l o r o m e t h a n e / I s o p r o p a n o l 15 1 81 D i o l II II 15 1 81 C a r t r i dges o b t a i n e d f rom A n a l y t i c h e m I n t e r n a t i o n a l , Ha rbo r C i t y , C A . , USA 120 Although this yield was similar to the previous solvent-solvent extraction, i t was more efficient i n that a smaller volume of organic solvent had to be evaporated. 3.3.2 Optimization of extraction method i n plasma The rate and extent of extraction may be different for a solute i n a test system than i n a practical sample. Partial association of drug substances with protein i n plasma samples i s one instance where the extraction efficiency may vary substantially from that obtained using water as a model system. Therefore, the subsequent step i n the development of an extraction procedure i s the isolation of the drug from the plasma sample. The solvent-solvent recovery experiments of digoxin from plasma are given in Table 10. Although only two solvents are listed i n this Table they were representative of the results obtained for a number of solvents. In a l l cases, endogenous materials from the plasma were found to co-elute near digoxin and would therefore prevent i t s quantitation. Since the polarity of digoxin was not clearly delineated, there was no consistency in the results from various bonded phases. Different solvents were evaluated to determine (a) how clean an extract was produced in terms of chromatographic background interferences and (b) how well digoxin was extracted. From the i n i t i a l evaluation from water, dichloromethane / isopropanol showed the best yield with least chromatographic interference. However, the presence of an interfering peak i n the plasma complicated the extraction procedure. Simple cartridge extraction was not enough to 121 TABLE .10. SOLVENT-SOLVENT EXTRACTION OF DIGOXIN FROM PLASMA P r o c e d u r e : 3mL p lasma c o n t a i n i n g 1 ng/mL d i g o x i n Add 10 mL o r g a n i c s o l v e n t ( see be l ow) Tumble tube f o r 10 min to e x t r a c t C e n t r i f u g e a t 1000 rpm f o r 5 min Remove o r g a n i c s o l v e n t E v a p o r a t e o r g a n i c s o l v e n t under n i t r o g e n R e c o n s t i t u t e w i t h 100 uL HPLC m o b i l e phase (methano l : w a t e r , 65 :35 ) O r g a n i c S o l v e n t s R a t i o o f Re cove r y O r g a n i c s o l v e n t s (%) D i c h l o r o m e t h a n e / I s o p r o p a n o l 15:1 _* D i c h l o r o m e t h a n e _* * Due t o i n t e r f e r e n c e f rom p lasma c o n s t i t u e n t s , q u a n t i t a t i o n c o u l d no t be done . 122 isolate the drug. Therefore, variation of the extraction methods using the 3 types of solid phases was investigated. They included; (a) precipitation of the protein i n the plasma with a variety of organic solvents before i t s introduction through the cartridge, (b) prewashing of the adsorbed plasma sample to remove interfering endogenous compounds before i t s f i n a l elution from the cartridge, (c) a clean-up procedure by extraction with organic solvent before i t s passage through the cartridge (Table 11, 12). The overall results from these extractions indicated that digoxin was either not readily removed from the cartridge or that digoxin in the presence of a small amount of water from the plasma, when combined with alcohol did not allow the drug to be adsorbed on the cartridge. The combination of solvent extraction and cartridge adsorption(method C Table 11) i n i t i a l l y showed some promise. However, method C which combined a clean-up step and a cartridge extraction procedure was proven to be unreliable when plasma from a different source was used. In a l l cases, the extractions were found to become tedious and endogenous materials from plasma were found to interfere with the digoxin peak; hence, this method was abandoned. It i s obvious that more clean-up procedures were required prior to the LC analysis of the f i n a l sample extract. Since proteins are known to interfere with many analytical techniques, a deproteinizing step was introduced into the scheme. Reagents used to -: effect deproteinization were acetonitrile, trichloroacetic acid, dilute hydrochloric acid, methanol, acetone and a combination of zinc sulfate and barium hydroxide. The effect that each 123 TABLE .11. SOLID PHASE CARTRIDGE EXTRACTION OF DIGOXIN IN PLASMA P r o c e d u r e : 3' .mL p lasma c o n t a i n i n g D i g o x i n (1 ng/mL) P r e c i p i t a t e p lasma p r o t e i n s w i t h 3 mL a c e t o n i t r i l e C e n t r i f u g e a t 1000 rpm f o r 5 min Pass s u p e r n a t a n t t h r ough c a r t r i d g e D i s c a r d f i r s t e l u e n t f rom c a r t r i d g e E l u t e c a r t r i d g e w i t h o r g a n i c s o l v e n t E v apo r a t e o r g a n i c s o l v e n t unde r n i t r o g e n R e c o n s t i t u t e r e s i d u e i n HPLC m o b i l e phase C a r t r i d g e O r g a n i c S o l v e n t s ' (10 mL) R a t i o o f s o l v en t s Re co ve r y {%) D i o l D i o l S i l i c a D i c h l o r o m e t h a n e Di c h l o r o m e t h a n e / i s o p r o p a n o l II II 15:1 15:1 1 0 0 0 124 TABLE 12 SOLVENT-SOLVENT EXTRACTION OF DIGOXIN COMBINED WITH SOLID PHASE CARTRIDGE ISOLATION Method A 3 mL p lasma c o n t a i n i n g 1 ng/mL d i g o x i n E x t r a c t w i t h 10 mL O r g a n i c s o l v e n t 1 (see nex t page) Pass o r g a n i c s o l v e n t t h r ough c a r t r i d g e Wash c a r t r i d g e w i t h 2 mL w a t e r E l u t e w i t h o r g a n i c s o l v e n t (15 mL) E v apo r a t e o r g a n i c s o l v en t R e c o n s t i t u t e w i t h 100 uL HPLC m o b i l e phase Method B 3 