UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Osteoporosis in cystic fibrosis : pathogenesis and clinical features Frangolias, Despina Daisy 2000

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

Item Metadata


831-ubc_2000-0399.pdf [ 5.17MB ]
JSON: 831-1.0089533.json
JSON-LD: 831-1.0089533-ld.json
RDF/XML (Pretty): 831-1.0089533-rdf.xml
RDF/JSON: 831-1.0089533-rdf.json
Turtle: 831-1.0089533-turtle.txt
N-Triples: 831-1.0089533-rdf-ntriples.txt
Original Record: 831-1.0089533-source.json
Full Text

Full Text

OSTEOPOROSIS IN CYSTIC FIBROSIS: Pathogenesis and Clinical Features by DESPEVA DAISY FRANGOLIAS B.P.E., The University of British Columbia, 1985 M.P.E., The University of British Columbia, 1993 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE IN THE FACULTY OF GRADUATE STUDIES (Department of Medicine; Experimental Medicine) We accept this thesis as conforming to the required standard T H E UNIVERSITY OF BRITISH COLUMBIA August 2000 © Despina Daisy Frangolias, 2000 In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. Department of ^ " f - H p g - C ' vy^e. The University of British Columbia Vancouver, Canada DE-6 (2/88) ii A B S T R A C T Objectives: Advances in the treatment of cystic fibrosis (CF) have lead to increased longevity and with it increased risk for a new set of complications including increased risk for osteoporosis. The purpose of the study was to elucidate the prevalence of low bone mineral density (BMD) in an adult C F population with heterogeneous severity of pulmonary disease and to ascertain clinical and laboratory correlates of low BMD. Methodology: The study design was a cross sectional investigation of B M D at one adult C F clinic. We measured spinal (Ll-4) and femoral B M D by dual energy x-ray absorptiometry (DEXA) in 68 (24 female) CF adults aged 18-55 years. The B M D standard scores for spine and femoral neck were averaged and the composite score used for analyses. Differences in disease severity, exercise capacity and physical activity level, dietary intake, body composition and body mass index, bone turnover, pubertal status, glucocorticoid use, vertebral and non-vertebral fracture rate, and back pain were investigated for associations with changes in BMD. Results: Out of our sample of 68 patients tested, we identified 58 patients with low B M D with 9 patients classified as osteoporotic. Low BMD was associated with declining pulmonary function and increasing age. Late diagnosis of C F was associated with low B M D in adulthood. Multiple regression analysis identified exercise capacity and body mass index as the most significant variables predicting changes in BMD. Current dietary intake was not predictive of present BMD, however inadequate intakes of calcium and vitamin D were documented. Body composition was moderately associated with BMD. Onset of puberty was delayed in those who showed decreased B M D in adulthood. Kyphosis and back pain were not predictive of low spinal BMD. Conclusions: Low B M D is common in adult C F patients. Increased severity of pulmonary disease, poor exercise tolerance and inadequate weight gain contribute to B M D deficits in adulthood. Delayed diagnosis of C F also contributes to low BMD in adulthood. Several other clinical and nutritional factors contribute to B M D deficits. iii T A B L E O F C O N T E N T S A B S T R A C T ii T A B L E OF CONTENTS iii LIST OF T A B L E S v LIST OF FIGURES vi G L O S S A R Y OF TERMS vii A C K N O W L E D G E M E N T S viii CHAPTER I OVERVIEW AND OBJECTIVES 1 1.1 Introduction 1 1.2 Statement of the Problem 7 1.3 Hypothesis 11 1.4 Significance 11 CHAPTER II METHODS A N D PROCEDURES 12 2.1 Sample and Eligibility Criteria .12 2.2 Study parameters 12 2.2.1 Bone densitometry 12 2.2.2 Body composition, physical growth and reproductive history 12 2.2.3 Exercise tests and physical activity questionnaire protocols 13 2.2.4 Lung function measurement 14 2.2.5 Brasfield and S-K scores derivation 15 2.2.6 Biochemical analysis 15 2.2.7 Radiology, fractures, posture and back pain determination 16 2.3 Study Design and Statistical Analysis 16 C H A P T E R III RESULTS 18 3.1 Descriptive characteristics of study sample and primary hypothesis 18 3.2 Prevalence of low B M D 18 3.3 Nutritional status and B M D 19 3.4 Exercise and B M D 20 3.5 Association of back pain, kyphosis and vertebral wedging with B M D 20 3.5.1 Kyphosis and cumulative vertebral fractures 21 3.5.2 Back Pain 21 3.6 Puberty and B M D 21 3.7 Prevalence of nonvertebral fractures and B M D 22 3.8 Steroid use and B M D 22 C H A P T E R IV DISCUSSION 44 4.1 Predictability of B M D and prevalence of low B M D in adult CF patients 44 4.2 Exercise and B M D 45 4.3 Growth and Reproductive function 48 4.4 Nutritional 49 4.5 Vertebral fractures, kyphosis and association with back pain 53 4.6 Nonvertebral fractures 55 4.7 Steroid use 55 iv C H A P T E R V CONCLUSION 57 5.0 Concluding comments 57 6.0 BIBLIOGRAPHY 58 7.1 APPENDIX A 64 7.2 APPENDIX B 83 LIST O F T A B L E S Table 1. Factors suggested affecting bone mineral density in CF patients 7 Table 2. Summary of literature findings 8 Table 3. Categorization of mode of exercise and summary of activities reported by study sample 14 Table 4. Total sample available for analysis 17 Table 5. Descriptive characteristics of the study sample 23 Table 6. Regression models for composite lumbar spine and femoral B M D 24 Table 7. B M D and disease severity between B M D classifications 27 Table 8. A N O V A results for nutritional predictor variables by B M D grouping 28 Table 9. Intercorrelations for composite lumbar spine and femoral B M D with nutritional predictor variables 30 Table 10. Dietary intake from foods and supplements for the 7 patients on dietary liquid replacement food 31 Table 11. Subjective rating of overall activity level 32 Table 12. V 0 2 m a x results by BMD grouping 32 Table 13. Exercise diary results by B M D grouping 33 Table 14. Exercise habits at puberty and adulthood 33 Table 15. Kyphosis and T8-T12 and B M D 33 Table 16. Regression models for lumbar spine B M D with kyphosis and T8-T12 34 Table 17. Intercorrelations for lumbar spine B M D and measures of kyphosis and pulmonary disease severity 35 Table 18. Tabulation of whether subjects experienced back pain more than once over the past year tabulated by B M D status 35 Table 19. Location of back pain by B M D status 35 Table 20. Severity of back pain and B M D status 36 Table 21. Frequency of back pain and B M D status 36 Table 22. Identification of activities patients reported back pain to interfere with 36 Table 23. A N O V A results for variables related to pubertal onset by B M D grouping 37 Table 24. A N O V A results for B M D by menstrual history 37 Table 25. Incidence of nonvertebral fractures by B M D group 38 Table 26. Corticosteroid use during puberty and adulthood 39 Table 27. Data from liver transplantation patient. The patient is on total enteral nutrition 85 ( vi L IST O F F I G U R E S Figure 1. Pancreatic sufficiency and B M D status 40 Figure 2. Comparison of daily caloric intake with estimated energy requirements and expenditures 41 Figure 3. Dietary intake of calcium from food and supplements by B M D status 42 Figure 4. Dietary intake of vitamin D from food and supplements by B M D status 43 G L O S S A R Y O F T E R M S A A Aquatic activities A C S M American College of Sports Medicine B M A D Bone mineral apparent density B M C Bone mineral content B M D Bone mineral density BMI Body mass index BS-APase Bone specific alkaline phosphatase CFTR Cystic fibrosis transmembrane regulator C T Computed tomograaphy D E E Estimated daily energy expenditure D E R Estimated daily energy requirement or intake D E X A Dual-energy x-ray absorptiometry D P A Dual-photon absorptiometry HR Heart rate L B - H Land-based load bearing-moderate to hard L B - L Land-based load bearing-light L B - S G Land-based load bearing-Sports/ Games %predFEV, Percent predicted forced expiratory volume in one second S-K Schwachman-Kulczycki clinical scores SPA Single-photon absorptiometry T8-T12 Thoracic vertebrae 8 through 12 T C -BMD (Lumbar spine B M D T score + femoral neck B M D T-score) + 2 T-lumbar spine Bone mineral density T-score for lumbar spine T-femoral neck Bone mineral density T-score for femoral neck vo2max Maximal oxygen consumption 25-OHD 25 hydroxy-vitamin D 1,25-di-OHD 1,25 dihydroxy-vitamin D Vlll A C K N O W L E D G E M E N T S I thank the patients who took the time out of their busy life to participate in the study. This study would not have run as smoothly if it weren't for the commitment to research and this study of the St. Paul's Hospital Cystic Fibrosis clinic and Nuclear Medicine staff. This study was funded by a grant by the British Columbia Lung Foundation. I thank my committee members (Drs. David Kendler, Darlene Reid, Peter Pare) and supervisor Dr. Pearce Wilcox for their help and support. A special thank-you to Peter for helping me fine-tune my writing skills. I would also like to acknowledge my husband Brian Charles for his support. I would also like to acknowledge and thank my mother and father in law Lynne and Jim Charles and my parents Steve and Toula Frangolias for, among other things, taking care of our son Stevie while I was writing up. I also have to acknowledge two very important four-legged friends of mine (Kesar the cat and Fendi the dog) who have always stayed up with me until the wee hours of the morning (when everyone else was asleep) when I was studying and as I was writing up this thesis. This thesis is dedicated to my dog Fendi who recently and suddenly passed away. He was my protector, my running partner, and my best friend. You will never be forgotten. You will always be in our hearts and thoughts. Cystic Fibrosis Pacific Spirit fun run at UBC (1995) November 8, 1988 - December 10, 1999 1 CHAPTER I OVERVIEW AND OBJECTIVES 1.1 INTRODUCTION Cystic fibrosis affects approximately 1 in 2000 live births in Canada and the United States (1). Cystic fibrosis is a systemic disorder transmitted as an autosomal recessive trait (2). In 1989, the CF gene responsible for the CF phenotype was identified (3). The CF gene codes for a cellular membrane protein, the CF transmembrane regulator (CFTR), which forms a chloride channel and also regulates other channel proteins. The CF gene is expressed in the epithelial cells of many organs including the pancreas, lungs and gastrointestinal (GI) tract. The mutation disrupts exocrine function by causing abnormal regulation of epithelial ion transport. Chronic and recurrent