mL p lasma c o n t a i n i n g 1 ng/mL d i g o x i n E x t r a c t w i t h 2x10 mL s o l v e n t 1 ( see nex t page) E v a p o r a t e o r g a n i c s o l v e n t 1 under n i t r o g e n R e d i s s o l v e d i g o x i n i n s o l v e n t 2 Pass s o l v e n t 2 t h r ough c a r t r i d g e E l u t e c a r t r i d g e w i t h o r g a n i c s o l v e n t 3 E v a p o r a t e o r g a n i c s o l v e n t 3 R e c o n s t i t u t e w i t h 100 uL HPLC m o b i l e phase Method C 3mL p lasma c o n t a i n i n g 1 ng/mL d i g o x i n E x t r a c t w i t h 2x10 mL s o l v e n t 1 (see n e x t page) E v a p o r a t e o r g a n i c s o l -v en t under n i t r o g e n R e d i s s o l v e i n s o l v e n t 2 Pass s o l v e n t 2 t h r o u g h c a r t r i d g e Wash c a r t r i d g e w i t h s o l v e n t 3 E l u t e c a r t r i d g e w i t h s o l v e n t 4 E v a p o r a t e o r g a n i c s o l v e n t under n i t r o g e n R e c o n s t i t u t e w i t h 100 uL HPLC m o b i l e phase c o n t i nued 125 Method o f E x t r a c t i o n C a r t r i d g e ORGANIC SOLVENT RATIOS S o l v e n t 1 S o l v e n t 2 AND VOLUME (a ) S o l v e n t 3 S o l v e n t 4 Re cove r y (%) (b) A DIOL DCM/MeOH ( 15 : 1 ) 10 mL N.D. B. DIOL DCM/IPA (15 :1 ) 2x10 mL wa t e r 1 mL DCM/IPA ( 15 : 1 ) 5 mL N.D. B S I L ICA CHL 1 mL IPA 5 mL 0 B C-18 " wa t e r 1 mL DCM/IPA ( 1 5 : 1 ) N.D. C DIOL n in ET20/ IPA ( 2 0 : 1 ) 3 mL DCM/IPA ( 15 : 1 ) 5 mL 0 C S I L ICA CHL 1 mL CHL 1 mL IPA 5 mL 0 C DIOL CHL 3 mL ETOAc 3 mL IPA 3x10 mL 75% (a ) S o l v e n t a b b r e v i a t i o n s : DCM = d i c h l o r o m e t h a n e ETOAc = e t h y l a c e t a t e IPA = i s o p r o p a n o l MeOH = methano l CHL = c h l o r o f o r m ET20 = d i e t h y l e t h e r (b) N.D. = none d e t e c t e d due to i n t e r f e r e n c e from p l a s m a . deproteinizing reagent could have on the assay of a drug must be carefully considered. For example, digoxin, an acid-labile drug would not withstand deproteinization with trichloroacetic acid and hydrochloric acid. Organic solvents, on the other hand, deproteinize by destroying quaternary and tertiary structures of the proteins and are thus unlikely to hydrolyze digoxin. Although acetonitrile was reported to precipitate 99.9% of proteins, the precipitate formed was found to trap some of the solvents therefore decreasing the extraction efficiency. Methanol, on the other hand, was found to be more d i f f i c u l t to evaporate. Among these deproteinizing agents, acetone was found to be the most convenient to use because i t has been reported to precipitate 99.1% of the protein (176) i n plasma, and i t i s water miscible and quite v o l a t i l e . By evaporating acetone from the plasma sample, subsequent extractions could be ef f i c i e n t l y performed with appropriate organic solvents. The presence of an interfering peak in some of the pooled plasma blanks, even after protein precipitation, prompted a search for an intermediate solvent to eliminate this endogenous interfering material. A variety of non-polar solvents were therefore examined. These were: hexane, heptane, benzene and isooctane. The most efficient solvent was found to be isooctane (Table 13) at a volume of 2mL isooctane to 3mL of plasma. Although liquid extraction i s simple and does not require complex equipment, i t i s not entirely free of practical problems. Extraction procedures using dichloromethane usually result i n an 127 TABLE 13. Recovery of Digoxin Using Different Non-Polar Solvent Washes Non-Polar Solvents Recovery n-2 hexane 55% heptane 65% benzene 40% iso-octane 70% 128 emulsion that i s not readily separated. More persistent emulsions were formed i n some of the plasma samples investigated and these could not be disrupted by high-speed centrifugation. As a consequence, the organic phase was f i l t e r e d through a Nylon 66 membrane f i l t e r . This step also served to protect the column f r i t from clogging due to debris contained i n the extractants. Finally, to optimize the extraction procedure with respect to recovery and specificity, the proportion and polarity of the alcohol in the extractant were re-examined. Quantities of 2, 6, 10, 20% isopropanol and 2% n-propanol i n dichloromethane were used i n this study. It was found that by increasing the amount of isopropanol i n dichloromethane, the area of the interfering peak also increased proportionately. By using n-propanol instead of isopropanol, a cleaner chromatogram could be obtained. Overall 2% n-propanol i n dichloromethane was found to be more selective as an extraction solvent than 2% isopropanol i n dichloromethane. In an attempt to f a c i l i t a t e the extraction procedure and further increase the extraction efficiency, the solid phase extraction was re-examined. This involved the deproteinization of the "spiked" plasma sample with acetone and delipidation with isooctane with subsequent evaporation of the acetone. The remaining supernatant was applied on the diol Bond Elut column and extracted with 10 mL 2% propanol i n dichloromethane. Although the extraction step was simplified, the recovery using the disposable mini-columns was found to be 30% less efficient than the manual extraction method. 129 The f i n a l extraction procedure used was based on the scheme described i n Figure 8. A typical HPLC elution pattern of digoxin standard and the internal standard, digitoxigenin, extracted from plasma i s shown in Figure 28. There were trace amounts of endogenous fluorescent contaminants which eluted before digoxin i n the plasma sample (Figure 28 blank). 3.3.3 Recovery and precision of the extraction method After an orderly logical progression through the development of the extraction procedure had been examined,1' an evaluation of this procedure i s always required. This involved the determination of i t s efficiency and accuracy. Percentage recoveries were determined using blank plasma samples spiked separately with digoxin at 2, 3, 5 ng/mL plasma levels. The results are shown in Table 14. "Spiked" plasma samples covering these same concentrations were also analyzed in t r i p l i c a t e on 3 different days to give the inter-assay coefficient of variation. An average recovery of 70% was found for these samples. 3.3.4 Calibration curve and method sensitivity In a linearity study, plasma samples at concentrations of 0.5 to 3.3 ng/mL were prepared, extracted and analyzed. Calibration samples were prepared from plasma samples taken from the blood bank of the Health Sciences Hospital, Acute Care Unit, UBC, Vane, Canada. The data obtained are summarized i n Table 15. Linear correlations between peak height ratios and weight ratios over the 130 Figure 2 8 . Typ i c a l Ch romatogram of a B lank and S p i k e d P l a sma i 1 1 — 0 3 0 6 0 m i n P lasma c o n t a i n i n g 3 ng o f d i g o x i n (1) and 80 ng o f d i g i t o x i g e n i n ( 2 ) ( i n t e r n a l s t a n d a r d ) . P lasma e x t r a c t e d a c c o r d i n g t o t h e e x t r a c t i o n p r o c e d u r e on F i g u r e 8 . 131 TABLE 14. R e p r o d u c i b i l i t y and Recovery Data C V . % Concentration Added % Recovery Intra-assay Inter-assay (ng/ 3mL) n=3 n=3 2 71 4 10 3 69 4 8 5 69 5 8 132 TABLE 15. Data for Plasma Standard Curve Weight Ratio (a) Peak Height Ratio (b) C.V.% .022 .277 + .025 cm 9' .029 .358 + .036 10 .044 .504 + .040 8 .073 .762 + .060 8 .103 1.16 + .098 8 .147 1.54 + .168 11 (a) digoxin weight/internal standard weight (b) digoxin peak height/internal standard peak height (mean of 5 days) 133 F igu re 29 134 concentration range were found. The coefficient of correlation calculated showed acceptable variation (Figure 29). The method's limi t of detection was found to be 0.5 ng/mL with a signal to noise ratio of 4 to 1. Therefore, the sensitivity of the method described i s appreciably the same as that observed i n the RIA method. 3.3.5 Specificity Patients undergoing therapy with digoxin may also be treated with a variety of other drugs depending on their overall state of health. Diuretics such as spironolactone and furosemide are frequently used i n cardiac patients to relieve the strain on the heart. On some occasions antiarrhythmic agents such as quinidine, verapamil or propafenone may be co-administered to aid i n regulating both the beat strength and rhythmicity of the heart beat. If present, these agents could interfere chromatographically with the digoxin peak. This interference problem could easily be circumvented by determining their potential for co-elution with digoxin. Drugs i n this study included furosemide, spironolactone, quinidine, procainamide, disopyramide, dipyridamole, verapamil, propafenone, captopril and other common drugs such as dioctyl sodium sulfosuccinate, trimethoprim and sulfamethoxazole. Only furosemide and spironolactone gave some response near the eluting peaks of interest (Figure 27). Included i n this chromatogram were the metabolites of digoxin present i n 5 ng amounts each. Although dihydrodigoxin was incompletely resolved from digoxin,' the interference was minor since the fluorescent intensity of this 135 metabolite was approximately one-half of the digoxin response. Figure 27 shows the separation of digoxin from i t s metabolites and co-administered drugs, indicating that the present method i s satisfactorily specific. The f i n a l test for the specificity of this method involved the extraction of a series of plasma samples obtained from the Red Cross, the blood bank of the Health Science hospital and samples from the staff and graduate students i n the Faculty of Pharmaceutical Sciences . It i s considered that a method that i s applicable to the pooled plasma of different sources i s indeed a robust method. The overall result of this experiment showed no interference i n the retention area of interest of digoxin. 3.3.6 Quality Control Procedure To assess the accuracy and precision of this method, a Lyphochek Radioassay control serum was used. This offers a t r i - l e v e l range of digoxin concentrations for intra-laboratory quality control. Assay values for Level I are in the low range, Level II i s mid-range and Level III i s elevated. The assay values obtained were subsequently compared to those determined by the Health Science hospital RIA method on the same sample (Figure 30). The results obtained during the analysis of the same sample by the two independent detection techniques are summarized i n Table 16. As shown i n Figure 30, i n levels I to I I I , a good correlation was found between the amounts of digoxin determined by HPLC post-column procedure and the RIA method. 136 Figure 3 0 . Compar i son of D igox in Dete rmina t ion Us ing HPLC -PC and RIA Method on Rad i oas say Con t ro l Serum ' i i 1 1 — / 0 .8 1.6 2.4 3.2 4 . 0 n g / m L RIA Method TABLE 16. Data f o r Q u a l i t y C o n t r o l Procedure LEVEL HPLC POST-COLUMN n=2 RIA (a) I 0.598 0.55 I I 2.07 2.25 I I I 3.09 3.24 (a) Nuclear Medical Labs (ng/mL) 138 3 . 