pulmonary infections and digestive disorders characterize the condition. A large number of mutations have been identified in the C F T R gene, but AF508 is the most common. Although some genotypes are associated with less severe disease, patients possessing the same genotype show great variation in disease severity and progression (4). With advances in treatment, patients who have CF are living beyond their adolescent years into adulthood. The median age of survival has increased to well into the third decade. A number of long-term medical complications are being seen in tins ageing group of patients including diabetes, liver disease, infertility and osteoporosis (5). As life expectancy continues to improve in this group of patients it is important to gain a better understanding of the risk factors for the development of these complications and explore possible mechanisms involved in their development. This review will focus on risk factors and mechanisms involved in the development of osteoporosis in this population. Osteoporosis is characterized by loss of bone mineral and increased risk of bone fracture. Osteoporosis is a common complication of organ transplantation and the condition is further aggravated if there is preexisting bone loss prior to transplantation (6). Peak bone mass is achieved between the ages of 20-30 years and is tied closely to events during childhood and adolescence (7). Parameters thought to contribute to attained peak bone mass in the normal population are dietary intake of calcium and other nutritional factors (vitamin D, caloric intake), weight, activity level, genetic factors and in women reproductive and menstrual history (8). The most common cause of premature death in C F patients is respiratory failure. However the pancreatic and intestinal disorders, which are also characteristic in this population, are likely to be important factors in bone development and the maintenance of bone mineral density (BMD) in these patients. The CF gene codes for the cellular membrane protein CFTR, which forms a chloride channel, regulated by phosphorylation by protein kinase A. The gene defect results in failure of chloride secretion and enhanced sodium absorption, which leads to excessive water absorption in the airway mucosal cells. The cilia on the bronchial epithelial walls are normal in CF and normal beat frequency is reported (9). In the 2 non-CF individual the cilia are bathed in a layer of fluid, which helps the cilia sweep inhaled particles and potential pathogens out of the lungs. The ability of the cilia to function effectively is compromised in CF and leads to the obstruction by mucus plugs of the glands and peripheral airways (10). Over time, changes in the lining of the epithelium occur and viscid mucous accumulation not only interferes with normal ciliary transport but also provides a breeding ground for bacteria, which accumulate in the lumen of the airways (1). The majority of studies have shown CF lungs to be normal in utero and immediately after birth (11-13), except for dilated airway submucosal gland ducts (11). Within the first few days postpartum, however histological abnormalities can be observed, even before onset of lung infection can be detected clinically. Oppenheimer and Esterly, showed submucosal gland hypertrophy, gland obstruction and mucous cell hyperplasia of the trachea and major bronchi in CF infants who died from non-respiratory causes, specifically from meconium ileus (14). With the onset of infection, the airways become filled with thick mucopurulent material containing bacterial colonies, neutrophils and thick mucus. Once such pathogens as Pseudomonas aeruginosa, Staph aureus and Burkholderia cepacia are isolated from the airways of CF patients, chronic colonization is the norm. CF patients are thus prone to repeated infectious pulmonary exacerbations which over time cause bronchiolitis and mucus impaction, which obstructs the bronchi and bronchioles and ultimately leads to bronchiectatsis (10). This pattern has been observed as early as 2-month postpartum (15). Chronic pulmonary infections can potentially influence B M D in C F patients. Increase in cytokines and nitric oxide, which occurs with infection, can affect B M D . An increase in interleukin-6 and tumor necrosis factor have been shown to increase bone resorption (16,17). Similarly, nitric oxide, which may increase with infection or inflammation, has also been shown to enhance bone resorption (18). Indirectly, recurrent and frequent pulmonary infections are likely to reduce the activity levels of patients as well as curb their appetite during their bouts of illness. Bed rest contributes to accelerated bone loss and with the frequent pulmonary infections and associated fatigue in this group, they are prone to prolonged bed rest (8). Increasing severity of pulmonary disease and the associated sense of breathlessness and fatigue may also contribute to reduced activity levels and prolonged bed rest and consequently contribute to accelerated bone loss. As with the respiratory manifestations of CF, the phenotypic characteristics in the pancreas and GI tract are related to the failure of epithelial ion transport. The main cause for abnormalities in the pancreas and gastrointestinal tract is blockage of the secretory ducts, which over time leads to atrophy, fibrosis and destruction of the tissue. Both the endocrine and exocrine pancreas are affected in CF, but it is the progressive interstitial fibrosis of the exocrine pancreas with inspissation of secretions in the ducts and 3 acini that are relevant to bone growth and maintenance. Morphological changes (eosinophilic secretions with dilatation of the ductules) in the exocrine pancreas can be seen as early as 18-20 weeks prenatally. The initial severity and progression of pancreatic insufficiency may differ in CF patients, but the majority will experience progressive obstruction of the pancreatic ducts by viscous secretions ultimately leading to complete acinar atrophy accompanied by fibrosis and lipomatosis of exocrine pancreatic tissue. At autopsy, the pancreas in patients who have CF is fibrosed and fatty with dilated ducts filled with secretions (14). With the destruction of the exocrine pancreas, enzymes required for digestion of protein and fat are lost and the consequence is malabsorption of protein, fat and fat-soluble vitamins. As stated, the pancreatic insufficiency, which affects the majority of CF patients, results in the malabsorption of fat, protein and vitamins and is expected to affect bone mineral density (BMD). Despite vitamin D supplementation at levels above those recommended for the normal population, reduced levels of plasma 25-hydroxyvitamin D (25-OH vitamin D) have been found in C F patients (19-22). Pancreatic enzyme supplementation improves, but does not always normalize absorption. Such factors as reduced fitness due to a decline in physical activity, malnutrition and weight loss (as in anorexia nervosa patients), diseased states (e.g., chronic renal failure), aging, and menopause have been shown to have deleterious effects on bone mineral density (BMD) in the non-CF population. Malnutrition and under nutrition due to malaise and inadequate pancreatic enzyme supplementation is seen in C F patients and is suggested to have similar effects on B M D as shown iii patients who have eating disorders. Increased survival of C F patients into dieir thirties and forties also introduces this patient group to the effects of aging and menopause on B M D . Early studies paint a contradictory picture of bone mineral status in CF. These discrepancies may reflect methodological differences such as lack of sensitivity of early bone density measurement methods, the skeletal site studied, subject selection criteria, and small sample size. Bone densitometry techniques have become increasingly useful in both the diagnosis and the management of low B M D . Early studies used single-photon absorptiometry (SPA) and dual-photon absorptiometry (DPA) to measure BMD. These methods use a radioisotope source and measure bone mineral content (in grams) and areal bone mineral density (g/cm2). The major disadvantage of SPA is that it incapable of distinguishing between absorption from tissue or bone and that this technique is limited to peripheral skeletal measurement (e.g., radius) (23). DPA allows one to distinguish between absorption from soft tissue and bone and can measure BMD of the lumbar spine, proximal femur and total body but is hindered by the low intensity of radiation provided by the isotope source (23). Isotope based absorptiometers have been recently replaced by dual x-ray absorptiometry (DEXA) which use an x-ray source. D E X A provides greater precision, faster scanning time, and can be used to measure areal bone mineral density of the lumbar spine, various regions of the femur (femoral neck, trochanter and Ward's triangle) and radius and also total body bone mineral 4 density. While with this method patients are exposed to radiation, the level of exposure is low (less than radiation exposure from a chest x-ray (25mrem) or full dental x-ray). Quantitative computed tomography (CT) has also been used for the non-invasive measurement of BMD. The major benefit of quantitative CT for B M D measurement is the ability to measure trabecular microarchitecture, however there are a number of shortcomings which include level of radiation exposure (200-1000mrem compared to 2-4mrem for D E X A ) and results are obscured by the presence of subcutaneous tissue including intramedullary fat. Ultrasound is another method being investigated for use in measurement of B M D and offers a number of advantages, which include portability, low cost, assessment of bone microarchitecture, and no radiation exposure (24). Forearm bone mass using single-photon absorptiometry was used in early studies (21,22,25) to look at B M D and consequently they may have missed bone mass deficits in the hip and spine. Gibbens and colleagues measured trabecular vertebral bone density using quantitative CT , but excluded from their study sample CF patients on corticosteroids, hormone replacement therapy, or vitamin D and calcium supplementation (26). The patient population evaluated in these studies may also have influenced reported B M D results; subjects were predominately young and were characterized as having less severe pulmonary disease. These early studies did show lower B M D values in some C F patients compared to age-matched healthy controls (22,25,26) and this difference was