4 D e t e r m i n a t i o n o f d i g o x i n i n p a t i e n t s ' p l a s m a b y H P L C p o s t - c o l u m n d e r i v a t i z a t i o n D i r e c t r a d i o i m m u n o a s s a y ( R I A ) o f d i g o x i n i n p l a s m a i s s u f f i c i e n t l y s e n s i t i v e , r a p i d a n d i n e x p e n s i v e b u t l a c k s t h e d e s i r e d s e l e c t i v i t y ( 1 1 0 ) . A p p a r e n t l y , h i g h c r o s s - r e a c t i v i t y v a l u e s h a v e b e e n r e p o r t e d b e t w e e n d i g o x i n a n t i b o d i e s a n d d i g o x i n m e t a b o l i t e s ( 1 1 0 ) . I n a d d i t i o n , t h e a n t i b o d i e s u s e d h a v e b e e n s h o w n t o c r o s s - r e a c t w i t h s e v e r a l c o - a d m i n i s t e r e d d r u g s ( 8 2 , 8 4 ) a s w e l l a s a " d i g o x i n - l i k e s u b s t a n c e " ( 1 1 8 ) p r e s e n t i n p l a s m a . T h i s i n t e r f e r e n c e m u s t b e e l i m i n a t e d , t o a v o i d t h e o v e r - e s t i m a t i o n o f d i g o x i n , e s p e c i a l l y b e c a u s e t h i s d r u g h a s a l o w t h e r a p e u t i c i n d e x . I n v i e w o f t h e o c c u r r e n c e o f f a l s e - p o s i t i v e d i g o x i n v a l u e s o b t a i n e d d u r i n g R I A a s s a y s , t h i s m e t h o d s h o u l d b e c o m p l e m e n t e d b y a n i n d e p e n d e n t c h e m i c a l m e t h o d w i t h a v e r y h i g h d e g r e e o f s p e c i f i c i t y . T h e m e t h o d p r e s e n t e d h e r e a l l o w s f o r t h e d e t e c t i o n o f d i g o x i n a t l e v e l s a t l e a s t e q u a l t o t h e u s e f u l d e t e c t i o n l i m i t e s t a b l i s h e d f o r t h e R I A p r o c e d u r e . I n a s s e s s i n g t h e a c c u r a c y o f t h e H P L C p o s t - c o l u m n m e t h o d , p a t i e n t s a m p l e s c o n t a i n i n g d i g o x i n , r a n g i n g f r o m s u b - t h e r a p e u t i c t o t o x i c c o n c e n t r a t i o n s w e r e p r o c e s s e d b y t h e H P L C p o s t - c o l u m n d e r i v a t i z a t i o n m e t h o d . R e s u l t s o b t a i n e d w e r e c o m p a r e d t o t h e R I A m e t h o d . A t o t a l o f 4 2 s a m p l e s ( 2 4 f r o m H e a l t h S c i e n c e A c u t e C a r e H o s p i t a l , 1 8 f r o m V a n c o u v e r G e n e r a l H o s p i t a l ) , c o l l e c t e d d u r i n g d i g i t a l i z a t i o n , w e r e e x t r a c t e d a n d a n a l y z e d f o r d i g o x i n . 1 3 9 TABLE 17. Digoxin i n Plasma o f D i g i t a l i z e d P a t i e n t s (Acute Care H o s p i t a l ) P a t i e n t s Digoxin Concentration ng/mL HPLC/RIA RATIO RIA HPLC-PC DETECTION 1 0.7 0.707 1.010 2 1.3 0.896 0.689 3 0.6 1.170 1.950 4 1.3 0.943 0.725 5 1.2 1.190 0.990 6 1.4 0.830 0.590 7 0.9 0.570 0.630 8 0.5 0.500 1.000 9 1.0 1.070 1.070 10 1.2 1.050 0.875 11 1.1 1.360 1.240 12 1.3 1.610 1.240 13 1.3 1.530 1.180 14 0.8 0.830 1.040 15 1.2 1.370 1.140 16 1.4 1.120 0.800 17 2.0 2.010 1.010 18 2.7 1.800 0.670 19 1.4 0.769 0.550 20 1.2 1.260 1.050 21 1.8 1.730 0.960 22 2.1 1.550 0.740 • 23 0.9 0.560 0.620 24 1.1 0.850 0.770 mean + S.D. (0.94 +0.30) continued 140 VGH P A T I E N T S D i g o x i n C o n c e n t r a t i o n n g / m L H P L C / R I A R A T I O R I A H P L C - P C D E T E C T I O N 1 0 0 0 2 0 . 3 0 . 3 0 0 1 . 0 3 2 . 3 1 . 7 2 0 0 . 7 5 4 2 . 0 1 . 0 7 0 0 . 5 3 5 5 2 . 5 2 . 2 2 0 0 . 8 8 8 6 1 . 7 0 . 8 7 6 0 . 5 1 5 7 1 . 3 1 . 9 7 0 1 . 5 2 0 8 0 . 8 1 . 4 0 0 1 . 7 5 0 9 1 . 2 0 . 8 2 5 0 . 6 8 8 10 1 . 3 1 . 1 4 0 0 . 8 7 7 11 0 . 7 0 . 7 8 0 1 . 1 1 0 12 1 . 3 1 . 1 2 0 0 . 8 6 2 13 0 . 6 0 . 8 6 0 1 . 4 3 0 14 1 . 1 1 . 4 0 0 1 . 2 7 0 1 5 0 . 8 0 . 9 1 1 . 1 4 0 16 0 . 9 0 . 8 5 0 0 . 9 4 0 17 2 . 0 1 . 5 2 0 0 . 7 6 0 18 0 . 5 0 . 5 0 0 1 . 0 0 0 mean + S . D . ( 1 . 0 + 0 . 3 4 ) 141 The results obtained during the analysis of the same samples by the two independent methods are summarized in Table 17. Average ratios of 0.94+0.3 (ACU) and 1.0+0.34 (VGH) were obtained. A range of 0.52 to 1.95 was found for these samples. This i s comparable with those results observed by Loo et a l . (7) (0.84+0.13) and the non-dialyzed patients observed by Gibson and Nelson (81) (1.06+0.09). From Table 17, ten of the samples showed a ratio of greater than one. Of the 6 patients from the Vancouver General Hospital whose medication profiles were followed, two patients who showed higher HPLC/RIA ratio (#13,14) were also on quinidine. The val i d i t y and significance of this finding requires further investigation. As shown in Figure 31, for samples with levels of 0.3 to 2.7 ng/mL, correlations of 0.78 (VGH) and .79 (ACU) were found between the amounts of digoxin determined by the two methods. A y-intercept of 0.334 with a T-test significance of less than 5% was found, suggesting that this intercept was not significantly different from zero. The regression line i n Figure 31 showed a consistent and appreciable bias between the two methods, with values being higher for the RIA procedure than for the HPLC post-column method. The quality c r i t e r i a (reproducibility i n section 3.3.3, linearity i n section 3.3.4, specificity in section 3.3.5 and comparison with standard sera with the RIA method i n section 3.3.6) showed that the HPLC post-column method i s accurate i n the determination of "true" digoxin concentrations. Although the results from Table 17 are comparable to the HPLC-RIA methods (7, 81), the method described here i s more advantageous in that i t allows for 142 Figure 3 1 . Co r r e l a t i on Be tween P lasma D igox in Leve l s by H P L C - P C and RIA Method 2.4 A i 1 1 1 j — 6 1.2 I B 2.4 3 .0 • V G H RIA Method (ng/mL) O A C U 143 unattended direct quantitation. The HPLC-RIA methods on the other hand, required a collection step after the chromatographic procedure. This step i s very c r i t i c a l and may introduce error i n the assay method. The reasons for the low correlation between the two methods may be due to several contributing factors. For example, i t has been reported that the digoxin metabolites cross-react extensively i n the RIA procedure (110). Dihydrodigoxin has also been shown to cross-react with the antibody(114). Recently, i t was also reported that substantial differences i n digoxin concentrations were found to exist when different immunoassay methods were used (118) . Such differences between these immunoassays raise the question about the determination of true, c l i n i c a l l y useful plasma digoxin concentrations by the current RIA method. The unavailability of a homogenous study population may have impaired the proper evaluation of the correlations between the two methods. Finally, while the RIA kits may be measuring an unknown endogenous substance i n the plasma, the HPLC method described here has been shown to be specific and accurate (as depicted in the quality control procedure section 3.3.6) for the measurements of the true plasma digoxin concentrations. 144 4. SUMMARY Aid CONCLUSIONS The potential of the proposed HPLC post^column method for the fluorescent detection of digoxin has been demonstrated. The method involves the separation of digoxin from i t s metabolites and post-column derivatization. The production of the fluorophore i s based on the reaction of hydrochloric acid with the steroid portion of the cardiac glycoside. Fluorescence of the derivative i s further enhanced by the addition of a hydrogen peroxide/ ascorbic acid mixture. The reactor described i s simple, easy to build and versatile. The method has the resolution and sensitivity desired for the analysis of digoxin i n plasma collected from patients who were undergoing digitalization. The relatively low correlation between the HPLC and the RIA methods supports the idea that the RIA procedure f a i l s to provide accurate measurement of digoxin levels i n plasma. This i s further supported by the knowledge that the RIA procedure measures drug and metabolite concentrations with equal reactivity. Hence, the HPLC procedure presented here provides for accurate determination of digoxin levels in plasma, free of interference from digoxin metabolites and many commonly co-administered drugs. 145 5 . REFERENCES 1. G .A . B e l l e r , T.W. S m i t h , W.H. Abe lmann , E. Habe r , W.B. Hood , J r . , New Engl J Med 284 , 989 (1971) 2 . F. Nachtmann, H. S p i t z y , and R.W. F r e i , Ana l Chem 48 ( 1 1 ) , 1576 ( 1 9 7 6 ) ; J Ch r oma t og r 122 , 293 (1976) 3 . Y . F u j i i , R. O g u r i , A . M i t s u h a s h i , and M. Y a m a z a k i , J Ch romatogr S c i • 2 1 , 495 (1983) 4 . J . C . K . L oo , I . J . M c G i l v e r a y , and N. J o r d a n , Res Comm Chem Path Pharm 1 7 ( 3 ) , 497 (1977) 5 . H.A . N e l s o n , S . V . L u c a s , and T . P . G i b s o n , J i Ch roma tog r 1 6 3 , 169 (1979) 6 . M.H. G a u l t , M. Ahmed, N. T i b b o , L. L o n g e r i c h , and D. Sugden , J Chromatogr 182 , 465 (1980) 7 . J . C . K . L oo , I . J . M c G i l v e r a y , and N. J o r d a n , J L i q Ch romatog r 4_(5), 879 (1981) 8 . J . A . M o r a i s , R .A . Z l o t e c k i , E. Sakmar , P . L . S t e t s o n , and J . G . Wagner , Res Comm Chem Path Pharm 31_(2), 285 (1981) 9 . J . C . G f e l l e r , G. F r e y , and R.W. F r e i , J Ch romatogr 142 , 271 (1977) 10 . J . E . Dohe r t y and J . J . Kane , C l i n i c a l Pha rmaco logy and T h e r a p e u t i c Use o f D i g i t a l i s G l y c o s i d e s . I n C a r d i o v a s c u l a r Drugs V o l . 1. A n t i a r r h y t h m i c , A n t i h y p e r t e n s i v e and L i p i d L owe r i n g D r u g s , ed by G . S . A v e r y , A d i s P r e s s , S ydney , ( 1 9 7 7 ) , p . 1 3 5 . 1 1 . K . - 0 . H a u s t e i n , Pharmac Ther 18, 1.( 1983) 12 . R. Thomas, J . B o u t a g y , and A . G e l b a r t , J Pharm S c i • 6 3 ( 1 1 ) , 1649 (1974) 1 3 . - B . F . Hof fman and J . J . B i g g e r J r . , D i g i t a l i s and A l l i e d C a r d i a c G l y c o s i d e s . In Goodman and G i l m a n ' s , The P h a r m a c o l o g i c a l B a s i s o f T h e r a p e u t i c s 6 th e d . by A . G . G i l man, L . S . Goodman, and A . G i l m a n , M a c M i l l a n P u b l i s h i n g C o . , I n c . , New York (1980) p. 7 30 . 14 . K. Repke , K l i n Wochenschr 4 2 , 157 (1964) 15 . G .A . L a n g e r , Fed P r o c Fed Am Soc Exp B i o l 3 6 ( 9 ) , 2231 (1977) 16 . D .G. A l l e n and J . R . B l i n k s , Na tu re 273 , 509 (1978) 1 7 . B . F . Hof fman and J . J . B i g g e r J r . , D i g i t a l i s and A l l i e d C a r d i a c G l y c o s i d e s . In Goodman and G i l m a n ' s , T h e P h a r m a c o l o g i c a l B a s i s o f T h e r a p e u t i c s 6 th ed by A . G . G i l m a n , L . S . Goodman, and A . G i l m a n , M a c M i l l a n P u b l i s h i n g C o . , I n c . , New York (1980) p. 750 . 