sometimes associated with serum 25-OH vitamin D and calcium concentrations that were below normal values even in CF patients who were on oral supplementation programs (21,22,25). Lower bone mineral density scores were also associated with greater pulmonary disease severity measured by spirometry (26). Ross and associates, who did not look at bone mineral density but instead at causes of back pain in CF patients aged 10-36 years, reported that back pain was due to postural abnormalities including decreased muscle strength and mobility in the shoulders, chest and truck, or to vertebral wedging. Grey and colleagues reported that C F patients who exhibited vertebral wedging were more likely to have spinal B M D in the osteopenic range (27). Henderson and Specter, found an increased prevalence of excessive thoracic kyphosis and an increased fracture rate in CF patients aged 4-22 years (28). In a follow-up study Henderson and Madsen evaluated bone mineral density (lumbar spine and proximal femur using DEXA) in their original sample (29). Patients who showed excessive thoracic kyphosis also had lower spinal B M D than those with normal kyphosis. They also found significant correlations between increasing age and pulmonary disease and declining BMD. They also showed poor nutritional status (i.e., weight) to be associated with low B M D . Other studies have shown similar relationships (19,27,30-33). Aris and colleagues measured B M D in 20 02-dependent CF patients who had end-stage respiratory disease, 15 patients who had COPD, and 20 patients who had a variety of other pulmonary disorders all of 5 whom were awaiting lung transplantation (6). The CF group displayed significantly lower B M D values than the other two groups both pre- and post-transplantation. The majority of the C F patients exhibited B M D values, which were below the fracture threshold (Z-score>2SD's below the mean). CF patients accounted for the only fractures in the pre-transplantation period (3 cases) and for 50% of reported fractures post-transplantation (5 cases). The COPD cases accounted for the other 50% of post-transplantation fracture cases. It is interesting to note that the CF patients were younger (age range: 19-35 years) than the COPD patients (age range: 49-57 years) who reported post-transplant fractures. Ribs, spine, ulna and radius were the most common fracture sites. Bachrach and associates and Bhudhikanok and colleagues reported rib, forearm and hip fractures in their study of CF patients who had moderate/severe pulmonary disease and who also had a history of long-term prednisone therapy and recurring pulmonary exacerbations (32,34). Donovan and associates also reported rib, wrist and digit fractures in their CF patients (31). In this latter study as in the above studies, they did not find a strong association between fracture rate and low B M D Bachrach and colleagues measured B M D in an older cohort of CF patients (aged 18-42 years, mean=27 years), but also expressed B M D corrected for bone size to account for the smaller skeletal frame of most CF patients (i.e., bone mineral apparent density: BMAD) (32). B M D and B M A D were both reduced in CF patients. The low B M A D confirmed that their cohort's low B M D was not related to their smaller body frame and similar findings have been reported by Shane and co-workers (35). This former research group has presented similar findings in a more recent larger scale longitudinal study as well (34). These findings concur with previous beliefs that poor nutrition and pulmonary complications are responsible for delayed puberty and skeletal maturation in this population. Consequently, the low B M D found in these studies is likely partially due to failure of B M D to increase at a normal rate in the CF patients during the developmental years. CF is frequently associated with malabsorption and poor nutrition. These factors may result in poor growth, low body mass, deficiency in fat-soluble vitamins (including vitamin D), low calcium and phosphorus intake. Results of studies have shown low B M D to be more common in C F patients who are underweight (26,32,34). Salamoni and associates measured B M D and lean body mass in 14 young CF patients who had mild pulmonary disease and were well nourished (36). The CF group was matched to normal controls for age, pubertal stage, gender, height and weight. They found no differences in B M D , lean body mass, and plasma levels of calcium and 25-hydroxyvitamin D between the CF group and matched normal controls. They showed a strong positive relationship between lean body mass and bone mineral content for both groups. Similarly, Flores and associates did not find a difference in lean body mass and fat mass between adolescent CF females (age range 12-17 years) who had mild disease and age matched normal controls, however they showed the CF group to have lower B M D values (33). 25-OH 6 vitamin D levels have been commonly reported to be in the lower range of normal or below normal in CF research study cohorts (19-22,27,29,31,32,34,35), despite oral supplementation (of 400-900 IU of vitamin D) in most patients (19,20,22,29,31,34). Despite marginal or deficient serum 25-OH vitamin D levels, serum calcium and phosphorus levels are usually within the normal range (19,27,29,31,32,34-36). Hypogonadism, delayed puberty and glucocorticoid use are additional risk factors for low B M D . Males who have CF and low serum testosterone levels have been shown to have lower B M D than patients with normal serum testosterone levels (32,34). The results of previous studies have also shown that C F patients who have a history of delayed puberty have lower B M D (21,22,27,29,31,32,34) and B M A D levels in the osteopenic range (31,32,34). There is some experimental support for a relationship between delayed puberty and lower B M D in adulthood in the CF population (29,37), although sample sizes reported in these studies have been small and thus difficult to make a definitive decision on the magnitude of the contribution of delayed puberty to reduced BMD. Recent studies have shown that female CF patients who are amenorrheic exhibit low B M D values (32,34,38). Based on these findings it has been suggested that the risk of developing low B M D in adulthood can be reduced if bone mineral acquisition is adequate during the first 2 decades of life (27,31,32,34,38,39). Bhudhikanok and colleagues showed a higher frequency of fractures in CF patients who were on long-term prednisone therapy. In most cases severity of pulmonary disease seems to be a good predictor of B M D (34). The results of previous studies have generally shown an inverse association between pulmonary disease severity (i.e., S-K scores or pulmonary function status) and B M D (21,22,26,27,29,31,33,34). Haworth and co-workers, in a sample of 143 C F patients showed that there might also be an association between B M D and CFTR genotype (19). They found an increased prevalence of low B M D in CF patients who were homozygous for the AF508 mutation, others (on smaller cohorts) have shown no specific association between CFTR genotype and B M D (34,36). Table 1 and 2 summarize research findings to date. 7 Table 1. Factors suggested affecting bone mineral density in CF patients. Parameter Affect B M D in CF Do not affect B M D i n C F Data unavailable for CF Lung disease severity V Pancreatic insufficiency V Nutritional status Body weight or BMI Lean muscle mass V Vitamin D intake Calcium intake V V Physical aaiviU V 0 2 m a x Exercise activity status (active or sedentary lifestyle) V V Ph\sical urowlli & repiodiicli\o function Tanner staging for pubertal development V Bone size development Testosterone levels V Menstrual cycle abnormalities Other ftctois CFTR genotype V Long term corticosteroid use V V 1.2 STATEMENT OF THE P R O B L E M As life expectancy for C F patients continues to increase, it is important to determine the potential increase in symptoms and disability related to decreasing bone mineral density in this population. Previous studies have shown increased prevalence of low B M D in children who have CF and in adult CF patients predominately with severe pulmonary disease. Two main questions that this study will address are: • What is the prevalence of low B M D in an adult CF population with heterogeneous severity of pulmonary disease? • What are the predictors of low BMD? 0 0 1 o cd 0 oo C N CJ Predictors of low BMD & CJ > pa j S 6 BMI %predFE Physical activity %predFE BMI Age BMI Height Weight S-K Physical activity Age Steroid use BMD assessment method DEXA DEXA DEXA DEXA DEXA Other categori zation 17 with Z- score<-2.0 48 with Z- score<-2.0 22 with Z- score<-2.0 O c / i - ^ C J O r t O V H O i / i r t t / i ON <n r o C N C N C N O c / ] + J t u o o < c j e . r t C d 0 0 T t VO 0 0 0 0 T T rt C N T- femoral BMD -1.0(0.1) -1.3(1.3) -2.1(0.3) -1.3(0.4) -1.0(0.6) -2.7(1.0) -2.4(1.9) -2.1(1.5) T-Lumbar BMD -0.8(0.1) -1.2(1.2) -2.0(0.4) -1.9(0.4) -1.2(0.6) -2.4(1.3) -2.1(1.9) -1.8(1.6) BMI 21.2(0.4) 21.2(2.5) 19.2(1.0) 18.0(0.6) 18.2(3.0) 18.8(1.2) 16.8(2.4) 19.8(2.6) Disease severity: S-K scores 66.9(2.1) 88(8) 65(22) 73(14) 65(18) Disease severity: %predFEVi 57.6(3.0) %predFVC 78.6(2.4) 58.8(24.0) 27(3) 22(2) 87(23) 51(9) 80(2) 55(19) Age (yrs) 28.6(1.0) 25.3(7.1). 17.7 <19 >19 30(2) 31(2) 13.3(2.7) 27.4(11.6) 13.7(2.7) 28.7(6.4) Sex (M/F) 44/24 79/64 62/72 T t VD rt r—{ Males 9 (<18yr) ll(>18yr) Females 15(>18yr) r~ VO r o T f Ti-ro CO <—I rt r~ vo o r o T t Reference Present study a Haworth etal (1999)6 rt ?