1 8 . N. Toda and T . C . Wes t , J Pharmaco l Exp Ther 1 5 3 ( 1 ) , 104 (1966) 146 19 . Y . I . K im , R . J . N o b l e , and D .P . Z i p e s , Am J C a r d i o l 36„ 459 (1975) 2 0 . M.R. R o s e n , H. G e l b a n d , and B . F . Ho f fman , C i r c u l a t i o n 4 7 , 65 (1973) 2 1 . D .H . S i n g e r , R. L a z z a r a , and B . F . Hoffman , C i r Res 21_, 537 (1967) 2 2 . J . K . A r o n s o n , C l i n Pha rmacok i n e t 5_, 137 (1980) 2 3 . J . L i ndenbaum, Pharm Rev 25_(2), 229 (1973) 24 . N. S a n c h e z , L . B . S h e i n e r , H . H a l k i m , and K . L . Me lmon, Po r Med J 4 , 132(1973) 25 . L. F l e c k e n s t e i n , M. W e i n t r a u b , B. K r o e n i n g , and L. L a s a g n a , C l i n Pharmaco l Ther 1 5 ( 2 ) , 205 (1974) 26 . A . J . J o u n e l a , P . J . P e n t i k a i n e n , and A . So thmann , Europ J C l i n Pharmaco l 8_, 365 (1975) 27 . V . Mann i n en , A. A p a j a l a h t i , J . M e l i n , M. K a r e s o j a , L a n c e t ^ , 398 (1973) 28 . D.D. B rown , R . P . J uh l , . New Eng J Med 295_, 1034 (1976) 2 9 . W.H. H a l l , S ' .D. S h a p p e l l , and J . E. D o h e r t y , Am J C a r d i o l 3_9, 213 (1977) 30 . J . L i ndenbaum, R .M. M a u l i t z , V . P . B u t l e r , G a s t r o e n t e r o l o g y 71_, 399 (1976) 3 1 . W.H. Hal 1, J . E . D o h e r t y , i j . G a m m i l l , and J . She rwood , Am J Med 56_, 437 (1974) 32 . L. N y b e r g , K . E . A n d e r s s o n , and A . B e r t l e r , A c t a Pharm S u e c i c a J J , 459 (1974) 3 3 . E . S . Weiss and A . N . W e i s s , C o n g e s t i v e Hea r t F a i l u r e . In Manual o f M e d i c a l T h e r a p e u t i c s , ed by N.V. C o s t r i n i and W.M. Thomson, 22nd . e d i t i o n , L i t t l e , Brown and C o . , B o s t on (1977) p. 8 5 . 34 . P. K r amer , E. K o t h e , J . S a u l , and F. S c h e l e r , Eu r J C l i n I n v e s t 4 , 53 (1974) 35 . S . W a l l a c e and B. W h i t i n g , B r J C l i n Pharmac 1, 325 (1974) 36 . E . E . Ohnhaus , P. S p r i n g , and L. D e t t l i , Europ J C l i n Pharmaco l 5_, 34 (1972) 37 . J . R . Koup, W . J . J u s k o , C M . E l w o o d , and R.K. K o h l i , C l i n Pharmaco l The r 1 8 ( 1 ) , 9 (1975) 3 8 . T.W. S m i t h , Am J Med 5J3, 470 (1975) 39 . W . J . Ju sko and M. W e i n t r a u b , C l i n Pharmaco l Ther 16_(3), 449 (1974) 4 0 . H .G. G u l l n e r , E . B . S t i n s o n , D .C . H a r r i s o n , and S . M . Ka lman , C i r c u l a t i o n 50_, 653 (1974) 4 1 . J . E . D o h e r t y , W.H. P e r k i n s , and W . J . F l a n i g a n , Ann . I n t e r n Med 6 6 , 11.6 (U 4 2 . R .H . R e u n i n g , R .A . Sams, and R . E . N o t a r i , J C l i n Pharmaco l 13_, 127 (1973) 4 3 . J . G . Wagner , J C l i n Pharmaco l 14 , 329 (1974) 147 4 4 . G. W e t t r e l l , K . - E . A n d e r s s o n , A . B e r t l e r , and N.R. L u n d s t r o m , A c t a ; Ped i a t , r Scand 6 3 , 705 (1974) 4 5 . F . I . M a r c u s , Am J Med 5 8 , 452 (1975) 46 . J . E . D o h e r t y , Ann I n t e r n Med 7 9 ( 2 ) , 229 (1973) 4 7 . W . J . J u s k o , C l i n i c a l P h a r m a c o k i n e t i c s i n D i g o x i n , In C l i n i c a l Pharmaco-k i n e t i c s , ed by G. L e v y , Am Pharm A s s o c , WA, USA ( 1 9 7 4 ) , p. 3 1 . 4 8 . J . R . Koup, D . J . G r e e n b l a t t , W . J . J u s k o , T.W. S m i t h , and J . Koch -Wese r , J P h a r m a c o k i n e t B i opha rm 3_, 181 (1975) 4 9 . F . I . M a r c u s , G . J . K a p a d i a , and G .G . K a p a d i a , J Pharmaco l Exp Ther 145 , 203 (1964) 5 0 . D . J . Sumner , A . J . R u s s e l l , and B. W h i t i n g , B r J C l i n Pharmac 3_, 221 (1976) 5 1 . U. Ab shagen , H. RennekamD, R. K u c h l e r , and N. R i e t b r o c k , Eu rop J C l i n Pharmaco l 7 , 177 (1974) 5 2 . D.R. C l a r k and S . M . Ka lman , Drug Metab D i spos 2 , 148 (1974) 5 3 . A . Hernandez J r , N. Kouchoukos , R .M. B u r t o n , D. G o l d r i n g , P e d i a t r 3 1 , 952 (1963) 5 4 . E. Wa t son , D.R. C l a r k , and S . M . Ka lman , J Pharmaco l Exp Ther 1 8 4 ( 2 ) , 424 (1973) 5 5 . R . J . L u ch i and J .W. G r u b e r , Am J Med 45_, 322 (1968) 56 . U. P e t e r s , L . C . F a l k , and S . M . Kaimany A r ch I n t Med 138 , 1074 (1978) 5 7 . J . F . D o b k i n , H i^R-. S a n a , V . P . B u t l e r J r . , H .C . Neu , J . L i ndenbaum, Trans A s s o c ; Am Phys 95_, 22 (1982) 5 8 . J . L i ndenbaum, D. T s e - E n g , V . P . B u t l e r , D.G. Rund , Am J Med 7^, 67 (1981) 5 9 . J . L i ndenbaum, D.G. Rund , V . P . B u t l e r , D. T s e - E n g , and J . R . S a h a , New Engl J Med 3 05 , 789 (1981) 6 0 . M.H. G a u l t , D. Sugden , C. M a l o n e y , M. Ahmed, and M. Tweedda l e , C l i n Pharmaco l The r 2 5 , 499 (1979) 6 1 . H . J . D e n g l e r , G . Bodem, H . J . G i l d f r i c h , D i g o x i n P h a r m a c o k i n e t i c s and t h e i r R e l a t i o n t o C l i n i c a l Dosage P a r a m e t e r s . In I n t e r n a t i o n a l Symposium on C a r d i a c G l y c o s i d e s , ed by G. Bodem and H.G. D e n g l e r . S p r i n g e r - V o r l a g , B e r l i n - H e i d e l b e r g , New York ( 1 9 7 8 ) , p. 2 2 . 6 2 . T.W. S m i t h , L . H . G r e e n , and G .D . Cu r fman , C l i n i c a l I n t e r p r e t a t i o n o f Serum C o n c e n t r a t i o n s o f C a r d i a c G l y c o s i d e s , I b i d , p . 226 . 6 3 . J . K . A r o n s o n , B r J C l i n P h a r m a c y , 55 (1978) 6 4 . G .D . Johnson and D.G. M c D e V i t t , B r J C l i n Pharmaco l 7_, 435 (1979) 148 6 5 . J . K a r j a l a i n e n , K. O j a l a , P. R e i s s e l , A c t a Pharmaco l T o x i c o l 3_4, 385 (1974) 66 . H .G. G u l l n e r , E .B . S t i n s o n , D.C. H a r r i s o n , and S .M . Ka lman , C i r c u l a t i o n 5 0 , 653 (1974) 6 7 . J . K . A r o n s o n , C a r d i a c G l y c o s i d e s and d rugs used i n dy s rhy thm ia s . . In Dukes (ed ) S i d e e f f e c t s o f Drugs Annua l I I C h a p t e r 17a ( E x e r p t a M e d i c a , Amsterdam, 1978) 6 8 . J . A . I n g e l f i n g e r and P. Go ldman, N Eng l J Med 294_, 867 (1976) 6 9 . G .A . Ewy, G .G . K a p a d i a , L . Y a o , M. L u l l i n , and F . I . M a r c u s , C i r c u l a t i o n 3 9 , 449 (1969) 7 0 . K. Boman, A c t a Med Scand 214 , 345 (1983) 7 1 . M.S . C r ox son and H.K. I b b e r t s o n , B r Med J 3_, 566 (1975) 72 . J . E . D o h e r t y , W.H. P e r k i n s , Ann I n t e r n Med. 64_(3), 489 (1966) 73 . J . R . L aw rence , D . J . Sumner , W . J . K a l k , W.A. R a t c l i f f e , B. W h i t i n g , K. G r a y , and M. L i n d s a y , C l i n Pharmaco l Ther 2 2 ( 1 ) , 7 (1977) 74 . G .M. S h e n f i e l d , J . Thompson, and D.B . H o r n , Europ J C l i n Pharmaco l 12 , 437 (1977) 7 5 . D .H . Hu f fman , C D . K l a a s s e n , and C R . Ha r tman , C l i n Pharmaco l T h e r . 2 2 ( 5 ) , 533 (1977) 76 . K. W a t t e r s and G .H . Tomk i n , B r Med J 4 , 102 (1975) 7 7 . M. V e r on i and G .M . S h e n f i e l d , C l i n Exp Pharmaco l Phys ]_, 159 (1983) 78 . G .M. S h e n f i e l d , C l i n Pharmacok i ne t 6_, 275 (1981) 7 9 . F. K e l l e r , M. M o l z a h n , and R. I n g e r o w s k i , Eur J C l i n Pharmaco l 18 ,^ 433 (1980) 8 0 . M. F. P a u l s e n and P .G . W e l l i n g , J C l i n Pharmaco l 16_, 660 (1976) 8 1 . T . P . G i b s on and H.A. N e l s o n , C l i n Pharmaco l Ther 2_7(2), 219 (1980) 8 2 . B. S i l b e r , L . B . S h e i n e r , J . L . Powe r s , M.E . W i n t e r , and W. Sadee , C l i n Chem 2 5 ( 1 ) , 48 (1979) 8 3 . S . W a l d o r f f , J . D . A n d e r s e n , N. H e e b o l l - N i e l s e n , O .G . N i e l s e n , E. M o l t k e , U. S o r e n s e n , and E. S t e i n e s s , C l i n Pharmaco l Ther 2 4 ( 2 ) , 162 (1978) 8 4 . J . T . D i P i r o , J . R . C o l e , C R . D i P i r o , and J . A . B u s t r a c k , Am J .Ho sp Pharm 37_, 1518 (1980) 8 5 . R.W. Thomas and R.R. Maddox, The r Drug Mon 3_, 117 (1983) 8 6 . R. D a h l q v i s t , G. E j v i n s s o n , and K. S c h e n c k - G u s t a f s s o n , B r J C l i n Pharmac 9 , 413 (1980) 149 8 7 . E . B . Leahey J r . , J . A . R e i f f e l , E -G .V . G i a r d i n a , J . T . B i g g e r J r . , Ann I n t e r n Med. 9 2 , 605 (1980) 8 8 . R. P e n e l o p e , B. F o s s , and S . A . B e n e z r a , Di go x in; . In A n a l y t i c a l P r o f i l e s o f Drug S u b s t a n c e s , V o l . 9 , ed by K l a u s F l o r e y , Academic P r e s s , New Y o r k , ( 1 9 8 0 ) , p. 208 . 8 9 . H. B a i j e t , Pharm Weekbl 5 5 , 457 (1918) 9 0 . W.P. Raymond, A n a l y s t 6 3 , 478 (1938) 9 1 . D . L . Kedde , Pharm Weekb lad 8 2 , 741 (1947) 9 2 . G. R a b i t z s c h , V . Tambor , Pha rmaz i e 24 , 262 (1969) 9 3 . H. K i l i a n i , A r ch Pharm (We i nhe im ) , 234 , 273 (1896) 9 4 . C. K e l l e r , Be r Pharm Ges 5_, 275 (1895) 9 5 . M.M. P e s e z , Ann Pharm F r an c 10_, 104 (1952) 96 . R.W. J e l l i f f e , J Ch romatog r 2 7 , 172 (1967) 9 7 . R. T s c h e s c h e , G. G r immer , F. S e e h o f e r , Chem Ber 8 6 , 1235 (1953) 9 8 . E . D o e l k e r , I . K a p e t a n i d i s , and A . M i r i m a n o f f , Pharm A c t a He l v 44_, 647 (1969) 9 9 . R.W. J e l l i f f e , J Lab and C l i n Med 6 7 , 694 (1966) 100 . T.W. S m i t h , V . P . B u t l e r J r . , and E. H abe r , N Eng l J Med 281, 1212 (1969) 101 . B. B e r g d a h l , L M o l i n , L L i n d w a l l , G . D a h l s t r o m , I - L . S c h e r l i n g , and ' A . B e r t l e r , C l i n Chem25_(2) , 305 (1979) 102 . L. M o l i n and B. B e r g d a h l , C l i n Chem 2 9 ( 4 ) , 734 (1983) 103 . Y . S h i s h i b a , M. I r i e , H. Yamada, and F. K i n o s h i t a , C l i n Chem 29_(8), 1501 (1981) 104 . R.W. F r e i and A . H . M . T . S c h o l t e n , J Ch romatog r S c i 1_7, 152 (1979) 105 . N .P . K u b a s i k , B .B . B r o d y , S . S . B a r o l d , Am J C a r d o l , 3 6 , 975 (1975) 106 . J . R . H a n s e l l , Am J C l i n P a t h o l 6f5, 234 (1976) 107 . D . L . V o s h a l l , L . H u n t e r , H . J . G r a d y , C l i n Chem 21_, 402 (1975) 108 . J . L . H o l t z m a n , R .B . S h a f e r , and R .R . E r i c k s o n , C l i n Chem 20_(9),1194 (1974) 109 . J . L . C r a v e r and R. V a l d e s J r . , Ann I n t e r n Med 9 8 J 4 ) , 483 (1983) 110 . R .G . S t o l l , M .S . C h r i s t e n s e n , E. Sakmar , and J . G . Wagner , Res Commun Chem P a t h o l Pharmaco l 4 , 503 (1972) 1 1 1 . P . L . M a l i n i , A . M . M a r a t a , E. S t r o c c h i , and E. A m b r o s i o n i , C l i n Chem 2 8 ( 1 2 ) , 2445 (1982) 150 112 . V . P . B u t l e r J r . , and J . L i ndenbaum, Am J Med , 5J3, 460 (1975) 1 13 . E .K . Oge and P .A . P o l o n u s , C l i n Chem 2 4 ( 6 ) , 1086 (1978) 114 . W.G. K ramer , M .S . B a t h a l a , and R .H . R e u n i n g , Res Commun Chem Pa th Pharmaco l 1 4 ( 1 ) , 83 (1976) 115 . W.G. K ramer , N . L . K i n n e a r and H.K. Mo rgan , C l i n Chem 2 4 ( 1 ) , 155 (1978) 116 . R.W. F r e i , J Ch romatogr 1 65 , 75 (1979) 117 . P . E . R o l a n , D .B . F r e w i n , S . R . H a g l e y , Ann I n t e r n Med 99_(2), 280 (1983) 118 . S.W. G r a v e s , B . B r own , and R. V a l d e s J r . , Ann I n t e r n Med 99^,604 (1983) 119 . M.R. Pudek , D.W. Seccombe, and M.F . W h i t f i e l d , New Eng l J Med 3 0 8 ( 1 5 ) , 904 (1983) 1 20 . M.R. Pudek , D.W. Seccombe, B . E . J a c o b s o n , and M.F . W h i t f i e l d , C l i n Chem 2 9 ( 1 1 ) , 1972 (1983) 1 2 1 . R. V a l d e s J r . , S.W. G r a v e s , B .A . B rown , and M. L a n d t , J P e d i a t r 1 0 2 ( 6 ) , 947 (1983) 122 . S . D . Brunk and H. V . M a l m s t a d t , C l i n Chem 2 3 ( 6 ) , 1054 (1977) 1 23 . R .H . D r o s t , Th . A . P l omp , A . J . T e u n i s s e n , A . H . J . Maas , and R . A . A . Maes , C l i n Chem A c t a 7 9 , 557 (1977) 124 . L. L i n d a y and D .E . D r a y e r , C l i n Chem 2 9 ( 1 ) , 175 (1983) 125 . K. B o r n e r and N. R i e t b r o c k , J C l i n Chem C l i n B iochem 16_, 335 (1978) 126 . J . C . S k o u , B iochem B i ophy s Ac ta . 