-CJ ^ £ £ CJ cv GO cv *-< rt 3 C-CO ^ _ Donovan et al (1998) 3 " 3 n *- ?s -3 CJ Cv 1 ^ -oa o w Cumulative steroid dose BMI Age at puberty Height 5 X < S CD Age at puberty7 Height Weight Physical activity co S < m DEXA DEXA DEXA DEXA DEXA DEXA DEXA DEXA Reduce dBMD 26 with score<-2.0 o T T o rs r-rs SO rs ON ro •—< rs' CN i i 1.99(1.0) -1.9(1.4) -0.7(0.2) -2.6(0.3) -2.5 -2.4 -2.2 -2.1(1.2) -1.7(1.4) -1.0(0.1) -2.5(0.3) -2.8 19.0(2.2) 17.4(2.2) 16.5(1.8) 18.6(1.9) 21.0(2.9) 18.9(2.5) 18.1(3.1) 17.0 18.1-19.1 83.8(7) 80.4(10) 72 (14) 70.3(16) 73.7(18) 68.0(15) 76(9) 55.6-56.5 33.2(5.8) 34.3(6.2) 80.2(22.4) 80.0(25.5) 70.3(33.0) Transplant candidates 74.9(27.2)6 78.8(17.8) 72.0(23.9) 19 subjects<60 % 42 subjects>60 % Transplant candidates 32.1(7.9) 24.9(6.7) 9.3(1.8) 14.2(2.5) 23.8(4.1) 18-56 20.6 10.7 (3.6) 12.2(3.5) 30(5) 26-28 CO CO 14/10 10/12 8/5 19/30 34/28 <n ON 14/8 o cs co (S (S ^ o rs ON • * rs SO cs rs Aris et al (1998) Baroncelli etal (1997) Aries et al (1996)K ^ to 3 *<3 (T-p tD OS £ v 2 ra "3 w Henderson & Madsen (1996) ° Salamoni etal (1996) Shane et al (1996) Bachrach etal (1994) BMI NIH clinical scores DEXA QCT SXA 9 below low-normal range NO tt C O . <o tt C O . Mean trabecula r vertebral density 20.0 (2.0) i— i m 69.3 w u r n NO ON r-' o 65 35-100 23(8) 11.6(5.4) 18 14.5-24.5 00 So 29/28 13/7 NO 1—1 t> o rs Gibbens et al (1988) Hanly et al (1985) 8 11 1.3 H Y P O T H E S I S Severity of lung disease (%predFEVi), malnutrition (BMI), low intake of dietary vitamin D and calcium, and limited physical activity (V0 2 m c u ) independently contribute to the development of low B M D in CF patients. Specific A ims: Primary aims: 1. Document the prevalence of low B M D in an adult CF population with heterogeneous pulmonary disease severity (S-K clinical scores, Brasfield scores, %predFEVi). Examine which variable of pulmonary disease severity best predicts BMD. 2. Examine whether CF patients who have poor nutritional status are more likely to have low BMD. Nutritional status will be defined as follows: a) Physical signs of poor nutrition: BMI, fat and lean body mass, and b) dietary and biochemical signs of poor nutrition: caloric intake, calcium and vitamin D intake, serum albumin and bone specific alkaline phosphatase. 3. Examine whether CF patients who engage in regular physical activity (as assessed by Y02max and exercise activity diary) will exhibit higher B M D values than sedentary C F patients. Secondary aims: 4. Examine whether CF patients who have low B M D are more likely to have chronic back pain, a higher degree of vertebral wedging, and higher kyphosis scores. 5. Examine whether CF patients showing delayed onset of puberty (age males and females reach Tanner stage 3 and 2, respectively and age of menarche in females) are more likely to exhibit lower B M D in adulthood. 6. Document the fracture rate and also the prevalence of steroid use during puberty and at the time of study participation (use of steroids for less dian or greater than 2 months per year). 1.4 S IGNIF ICANCE The proposed study will extend current knowledge concerning the causes of osteopenia in CF. The associations that are exhibited will have implications for the clinical therapies that adult and adolescent CF patients will be prescribed and will help define the necessity to respond to lower B M D values with treatment (i.e., prescribing or increasing vitamin D and calcium supplementation, intervening with bisphosphonates and calcium supplementation, adequate caloric intake during adolescent years, reassessing current pancreatic enzyme replacement and increasing exercise). 12 CHAPTER n METHODS AND PROCEDURES 2.1 SAMPLE AND E L I G D 3 H J T Y CRITERIA This was a prospective study, recruiting patients from those attending the Saint Paul's Hospital Adult Cystic Fibrosis clinic. Patients with a diagnosis of C F on the basis of clinical signs and two elevated sweat chloride values according to published standards by Shwachman and Mahmoodian (1967) were recruited for participation in the study. An attempt was made to recruit all C F patients attending the Saint Paul's Hospital Adult C F clinic. The sample population included patients 18 years and older, who showed mild to severe airway obstruction. Patients who agreed to perform all tests in the study were recruited and measurements made during periods of clinical stability without hospitalization or home therapy with intravenous antibiotics for pulmonary infection within 2 months of entry. C F patients who were already being treated for osteoporosis, or had received a lung transplant or were physically unable to complete the exercise test because they were on supplemental oxygen were not asked to participate in the study. 2.2 STUDY PARAMETERS 2.2.1 Bone densitometry Dual-energy x-ray absorptiometry (DEXA) was used to measure lumbar spine (Ll-4), proximal femur (femoral neck and total hip including trochanter, Ward's triangle), and total body B M D . Measurements were expressed as grams of bone mineral per square centimeter of bone and data was finally expressed as z-scores (number of SD's that the B M D measurement is from age and sex matched control subjects). A composite score for lumbar spine and femoral neck B M D (i.e., (Lumbar spine T score + femoral neck T-score)4-2) will be calculated. Reference data used were those provided by D E X A manufacturers. Reference values were not available for total body B M D . Standards accepted by the World Health Organization (Scientific advisory board and Osteoporosis Society of Canada; 1996) were used to define normal bone density, osteopenia and osteoporosis. Normal bone density was defined as a B M D up to 1 SD below the mean for young healthy adults. Osteopenia was diagnosed as having a B M D of 1-2.5 SD's below the mean for young healthy adults. Osteoporosis was defined as a B M D > 2.5 SD's below the mean for young healthy adults. 2.2.2 Body composition, physical growth and reproductive history Dual-energy x-ray absorptiometry was used to measure lean body mass and body fat. Physical maturation was assessed to elucidate if there was a delay in the onset of puberty in childhood, which may have permanentiy retarded childhood growth. A retrospective review of the patients' medical charts at B C Children's Hospital examining Tanner staging for these patients at the time of onset of puberty was 13 documented. Female patients were asked to complete a reproductive questionnaire, which addressed age of menarche and menstrual history. The Reproductive History section from the questionnaire developed for the Canadian Multi-center Osteoporosis Study (1995) was used. See Appendix A for a sample of the questionnaire. 2.2.3 Exercise tests and physical activity questionnaire protocols Subjects performed a maximal oxygen consumption ( V 0 2 w o x ) test during stable clinical status. Subjects were asked to refrain from eating 2-3 hours prior to testing and from heavy exercise 2 days before V®2max testing. The V 0 2 m a x test was performed on a Monark™ stationary bicycle and followed a continuous progressive incremental regimen. The workload was increased each minute by 0.25 kiloponds (kp) and patients were asked to maintain pedal frequency within the range of 60-80 revolutions per minute (rpm) throughout the test. Initial workload was individually set and was dependent on physique, fitness level and severity of disease for each patient and ranged from 0 kp to 1.5 kp and increased thereafter by 0.25 kp/min.. Metabolic measurements (i.e., oxygen consumption, ventilation, tidal volume, respiratory rate, and respiratory exchange ratio (RER)) were obtained at 30-second intervals utilizing a Beckman metabolic cart. Subjects wore a nose clip and breathed through a Rudolph 2-way valve for the duration of the test. Heart-rate (HR) and oxygen saturation (Sp02) were monitored throughout the test with an Ohmeda finger Pulse Oximeter (a Polar HR monitor was also used to monitor HR) and average HR and Sp0 2 values were recorded for each minute of exercise. The 10-point Borg scale was used to report leg and chest scores for perceived exertion and dyspnea, every 2 minutes during the test. The test was terminated by the subject due to fatigue or by the researchers if there was a plateau in VO2 and HR for more than 1 minute, or Sp02 levels fell below 82%. V02max w a s substantiated with subjects meeting at least two of the following 3 criteria (except in cases where the test was terminated due to Sp02<82%): a plateau for more than 1 minute of VO2 (±2 mlkg'^min"^) and HR (+3 bpm) with increasing workload and a respiratory exchange ratio (RER)>1.10. Resting measures were obtained with the subjects seated. Subjects who were going to take Ventolin or some other form of bronchodilator were instructed to take their medication following resting data collection. The exercise activity questionnaire required the subjects to maintain a diary of their exercise activities for 7 days for a minimum 2 weeks or up to a 4-week period if they did not have an established exercise routine. In the diary they recorded the mode of exercise, the duration of exercise, and the intensity of exercise. The intensity of exercise was measured by the subjects' measuring their heart rate during the activity. Subjects were instructed to take HR measurements during their activity by palpating their left carotid artery and recording a 10-second reading. Subjects were instructed to take one heart-rate measure for every 15-20 minutes of exercise and average over the total activity. 14 Data collected from the exercise activity questionnaire was quantified for exercise intensity, duration, frequency and mode of exercise. Mode of exercise was quantified as presented in table 3. Categorization was based on whether the activity was land or water based (i.e., land-based load bearing (LB) or aquatic activities (AA)). Activities were further classified by intensity level into hard (high impact (LB-H)) and light (low-impact (LB-L) and on whether die activity was continuous (i.e., L B - H and LB-L) or intermittent (land-based load bearing sports and games (LB-SG)). Weight training was not included in the calculation of exercise intensity, duration and frequency, but frequency of weight training was recorded separately. The reasoning was that we wanted to be able to quantify aerobic activities and then calculate the relative exercise intensity for each subject's aerobic exercise based on V 0 2 m a x and achieved heart rate. Calculation of exercise duration was based on each individual aerobic exercise session(s) and the average duration of an exercise session calculated for each week. The average duration of an exercise session over the 2 to 4 week diaries was then calculated. In a similar manner we calculated the average exercise intensity and average frequency of exercise per week. Table 3. Categorization of mode of exercise and summary of activities reported by study sample. Land-lased load bearing-moderate to hard (LB-H) Land-lased load bearing-light (LB-L) Aquatic activities (AA) Land-lased load bearing-Sports/ Games (LB-SG) 1. Running Walking Swimming Basketball/volleyball 2. Cycling' Hiking Other water based activities (i.e., diving, scuba diving, water exercise) Softball/Baseball 3. Rowing/kayaking Golf Hockey (ice/ inline skate, street) 4. Aerobic class Horseback riding Racket sports (badminton, tennis, racketball etc.) 5. Social / Ballroom dancing 2.2.4 Lung function measurement The Respiratory Therapy Services at Saint Paul's Hospital performed spirometry tests on study subjects during the patient's regular clinic visit. Measurements used in the study were obtained during stable 15 clinical periods and commonly exercise and B M D testing were scheduled for the same day. These measurements included forced expiratory volume in 1 second (FEVi) and forced vital capacity (FVC) carried out according to ATS criteria (40). The values for spirometry were expressed as the percentage of normal values based on age, gender and height. The best value of a minimum of 3 post-bronchodilator measurements was taken with the FEVi and the F V C of the best 2 of these efforts not differing by more than 5% or 100 ml whichever was greater. For the purposes of this study, only post-bronchodilator measurements were selected. All spirometers fulfilled minimal accepted standards (40). Spirometric values were selected close to the time of exercise testing when patients were stable based on review of the patients' charts. 2.2.5 Brasfield and S-K scores derivation Schwachman-Kulczycki (S-K) and Brasfield scores were calculated. S-K scores were calculated for clinical status at the time of testing, which represented stable (non-pulmonary exacerbation) clinical status. Radiographic data were obtained from the patients' medical charts, as was information on activity patterns, pulmonary and nutritional status for S-K scoring. From a maximum score of 100 (maximum 25 per category), points were deducted for the level or degree of: 1. Inactivity, fatigue and non-participation in daily living. 2. Pulmonary symptoms and clubbing. 3. Growth and nutritional deficiencies. 4. Chest radiographic abnormalities A pulmonary exacerbation was defined as a pulmonary infection that required intravenous antibiotic intervention. S-K scoring was performed by myself and the results were reviewed with Dr. Wilcox as described by Schwachman and Kulczycki (41). Chest radiographs were scored utilizing the Brasfield clinical scoring system (42) by Dr. Wilcox, who was blinded to the identity of the patient. The severity of lung disease was determined from a maximum score of 25, with points deducted for: 1. Degree of hyperinflation. 2. Peribronchial thickening. 3. Nodular cystic structures (bronchiectasis). 4. Areas of atelectasis or pneumonia. 5. Assessment of overall severity. 2.2.6 Biochemical analysis Patients participating in the study completed a 3-day dietary recall. Patients were asked to record all foods, beverages and supplements taken for 2 weekdays and one weekend day. This dietary history was then analyzed using the Saint Paul's Hospital's Healthy Heart program dietary computer software package to determine caloric intake, levels of intake of vitamin D and calcium, carbohydrate, fat, protein, 16 phosphorus and sodium intake, alcohol consumption. Mr. Donald Barker analyzed the 3-day dietary recall. Supplements taken by patients were incorporated in the totals of the specific nutrients. Daily energy requirement (DER) and daily energy expenditure for CF patients were estimated as described in Ramsey and associates (43). The equations incorporate patient activity level, level of pulmonary disease (i.e. "/opredFEVO and pancreatic function (i.e., pancreatic sufficient or insufficient). Additionally, for pancreatic insufficient patients an approximate value of 0.85 was used as the coefficient of fat absorption since we did not perform direct measurements of degree of steatorrhea. Kodak AutoAnalyzer was used to measure serum levels of alkaline phosphatase and albumin. Tests were performed by the Saint Paul's Hospital Medical Laboratory. The Saint Paul's Hospital Medical Laboratory also collected serum, which was used for measurement of bone specific alkaline phosphatase (BS-APase). The Metra Biosystems Inc assay kit was used for measurement of BS-APase. Procedures were performed are described in Liberman and co-workers (44). 2.2.7 Radiology, fractures, posture and back pain determination Patients were asked to complete a questionnaire on back pain and history of fractures (see Appendix A for sample of the questionnaire). Thoracic and lumbar spine radiographs were examined for evidence of thoracic vertebral wedging. Anterior, mid and posterior vertebral heights were measured from the anteroposterior chest radiographs and the difference expressed as a percentage of posterior vertebral height. Thoracic vertebras 8-12 (T8-12) were measured and the average percentage decrease in height of T8-12 entered as the subject's score. A 20% decrease in any of the heights of a vertebra was considered indicative of vertebral wedging. Kyphosis was measured with a surveyor's flexicurve. A description of the instrument and validity studies of the instrument is provided in Milne and Lauder (45). Measurement of kyphosis was performed as described in Chow and Harrison (46). 2.3 S T U D Y DESIGN A N D S T A T I S T I C A L A N A L Y S I S A prospective non-experimental design was used. Due to the limited number of C F patients who can conveniently access the St Paul's hospital facilities, we were limited by the number of patients available for study recruitment. A total of 50-70 patients were recruited for various aspects of the study from the St. Paul's Hospital Adult Cystic Fibrosis clinic from potentially 140 patients who attend the clinic. Data analysis was undertaken using SPSS version 7.0 (SPSS Inc., Chicago, Illinois, USA). The relationships between covariates are described using Pearson product moment correlation. Multiple regression analysis was used to identify factors diat influence B M D . A composite score for lumbar spine and femoral neck B M D (i.e., (Lumbar spine T score + femoral neck T-score)+2) was used as the dependent 17 variable in regression analyses. Backward hierarchical linear regression was used to examine each of our hypotheses and primary aims reporting the initial regression equation with all variables stated in the main hypothesis and primary aims included, respectively. The independent variables, which made a statistically significant contribution to the main hypothesis, were used in the final regression model. Similarly, the independent variable that made the most significant contribution in predicting B M D from each of the 3 specific aims was used to develop a secondary regression equation and the independent variables that made a significant contribution to this model were then used for the tertiary regression equation. Age and gender were included in all regression analyses as covariates. Univariate analysis of variance was used to examine differences between bone density groupings and post-hoc analyses using Bonferonni correction for significant findings. The number of subjects available for analysis is provided in Table 4. Column 2 shows total sample available for each predictor and column 3 shows total number of patients available for each predictor(s) who also had complete B M D results. Table 4. Total sample available for analysis. Total N BMD-hip/spine/body V 0 2 m a x test V 0 2 m a x test and exercise diary Dietary recall Blood Predictors 71 55 51 46 67 55 51 46 18 C H A P T E R I I I R E S U L T S 3.1 D E S C R I P T I V E C H A R A C T E R I S T I C S O F S T U D Y S A M P L E A N D P R I M A R Y H Y P O T H E S I S Table 5 shows the descriptive characteristics of the study population. The Saint Paul's Hospital Adult CF clinic serves approximately 140 patients mainly located in the Vancouver and surrounding suburbs, with patients attending as well from other parts of the province of British Columbia. It was estimated that we would be able to detect a correlation of 0.35 with 80% power and a significance level of 0.05 from a recruited sample of 50-70 patients. None of the patients recruited for the study had had a lung transplant, but one patient had received a liver transplant. Comparison of demographic characteristics of participants (shown in table 5) versus non-participants for age (29.8(1.3) yrs, p=0.89), BMI (21.4(0.40) kg/m 2, p=0.46), and gender (M/F=34/30, p=0.27) yielded no significant differences between the study group and non-participants attending the clinic. For severity of pulmonary disease there was a statistically significant difference between the study group and the non-participants (%predFEVi: 63.9(3.7)%, p=0.001), however the difference is not clinically significant. Patients who had received lung or lung and heart transplants were not included in the above analyses. It is however possible that our selection criteria did contribute to bias our study sample by over selecting healthier CF patients. Variables reported in this paper were checked for normality and are normally distributed. Variables used to investigate our primary hypothesis were %predFEV,, BMI, V 0 2 m a x , and for dietary predictors we used vitamin D and calcium intake. Hierchical linear regression was used to predict the composite score of lumbar and femoral B M D . Results for the main hypothesis and primary aim questions are presented in Table 6. The main hypothesis model accounted for only 33% of the variance for B M D (model 1). With stepwise regression, the final model generated included BMI and V02max, together with age and gender accounted for 35% of the variance for B M D (model 7). Using the independent variables from primary aims 1-3 with the highest regression coefficients for each model we generated an alternate model (model 6). S-K clinical scores, total dietary vitamin D intake and lean muscle mass, together with age and gender accounted for 25% of the variance for BMD. Stepwise regression applied to model 6 generated the final model which included variables V02max and lean muscle mass, together with age and gender accounted for 23% of the variance for B M D (model 8). 3.2 P R E V A L E N C E O F L O W B M D Table 7 shows the prevalence of low B M D in die study population, the overall pulmonary disease severity (as described by S-K and Brasfield scores and lung function), age at time of testing for the study and at the time of CF diagnosis, B M D values as well as die distribution of our sample by gender. Out of the 67 19 patients tested, 17 had B M D z-scores of -2 or less. In univariate analysis we showed significant positive associations between parameters of pulmonary disease severity and B M D (i.e., %predFEV,: r=0.35, p=0.004; S-K clinical score: r=0.40, p=0.001 and Brasfield score: r=0.38, p=0.001), however these correlations were weak even though they were statistically significant. Parameters of pulmonary disease severity accounted for 21% of the variance for B M D and S-K clinical score was the best predictor of the three for B M D (Table 6, model 2). There were significant differences in pulmonary disease severity scores in patients grouped by bone mineral density (Table 7). Although no significant differences were seen between the normal and osteopenic groups, diese 2 groups had significantly less severe pulmonary disease than the osteoporotic group. Interestingly, patients classified as osteoporotic were not only significantly older at the time of testing, but were also diagnosed with CF later in life. The latter factor may have implications for adult B M D and attainment of peak bone mass. The misfortune of late diagnosis of CF in this group may have jeopardized pulmonary and nutritional status during childhood and had lasting implications to adulthood pulmonary and B M D reserve. 