4 2 , 6 (1960) 127 . F . I . M a r c u s , L. B u r k h a l t e r , C. C u c c i a , J . P a v l o v i c h , G .G . K a p a d i a , C i r c u l a t i o n 3 4 , 865 (1966) 128 . G .G . B e l z and W. P f l e d e r e r , . B a s i c Res C a r d i o 7 0 , 142 (1975) 1 29 . G .G . B e l z and.W. H e i n z , A r z ne im F o r s c h / D r u g Res 27_ ( l ) , 653 (1977) 1 30 . S . L a d e r , A . B y e , and P. M a r s d e n , Eur J C l i n Pharmaco l 5_, 22 (1972) 1 3 1 . E. Wa t s on , and S . M . Ka lman , J Ch romatogr 5 6 , 209 (1971) 132 . R.W. J e l l i f f e , D .H . B l a n k e n h o r n , J Ch romatogr 6 9 , 157 (1972) 1 3 3 . E. Wa t son , P. T r eme l l , and S . M . Ka lman , J Ch romatogr 69_, 157 (1972) 134 . 0 . E i c h h o r s t and P . H . H i n d e r l i n g , J Ch romatog r 224 , 67 (1981) 135 . W;. L i n d n e r and R.W. F r e i , J Ch romatogr 117 , 81 (1976) 136 . F . J . E v a n s , J Ch romatogr 8 8 , 411 (1974) 137 . M.C. C a s t l e , J Ch romatog r 115 , 437 (1975) 151 138 . B. D e s t a , E. Kwong, and K.M. M c E r l a n e , J Chromatogr 240 , 137 (1982) 139 . P. S chauwecke r , R.W. F r e i , and F. E r n i , J Ch romatogr 136 , 63 (1977) 140 . B .M. E r i k s s o n , L . T e k e n b e r g s , J . O . Magnusson , and L. M o l i n , J Chromatogr 223 , 401 (1981) 1 4 1 . M. D o n i k e , J Ch romatogr 115_, 591 (1975) 1 42 . W.C. B u t t s and R . L . J o l l e y , C l i n Chem 16_, 722 (1970) 1 43 . F. B e n n i n g t o n , S . T . C h r i s t i a n , R .D . M o r i n , J Ch romatog r 106_, 435 (1975) 144 . R.W. F r e i and J . F . L aw rence , Ch r oma t og r aph i a 83_, 321. (1973) 145 . D.V. N a i k , J . S t ephen G r o o v e r , and S . G . Shu lman , Ana l Chim A c t a 7 4 , 29 (1975) 146 . I . M . J a k o v l j e v i c , Ana l Chem 35_(10) , 1513 (1963) 147 . The U n i t e d S t a t e s P h a r m a c o p e i a , T w e n t i e t h R e v i s i o n ( 1 9 8 0 ) , U n i t e d S t a t e s Pha rmacope i a l C o n v e n t i o n , I n c , R o c k v i l l e , Md, p . 2 4 3 . 148 . T . S e k i and Y Yamaguch i , J L i q Chromatogr 6_(6), 1131 (1983) 149 . R. H o r i k a w a , T . T a n i m u r a , and Z. Tamura , Ana l B iochem 85_, 105 (1978) 150 . G. E j v i n s s o n , B r Med J i : 279 (1978) 1 5 1 . W. D. Hage r , P. F e n s t e r , M. M a y e r s o h n , D. P e r r i e r , P. G r a v e s , F . I . M a r k u s , and S . Go ldman, N Eng l J Med 300 , 1238 (1979) 152 . J . K . A r o n s o n , J . G . C a r v e r , L a n c e t i : 1418 (1981) 1 5 3 . K. S c h e n c k - G u s t a f s s o n , R. D a h l q v i s t , B r J C l i n Pharmaco l (1981) _ U , 181 154 . H . 0 . K l e i n , R. Lang,, E .D . S e g n i , E. K a p l i n s k y , N Eng l J Med 3 0 3 , 160 (1980) 155 . G .G . B e l z , P . E . A u s t , R. Munkes , Lancet , - i : 844 (1981) 156 . J . O . Moysey , N . S . V . J a g g a r a o , E i .N. G rundy , D.A. C h a m b e r l a i n , B r Med J 2 8 2 , 272 (1981) 157 . E .G . Mandl;as,,D. Hun t , G. S l o m a n , A u s t NZ J Med 10 , 426: (1980) 158 . G. Schomburg , C a p i l l a r y Ch roma tog r aphy , F ou r t h I n t e r n a t i o n a l Sympos ium, May 3 - 7 , 1 9 8 1 , H u n d e l a n g , West Germany, K a i s e r , RE , e d , I n s t i t u t e o f Ch r oma tog r aphy , Bad. Dunkhe im, West Germany, 1 9 8 1 , p p . 3 7 1 - 4 0 4 . 1 5 9 . . Depar tment o f M e t a l l u r g y , p e r s o n a l c ommun i c a t i on 160 . L . R . Snyde r and J . J . K i r k l a n d - . In I n t r o d u c t i o n t o Modern L i q u i d Ch roma tog r aphy , John W i l e y and S o n s , I n c , New Y o r k , New Y o r k , 1979 , Chap 2 1 6 1 . I b i d , p . 260 152 162 . R. W. F r e i , L. M i c h e l , and W. S a n t i , J Ch romatogr 126 , 665 (1976) 1 6 3 . R.W. F r e i , L. M i c h e l , and W. S a n t i , J Ch romatogr 142 , 261 (1977) 164 . R . S . D e e l d e r , M .G . F . K r o l l , A . J . B . B e e r e n , J . H . M . van den B e r g , J Chromatogr 149 , 669 (1978) 165 . R.W. F r e i , R e a c t i o n D e t e c t o r s i n L i q u i d Ch roma tog raphy . In Chemica l D e r i v a t i z a t i o n i n A n a l y t i c a l C h e m i s t r y V o l . 1, Ch romatog raphy ed by R.W. F r e i and J . F . L a w r e n c e , P lenum P r e s s , New Yo,rk ( 1 9 8 1 ) , chap 4 . 166 . W.B. Furman: In Con t i nuou s F low A n a l y s i s , Theory and P r a c t i c e , Ma r c e l D e k k e r , I n c . , New Y o r k , New York ( 1 9 7 6 ) , p . 126 167 . K. Hofmann and I . H a l a s z , J Ch romatogr 199 , 3 (1980) 168 . M. U i h l e i n and E. Schwab, Ch r oma t g r aph i a 15_(2), 140 (1982) 169 . R . S . D e e l d e r , A . T . J . M . K u i j p e r s , and J . H . M . Van den B e r g , J Ch romatogr 255 , 543 (1983) 170 . R .T . K r a u s e , J Ch romatogr S c i 1_6, 281 (1978) 1 71 . L .R . S n y d e r , J Ch romatog r 125 , 287 (1976) 172 . F. C r ame r , W. S a e n g e r , H . C . S p a t z , J Am Chem Soc 8 9 , 14 (1967) 1 73 . T . K i n o s h i t a , F. I i n u m a , A . T s u j i , B iochem B i ophy s Res Commun 51_, 666 (1973) 174 . W. S a e n g e r , Angew Chem I n t Ed Eng l 1_9, 344 (1980) 175 . P r o d u c t I n f o r m a t i o n , A n a l y t i c h e m I n t e r n a t i o n a l , I n c . B o n d - e l u t 1 76 . J . B l a n c h a r d , J Ch romatogr 226 , 455 (1981) 177 . D. W e l l s , B. K a t z u n g , and F . H . M e y e r s , J Pharm Pharmaco l 13j" 389 . (1961 ) 153 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
https://iiif.library.ubc.ca/presentation/dsp.831.1-0096571/manifest

Comment

Related Items