3.3 NUTRITIONAL STATUS AND B M D Eighty seven percent of study patients were pancreatic insufficient (see Figure 1). The distribution of BMI, total fat and lean muscle mass, serum albumin and BS-APase by bone mineral density status are presented in Table 8. Univariate analysis of variance for each independent variable by B M D is shown in Table 8. There were no significant differences in caloric intake in patients grouped by B M D and similarly no differences between groups for estimated dietary intake (DER) and expenditure (DEE). A comparison between actual dietary intake and DER, actual dietary intake and D E E and the difference between actual dietary intake and estimated energy expenditure (kcal-DEE), also did not show any significant differences across B M D groups (Figure 2). Figures 3 and 4 show the dietary intake for calcium and vitamin D (from diet and supplements). Thirty and 50 percent of our patient study population were not on oral vitamin D and calcium supplementation, respectively. There were significant differences between B M D groups for total vitamin D and calcium intake. This was mostly related to supplemental intake of vitamin D and calcium and not to differences in food intake. This observation is paradoxical but is probably related to practice in the clinic to have patients with more severe pulmonary disease take oral vitamin D and calcium supplements. Since patients with more severe pulmonary disease had lower B M D this resulted in a negative correlation between dietary and supplemental vitamin D and calcium with B M D (Table 9). Specifically, total vitamin D dietary intake was significantly higher in the osteoporotic group compared to the osteopenic and normal B M D groups. We had one patient in our sample that had received a liver transplant and this patient's values for dietary intake parameters are presented separately in Appendix B. There were 7 patients who were also prescribed Ensure Plus™ as a food supplement to increase caloric 20 intake and although.they have been included in the analysis their results are also presented in Table 10 broken down to dietary intake from food and supplements. Nutritional aim 2 was addressed as 2 separate models: the physical and the dietary characteristics. The physical characteristics of the nutrition model accounted for 29% of the variance for B M D with lean muscle mass out of the three physical nutritional parameters accounting for most of the variance shown (Table 6, model 3). The dietary characteristics of the nutrition model accounted for only a very low percentage (i.e., 10.4%) of the variance for B M D and the values were all negative (Table 6, model 4). We were limited in terms of the number of individuals from whom we had results to examine our hypothesized prediction equations. The intercorrelations of the dependent (i.e., composite BMD) and independent variables are presented in Table 9. 3.4 E X E R C I S E A N D B M D The activity level model accounted for only 14% of the variance for B M D (Table 6, model 5). V 0 2 m a x was the important exercise parameter in the model with the categorical variable, exercise status (i.e., whether patients exercised or not) not contributing much to the equation. Of the subjects who reported their exercise status there were 6 out of 10 classified as having normal B M D , 25 out of 47 classified as osteopenic and 4 out of 9 classified as osteoporotic who regularly engaged in exercise. Table 11 shows the breakdown of the subjects' perceived activity level and B M D classification. Table 12 shows measured and predicted V 0 2 m a x results, Table 13 exercise diary results and Table 14 shows the breakdown of subjects based on exercise status (at puberty and adulthood) and B M D classification. Although statistical analysis (chi square) yielded no significant pattern (p=0.06), the osteoporotic group did show a trend of perceiving that they had lower overall activity levels and this was substantiated by V 0 2 m a x results. While patients were asked to rate their activity level in this study, we found no association between the patients' subjective rating of current activity level and BMD. Only the patients' measured V 0 2 m a x and B M D yielded a significant positive association (r=0.42, p=0.001). On a subset of our sample (N=30) where we were able to retrospectively review the patients' childhood medical charts, we were able to ascertain their activity patterns at the time of puberty (Table 14). However, we did not show any significant associations between childhood exercise or inactivity and adult B M D (r=0.21, p=0.26) and there was no association between activity patterns in childhood and exercise status in adultiiood (r=-0.21, p=0.29). 21 3.5 ASSOCIATION OF B A C K PAIN, KYPHOSIS AND V E R T E B R A L WEDGING WITH BMD 3.5.1 Kyphosis and cumulative vertebral fractures Table 15 shows descriptive results of the cumulative vertebral fractures of T8-T12 and kyphosis by bone density grouping, respectively. We found no significant differences between the B M D groups for vertebral fractures or kyphosis. Aris and colleagues have suggested an association between lumbar spine B M D and degree of kyphosis (47). We decided to look at this relationship in our study sample as well. For this reason we performed stepwise regression analysis to look at whether low lumbar spine B M D could be predicted from cumulative vertebral fractures of T8-T12 and kyphosis scores, however neither of the independent variables were predictive (Table 16). It has been suggested that age and severity of pulmonary disease are associated with more pronounced kyphosis (48). As an aside we looked at the ability of these two parameters to predict kyphosis and cumulative vertebral fractures of T8-T12. For both kyphosis and cumulative vertebral fracture of T8-T12, %predFEV, was the best predictor, however this association could only explain 12.2% and 8.1%, respectively, of the variability for predictor variables kyphosis and cumulative vertebral fracture of T8-T12 (Table 16). Lastly we performed a regression analysis including age and pulmonary function as independent variables together with kyphosis and T8-12 and looked at lumbar spine B M D predictability. Following stepwise regression analysis %predFEVi was the only parameter remaining, however %predFEVi could only explain 10% of the variability in lumbar spine BMD. The intercorrelations for the above analyses are presented in Table 17. 3.5.2 Back Pain Back pain is common among CF patients and we found this complaint to be present regardless of BMD status (see Table 18). Back pain is localized in the thoracic and lumbar region (Table 19). The severity and frequency of back pain are presented in Tables 20 and 21. Although these data are only descriptive, they suggest that CF patients who have back pain tend to experience episodes of pain on a regular basis. Activities that the patients identified which their back pain interfered with are presented in Table 22. 3.6 PUBERTY AND BMD We were unable to perform regression analysis or univariate analysis of variance due to the small number of patients in the normal and osteoporotic groups for whom data were available. The results are presented descriptively. We attempted to retrospectively obtain data from the patients' adolescent medical charts. Due to the policy of the hospital to destroy medical files 10 years following transfer of the patient to the adult clinic we were only able to obtain data on 30 patients regarding Tanner staging for adolescent growth spurt. This caused a definite bias in our data collection, since we were unable to obtain data on the majority of our osteroporotic group who were older. Age of menarche was obtained through 22 questionnaire for 22 of the female patients participating in the study (Table 23). Table 23 also presents study sample data for Tanner staging for sexual maturation (i.e., scored as stage 2 for females and stage 3 for males for pubic hair). The average age range for the adolescent growth spurt for male and female participants and average age of menarche are also presented in Table 23 as described in Tanner (7). Based on questionnaire responses we identified 12 out of the 22 females (Table 23) to have experienced amenorrhea or oligomenorrhea. However, there were no significant differences in B M D status in those females with normal versus abnormal menses (Table 23 and 24). Data for age of menarche seem to suggest that onset of menstruation is delayed in CF females and that this is more pronounced in those females who as adults have low BMD. A similar trend is seen for onset of puberty in males. One-way A N O V A did show that overall CF patients who have low B M D do reach puberty at an older age (Table 23). Our analysis did show that our B M D groups did differ in age of onset of puberty and that patients classified as osteopenic were significantly delayed in their adolescent growth spurt compared to patients with normal BMD. It is likely that adolescent growth spurt would be further delayed in osteoporotic patients, however we are limited by our small sample size for this group. 3.7 P R E V A L E N C E O F N O N V E R T E B R A L F R A C T U R E S A N D B M D Table 25 presents the prevalence of nonvertebral fractures in the 3 groups. We did not observe an increased fracture risk in our osteoporotic group; however, this may be related to our small sample size or to the sedentary nature of this group. We did observe a greater incidence of refracture in the osteopenic group. Fracture and refracture incidences were more common in those subjects who lead a more active lifestyle (regular participation in sports). Two out of the 3 normals and 11 out of the 12 osteopenic subjects who experienced fractures classified themselves as active (moderately active to heavy labor). These individuals also engaged in regular sports, which likely increased their risk of fracture. 3.8 STEROID U S E A N D B M D We examined corticosteroid use both during puberty and at the time of measurement of BMD. Patients were classified as taking corticosteroids if they were using them for greater or equal to 2 months per year and were classified as not on corticosteroids if they were taken for less than 2 months per year or were not taken at all. The prevalence of corticosteroid use at puberty and adulthood are presented in Table 26. There were no significant differences in uses of oral and inhaled corticosteroids during adulthood by the bone density groups (x2=1.6, p=0.4). We showed no significant association between corticosteroid use and low B M D , however the small sample sizes in the normal and osteoporotic groups limited the power to detect differences in our results. The majority of patients classified as taking corticosteroids were taking 23 inhaled preparations (Beclovent™, Beclofort™, Flovent™, Pulmacort™). One patient in the osteopenic group had received a liver transplant and was on antirejection medications. Table 5. Descriptive characteristics of the study sample. Variable Mean Range (min-max) S E M Age (yrs) 28.6 18-55 1.0 Gender (male/female) 44/24 BMI (kg/m2) 21.1 16.0-31.7 0.4 S-K score 66.9 25-100 2.1 Brasfield score 15.0 5-23 0.5 %predFEV, (%) 57.6 21.0-119.0 3.0 %predFVC (%) 78.6 33.0-117.0 2.4 T-lumbar spine -0.8 -3.1 - 1.8 0.1 T-femoral neck -1.0 -3.9-2.0 0.1 Lumbar B M D (g/cm2) 0.67 0.67-1.45 0.02 Femoral B M D (g/cm2) 0.90 0.55-1.26 0.02 Total B M D (g/cm) 1.0 0.7-1.3 0.01 Total B M C (g) 2622.8 1650.0-3504.0 50.1 24 Table 6. Regression models for composite lumbar spine and femoral B M D . Model # Model Covariates Coefficient p-value 1 Main Hypothesis Intercept -3.85 0.003 V 0 2 m a x 0.05 0.03 %predFEV, -0.005 0.51 Age -0.001 0.93 Sex -0.18 0.60 B M I 0.12 0.003 Total dietary Vitamin D intake -0.0004 0.21 Total dietary calcium intake -0.0002 0.36 N=49; Adjusted R2=0.33; F-value=4.4, p=0.001 2 Primary aim 1: Pulmonary disease severity Intercept -1.83 0.008 Age -0.02 0.29 Sex -0.57 0.02 %predFEV, 0.004 0.56 Brasfield scores 0.01 0.79 S-K scores 0.02 0.15 N=66; Adjusted R 2=0 .21; F-value=4.6t p~0.001 3 Primary aim 2: Nutritional-physical characteristics Intercept -3.87 0.003 Age -0.025 0.06 Sex 0.20 0.62 B M I 0.044 0.51 Lean muscle mass 0.00005 0.05 Fat mass 0.00003 0.26 N=65; Adjusted R 2 -0 .29 ; F-valuc=6.2, p=0.001 25 ... Table 6 continued 4 Primary aim 2: Nutritional-dietary characteristics Intercept -0.11 0.96 Age -0.03 0.74 Sex -0.54 0.13 Total dietary Vitamin D intake -0.0006 0.14 Total dietary calcium intake -0.00009 0.78 BS-APase -0.006 0.55 Serum albumin 0.02 0.66 Daily caloric intake -0.00002 0.85 N==45; Adjusted R2=0.10: F-valuc= 1.8. p=0.13 5 Primary aim 3: Activity level Intercept -1.51 0.04 Age -0.01 0.36 Sex -0.24 0.36 V 0 2 m ! l x 0.04 0.16 Exercise status 0.01 0.97 N=f»5; Adjusted R2-0.14: F-vnhic=3.8. p=0.009 26 ... .Table 6 continued 6 Predicting BMD from the statistically significant contributor variables for models generated for primary aims 1-3. Intercept -3.03 0.05 Generated model based on models 2-5. Age -0.02 0.32 Sex 0.05 0.93 S-K scores 0.01 0.21 Total dietary Vitamin D intake -0.0005 0.13 Lean muscle mass 0.00004 0.19 V 0 2 m a x 0.008 0.69 N=49; Adjusted R2-0.25; F-valuc=3.7, p=0.005 7 Predicting BMD from statistically significant variables generated in model 1 (stepwise regression). Intercept -4.59 0.001 Age -0.01 0.34 Sex -0.13 0.58 V0 2 m i i X 0.04 0.003 BMI 0.14 0.001 N=65; Adjusted R2=0.35; F-value=9.7, p=0.001 Intercept -3.99 0.002 8 Predicting BMD from statistically significant variables generated in model 6 (stepwise regression). Age -0.02 0.28 Sex 0.52 0.18 V 0 2 m a x 0.02 0.12 Lean muscle mass 0.00006 0.01 N=65; Adjusted RM).23; F-valuc=5.7, p=0.001 27 Table 7. B M D and disease severity between B M D classifications. Variable G, : Normal G 2 : Osteopenic G 3 : Osteoporotic p-value Significant (Z score <-1.0) (Z score:-1.0 to 2.5) (Z score > 2.5) post-hoc differences Sample size 11 48 9 Gender 6/5 32/16 6/3 (male/female) Age at time of 27.0 (1.6) 30.6 (1.1) 37.5 (4.0) 0.01 G, < G 3 testing (yrs) 21.7-39.5 20.3-55.8 20.5-56.6 Age at time of 1.4 (0.9) 2.8 (0.8) 11.5 (4.6) 0.003 G, < G 3 CF diagnosis 0-7.0 0-31.2 0-35.0 G 2 < G 3 (yrs) %predFEV, 69.5 (7.4) 59.0 (3.5) 35.4 (4.8) 0.006 G , > G 3 27.0-98.0 21.0-119.0 23.0-62.0 G 2> G 3 S-K scores 74.1 (5.3) 68.6 (2.1) 48.9 (6.7) 0.001 G,> G 3 45.0-100.0 40.0-100.0 25.0-80.0 G 2> G 3 Brasfield 18.0 (1.3) 15.2 (0.6) 10.9 (1.6) 0.001 G , > G 3 scores 12.0-23.0 7.0-23.0 5.0-17.0 G 2> G 3 BMD-lumbar 1.1 (0.05) 1.0 (0.02) 0.9 (0.06) 0.001 G,> G 3 (g/cm) 0.9-1.4 0.7-1.4 0.7-1.2 G 2> G 3 BMD-femoral 1.1 (0.04) 0.9 (0.01) 0.6 (0.02) 0.001 G,> G 2 (g/cm) 0.9-1.3 0.7-1.1 0.5-0.7 G , > G 3 G 2> G 3 BMD-total 1.1 (0.03) 1.0 (0.02) 0.8 (0.02) 0.001 G,> G 3 (g/cm) 0.9-1.2 0.8-1.3 0.7-0.9 G 2> G 3 BMC-total (g)* 2875.5 (105.1) 2650.2 (56.7) 2195.8 (86.5) 0.001 G , > G 3 2307.0-3340 1844.0-3504.0 1650.0-2491.0 G 2> G 3 * - Total bone mineral content. 28 Table 8. A N O V A results for nutritional predictor variables by B M D grouping. Variable G, : Normal G 2 : Osteopenic G 3 : Osteoporotic p- Significant (Z score <-1.0) (Z score:-1.0 to 2.5) (Z score > 2.5) value post-hoc differences Height (m) 1.7 (0.02) 1.7 (0.03) 1.7 (0.02) 0.61 1.6-1.8 1.5-1.9 1.6-1.8 Weight (kg) 68.0 (3.9) 61.6 (1.4) 54.6 (2.5) 0.02 G, > G 3 51.8-87.3 41.5-98.0 41.0-67.9 BMI 23.4 (1.4) 21.1 (0.4) 19.2 (0.7) 0.01 G, > G 3 18.3-31.7 16.6-30.6 16.0-22.0 % Body fat 22.7 (3.8)* 17.0 (1.1) 17.5 (2.7) 0.16 5.2-42.2 7.0-31.6 7.0-32.1 Fat mass 21307.7 13946.4 (883.9) 12399.7 (2164.4) 0.01 G, > G 2 (4055.9)* 6752.0-32059.0 5406.0-22095.0 G , > G 3 6763.0-46608.0 Lean mass 44826.1 44155.6 (1213.0) 40367.2 (2387.4) 0.41 (2619.6)* 25658.0-61699.0 28210.0-47791.0 28150.0-56554.0 Dietary calcium 1063.6 (149.7) 1482.8 (119.5) 1531.4 (234.7) 0.21 (mg) 485.4-1951.0 19.5-3325.0 667.5-2387.0 Supplemental 111.1 (73.5) 175.9 (66.7) 664.3 (2008.9) 0.01 G, < G 3 calcium (mg) 0-500.0 0-1500.0 0-1500.0 G 2 < G 3 Total calcium 1174.7 (156.7) 1657.8 (140.8) 2195.6 (303.7) 0.04 G, < G 3 intake (mg) 485.4-1951.0 19.5-3571.0 1167.5-3314.0 Dietary vitamin 243.4 (62.4) 318.4 (36.9) 424.9 (99.6) 0.26 D(IU) 45.1-646.0 17.3-825.6 103.3-812.0 Supplemental 311.1 (111.1) 321.5 (57.7) 700.0 (125.4) 0.03 G 2 < G 3 vitamin D (IU) 0-800.0 0-800.0 400.0-1300.0 Total vitamin D 554.5 (126.9) 639.9 (72.9) 1124.9 (156.4) 0.02 G, < G 3 intake (IU) 45.1-1103 85.1-1625.6 503.3-1661.0 G 2 < G 3 Kcal (calories) 2516.4 (307.6) 3122.3 (313.5) 3003.0 (300.1) 0.61 1334.0-4024.0 1669.0-12665.0 2175.0-4403.0 29 DER (calories)** 2942.1 (136.6) 2940.3 (67.9) 2841.3 (143.3) 0.77 2338.9-3777.9 2023.1-3732.3 2163.8-3383.2 D E E 2733.5 (115.7) 2716.5 (60.5) 2618.0 (120.6) 0.83 (calories)*** 2137.7-3452.9 1889.5-3411.3 1977.7-3092.2 Kcal-DEE -252.3 (309.2) 385.6 (300.3) 373.3 (235.0) 0.63 (calories) -1425.9-1478.2 -943.1-9557.9 -525.5-1310.8 BS-APase (u/L) 19.9 (1.5) 23.0 (2.6) 25.5 (9.5) 0.78 13.2-24.6 4.0-86.6 10.0-78.4 Albumin (g/L) 40.4 (0.7) 39.9 (0.5) 37.9 (1.5) 0.24 37.0-45.0 28.0-45.0 29.0-42.0 Total sample 11 48 9 available for analysis * -Data available for only 10 patients. **- DER: estimated dietary intake. *** -DEE estimated dietary expenditure. 30 Table 9. Intercorrelations for composite lumbar spine and femoral B M D with nutritional predictor variables. T C - B M D BMI BS-APase Fat mass Lean mass Calcium Vitamin D Kcal Albumin T C - B M D 1.00 0.42+ 0.02 0.29* 0.42+ -0.25 -0.34* -0.03 0.25* B M I 1.00 -0.18 0.74+ 0.37+ -0.13 -0.21 -0.05 0.14 BS-APase 1.00 -0.21 0.16 0.006 0.04 0.06 0.03 Fat mass 1.00 -0.10 -0.19 -0.21 -0.26 0.16 Lean mass 1.00 0.12 -0.17 0.30* 0.06 Calcium 1.00 0.55+ 0.52+ -0.35* Vitamin D 1.00 0.19 -0.19 Kcal 1.00 -0.20 Albumin 1.00 . * -Correlation is significant at the 0.05 level (2 tailed). + -Correlation is significant at the 0.01 level (2 tailed). 31 Table 10. Dietary intake from foods and supplements for the 7 patients on dietary liquid replacement food. ID 5 13 33 37 53 82 84 Sex Male Female Male Male Female Male Male Bone status+ G, G 2 G 2 G 3 G 2 G 2 G 2 %predFEV, 30 44 30 27 37 45 23 (%) Dietary 891 972 (53.0) 246 1001 1447 1146 1132 Vitamin D (IU) (51.5)** (92.5) (53.8) (63.0) (58.2) (70.2) Dietary calcium 869 664 (55.5) 1207 2394 1985 2089 2738 (mg) (52.4) (89.3) (55.3) (84:6) (93.6) (63.5) Supplemental 800 800 (43.7) 0 800 800 800 400 vitamin D (IU) (46.2) (43.0) (34.8) (40.7) (24.8) Supplemental 500 100 (8.4) 0 1500 0 0 1000 calcium (mg) (30.2) (34.7) (23.2) Ensure vitamin 40 (2.3) 60 (3.3) 20 (7.5) 60 (3.2) 50 (2.2) 20 (1.0) 80 (5.0) D (IU) Ensure calcium 288 432 144 432 360 144 (6.4) 576 (mg) (17.4) (10.7) (10.0) (15.4) (13.3) Dietary kcal* 2034 1204 2561 3693 2568 3206 2186 (81.1) (91.5) (83.9) (81.3) (93.1) (69.8) Ensure kcal* 473 710 237 710 592 237 (6.9) 947 (18.9) (8.5) (16.1) (18.7) (30.2) Total kcal* 2507 1914 2798 4403 3160 3443 3133 DER* 3079 2264 3446 3383 2243 3515 3312 D E E * 2814 2069 3150 3092 2103 3213 3027 + - G, =Normal; G 2 = Osteopenic; G 3= Osteoporotic. * -Unit of measurement are calories. ** - Values in parentheses are percent of total vitamin D, calcium and caloric intake. 32 Table 11. Subjective rating of overall activity level. Activity level rating Normal Osteopenic Osteoporotic Sedentary 1 6 1 Light activity 3 10 6 Moderate activity 3 19 0 Very active 3 10 0 Heavy manual labor 0 2 0 Total sample available 10 47 7 for analysis Table 12. V 0 2 m a x results by B M D grouping. Variable Gi: Normal G 2 : Osteopenic G 3 : Osteoporotic P-value Significant post-hoc differences Sample size 10 47 9 V 0 2 m a x (ml kg"1 min"1) 32.4 (3.0)* 21.5-48.1 30.5 (1.2) 10.9-48.1 22.6 (3.3) 13.4-44.6 0.03 G,> G 3 G 2> G 3 V0 2 m a x ( lmin" 1 ) 2.2 (0.2) 1.1-3.3 1.9 (0.09) 0.8-3.1 ' 1.2 (0.2) 0.7-2.7 0.004 G , > G 3 G 2> G 3 Predicted V 0 2 m a x ( lmin 1) 2.8 (0.2) 2.0-3.4 2.6 (0.09) 1.2-3.8 2.2 (0.2) 1.0-3.1 . 0.2 % of predicted V 0 2 m a x (%) 78.5 (5.8) 55.5-106.1 73.7 (2.7) 29.9-113.8 57.3 (7.8) 30.5-95.6 0.04 * - Values are mean (SEM) followed by range. 33 Table 13. Exercise diary results by BMD grouping. Variable Normal Osteopenic Osteoporotic p-value Sample size 3 20 3 Exercise HR (beats/min) 107.4 (22.2) 82.8-151.8 135.5 (5.4) 93.7-180.0 137.1 (3.2) 130.8-141.0 0.20 Exercise HR as % of H R ^ (%) 62.3 (14.5) 43.6-90.9 85.2 (3.2) 59.7-110.0 88.5 (6.7) 80.7-101.9 0.06 Exercise frequency (# of times per week) 3.5 (0.4) 3.0-4.2 3.8 (0.4) 1.0-7.0 2.8 (0.4) 2.0-3.2 0.53 Exercise duration per session (min) 48.2 (10.7) 26.8-60.0 61.5 (8.8) 13.3-151.2 48.0 (11.4) 25.6-62.8 0.31 Table 14. Exercise habits at puberty and adulthood. Normal Osteopenic Osteoporotic Durmu pubcrls Regular exercisers 4 12 0 Sedentary 4 8 2 Total sample available 8 20 2 During adulthood Regular exercisers 4 26 4 Sedentary 6 22 5 Total sample available 10 48 9 Table 15. Kyphosis and T8-T12 and BMD. Variable Normal Osteopenic Osteoporotic p-value Sample size 11 48 9 Kyphosis (cm) 12.2 (0.9) 8.0-19.4 10.9 (0.4) 4.6- 16.8 13.1 (0.8) 11.3 - 16.1 0.14 T8-12 0.95 (0.006) 0.93-0.98 0.95 (0.005) 0.85- 1.00 0.95 (0.004) 0.94-0.97 0.75 Total sample available for analysis was N=68. 34 Table 16. Regression models for lumbar spine BMD with kyphosis and T8-T12. Model Model Covariates Coefficient p-value 1 Prediction of spine BMD from kyphosis and T8-T12 Intercept -0.91 0.84 Forced variable entry* T8-T12 -0.10 0.98 Kyphosis 0.02 0.68 N=59: Adjusted R^-u.03: F-value=0.09, p~0.91 2a Prediction of kyphosis from age and %predFEV, Intercept 13.07 0.001 Forced variable entry %predFEV, -0.04 0.009 Age 0.02 0.67 N=o9; Adjusted R2-0.10: F-valuc=4.3, p=0.02 2b Prediction of kyphosis from age and %predFEV, Intercept 13.76 0.001 Stepwise analysis results %predFEV, -0.04 0.005 N=59; Adjusted R2=0.11; F-valuc=8.5, p=0.005 3a Prediction of T8-T12 from age and %predFEV, Intercept 0.94 0.001 Forced variable entry %predFEV, 0.0003 0.03 Age -0.0004 0.41 N=59; Adjusted R2=0.08; P-value=3.7, p=0.()3 3b Prediction of T8-T12 from age and %predFEV, Intercept 0.93 0.001 Stepwise analysis results %predFEV, 0.0003 0.01 N=59; Adjusted R2=0.08; F-value=6.7, p=().()l 4a Prediction of spine BMD from kyphosis, T8-T12, age and %predFEVi Intercept 1.06 0.81 Forced variable entry %predFEV, 0.02 0.001 Kyphosis 0.07 0.13 - T8-12 -4.21 0.35 Age 0.01 0.54 N=59; Adjusted R2=0.12: F-value=2.9, p=0.03 4b Prediction of spine BMD from kyphosis, T8-T12, age and %predFEV, Intercept -1.54 0.001 Stepwise analysis results %predFEV, 0.01 0.008 N-59; Adjusted R2=0.10; F-valuc=7.7. p=0.008 * -Independent variables forced into the regression model. 35 Table 17. Intercorrelations for lumbar spine B M D and measures of kyphosis and pulmonary disease severity. T-BMD-spine Kyphosis T8-12 %predFEV, Age T-BMD-spine 1.00 0.11 -0.02 0.32* 0.04 Kyphosis 1.00 -0.13 -0.37* 0.18 T8-T12 1.00 0.33* -0.22 %predFEV, 1.00 , -0.274" Age 1.00 * -Correlation is significant at the 0.01 level (2 tailed). + -Correlation is significant at the 0.05 level (2 tailed). Table 18. Tabulation of whether subjects experienced back pain more than once over the past year tabulated by B M D status. Normal Osteopenic Osteoporotic Cumulative % * Yes 7 27 6 61.5 No 3 18 1 33.8 Not sure 0 2 1 4.6 Total sample available 10 47 8 l i i i i i i i i i i i i i i i i * -Cumulative percent of total available sample. Table 19. Location of back pain by B M D status. Location of back pain Normal Osteopenic Osteoporotic Cumulative % * Cervical 0 2 0 5.1 Thoracic 2 6 4 30.8 Lumbar 4 13 2 48.7 Thoracic and lumbar 1 3 0 10.3 Cervical and lumbar 0 1 0 2.6 Cervical, thoracic and lumbar 0 1 0 2.6 No back pain ::::::::::::^>::::::::::::::::::::::::: 13 |:||§§||§||t|||| Total sample available 10 39 11111111111 ll^lHiiilllllllllill * -Cumulative percent of total available sample. 36 Table 20. Severity of back pain and B M D status. Severity of back pain Normal Osteopenic Osteoporoti c Cumulative % * Very mild 2 2 1 12.8 Mild 0 9 1 25.6 Moderate 5 12 3 51.3 Severe 0 1 1 5.1 Very severe 0 2 0 5.1 No back pain f l i l l l l l i ! 13 iiiitiiiiiiiiiiiiiiiiii Total sample available for analysis 10 39 i i i i i i i i i i i i i i i i i N=56 * -Cumulative percent of total available sample. Table 21. Frequency of back pain and B M D status. Frequency of back pain Normal Osteopenic Osteoporotic Cumulative % * Daily 3 7 2 30.7 Weekly 2 7 2 28.2 Monthly 0 6 1 17.9 Yearly 0 3 0 7.7 Not sure 1 4 1 15.4 No back pain : : : : : : : : : : : : : : : : . 13 l l f l l t l l l l l l l H l i l l l l l l l l l l l l i l ; Total sample available for analysis 10 39 ii|5§!i|ii!!!ii|i| * -Cumulative percent of total available sample. Table 22. Identification of activities patients reported back pain to interfere with. Interferes with: Normal Osteopenic Osteoporotic Cumulative % * Ability to effectively cough 0 8 1 23.1 Breathing 0 4 1 12.8 Exercise 2 9 2 33.3 Work 1 8 0 23.1 Chest physiotherapy 0 3 1 10.3 No back pain 3 13 llliillillllllllii1!! K l l i l l l l l l l l l l l l l i Total sample available for 10 39 llliillllllllllllll l i i l l l lP analysis * -Cumulative percent of total available sample. 37 Table 23. A N O V A results for variables related to pubertal onset by B M D grouping. Variable Normal (Z score <-1.0) Osteopenic (Z score: -1.0 to 2.5) Osteoporotic (Z score > 2.5) p-value Tanner staging standards (years old) Pubertal age (yrs)+ 12.7 (0.8) 9.9-15.7 14.4 (03) 11.3-17.1 13.3 (0.6) 12.7-13.9 0.03* Sample size Pubertal age (yrs) Males only 8 13.1 (1.3) 9.9-15.7 20 14.8 (0.39) 13.1-17.1 2 13.9 0.13 13.0-15.5 Sample size Pubertal age (yrs) Females only 4 13.6 (0.90) 1.0-15.0 16 14.0 (0.37) 12.0-16.5 1 15.5 (0.50) 16.0-16.0 0.72 11.0-13.5 Sample size Age at menarche (yrs) 5 13.5 (0.71) 11.0-15.0 13 14.0 (0.37) 12.0-16.5 2 15.5 (0.50) 15.0-16.0 0.21 12.0-13.0 Menstrual history ,(normal/ab normal menses)** 4/1 5/10 1/1 0.18*** Sample size 5 15 2 + Total sample available including male and female subjects. * Significant post-hoc difference: Normal < Osteopenic. * * Number of subjects with normal versus abnormal menstrual cycles. * * * Chi-square analysis. Table 24. A N O V A results for B M D by menstrual history. Variable Normal Amenohreic p-value menses B M D - TwT(6~4) -1.2 (0.2) 059 composite -3.5-1.7 -2.2-0.1 Sample size 10 12 38 Table 25. Incidence of nonvertebral fractures by B M D group. Fracture site Normal Osteopenic Osteoporotic Ribs 0 1T 1 Hand/finger 1° 4V5 0 Forearin/wrist l p 5< 0 Humerus/elbow 0 2* 0 Skull/maxilla/nose l p 2e 0 Clavicle l a 1 0 Foot/ankle/toe l p ln 0 Tibia/fibula 1 0 Number of subjects who fractured a bone 3/11 (27.3%) 12/48 (25.0%) 1/9(11.1%) Trauma code (minimal/severe) 3/3 9/10 1/0 Age at 1 s t fracture (yrs) 9.3 (2.7) 12.7 (2.3) 12.0 Fracture age range (yrs) 4.0-19.0 4.0-33.0 12.0 a -Fractures of finger and clavicle experienced by same patient at 7 and 11 years of age, resulting from minimal and severe trauma, respectively. P -Fractures of foot, forearm and skull experienced by same patient at 4, 12 and 19 years of age, resulting from minimal and severe (for the latter two episodes) traumas, respectively. y -Fractures of hand and ribs experienced by same patient at 15 and 28 years of age, resulting from minimal trauma. 8 -Fractures of hand and tibia experienced by same patient at 12 and 29 years of age. Fractures were classified as resulting from minimal and severe trauma, respectively. e -Fractures of forearm, skull, skull and elbow experienced by same patient at 12, 21 and 33 years of age. Fractures were classified as the result of severe trauma at 21 and 33 years of age. C, -Fractures of forearm and elbow experienced by same patient at 4 and 18 years of age. Both fractures were classified as the result of minimal trauma. r| -Fractures of tibia and foot experienced by same patient at 4 and 18 years of age. Both fractures were classified as the result of severe trauma. Table 26. Corticosteroid use during puberty and adulthood. Normal Osteopenic Osteoporotic During puberty On steroids 1 2 0 Not on steroids 7 18 2 Total sample available 8 20 2 D i n i n g adulthood On steroids 5 17 5 Not on steroids 5 31 4 Total sample available 10 48 9 40 42 • Pancreatic insufficient I Pancreatic sufficient 1 Normal Osteopenic B o n e mineral densi ty status Osteoporotic Figure 1. Pancreatic sufficiency status and BMD status. Eighty seven percent of the study sample were pancreatic insufficient. 41 •.o i o i: CO. : ™8S^ n o Q . O CU to O o co Q TT CO CU o_ O cu cn O c 'a. o >> CU c c o m 0 5 II CO £ o o o CD CO o o TT co o o CM co o o o CO o o oo CM o o CD CM O O CM O O CM CM O O O CM ao o a -d a <u o< X a W T3 W I Q fi <3 CJ 6 W u W rt T J CD fi 1-. CC <U 00 M •5 3^ O <D I/] rt^ T3 c*-< rv 0 g< fi O 8 OO •c fi rt £ 1 '5 <-> S ^ M op qH (|B0>|) J U n O O O U O | B 0 42 Idietary Ca • supplemental calcium £664.33 Normal (N=9) Osteopenic (N=34) Bone mineral density status Osteoporotic (N=7) Figure 3. Dietary intake of calcium from food and supplements by BMD status. Dashed line indicates recommended nutrient intake (700-800mg) for the general population. Osteoporotic group consumed more calcium and was due to higher supplemental intake. 1400 Idietary vitamin D Dsupplemental vitamin D 1200 Suggested vitamin D intake from foods and supplements. 1000-800 600 400 200 312.3 321 5-700 Normal (N=9) Osteopenic (N=34) Bone mineral density status Osteoporotic (N=7) Figure 4. Dietary intake of vitamin D from food and supplements by BMD status. The osteoporotic group showed a higher intake of total vitamin D which was due to higher supplemental intake. 44 CHAPTER I V DISCUSSION 4.1 PREDICTABILITY OF BMD AND P R E V A L E N C E OF L O W BMD IN ADULT CF PATIENTS We were able to show that pulmonary function, nutritional status, dietary intake and exercise capacity are moderately predictive of BMD. Exercise capacity and BMI were the best predictors but still only explained a small portion of the variance for hip and spine BMD. The prevalence of low B M D in our current sample was high, only 15% of our sample had normal B M D for their age and gender. Thirteen and 72% of our sample were classified as osteoporotic and osteopenic, respectively. Table 2 compares our findings with the literature. Twenty five percent of our sample had B M D Z scores of -2.0 or less, which according to Cummings and colleagues suggests a threefold increased risk of fracture compared to individuals of the same age and gender with normal B M D (49). Haworth and co-workers and Bhudhikanok and associates showed higher prevalence of B M D Z scores of -2 or less (43% and 53%, respectively) in their cohorts of younger CF patients but similar pulmonary disease severity (19,34). This value was only higher in the study by Donovan and associates on CF patients with end-stage pulmonary disease (73.3%) (31). We showed moderate positive relationships between measures of disease severity and BMD. Our results agree with the majority of the previous literature on adult CF patients (19,27,34,37,39,50). The lack of significant correlations between disease severity and B M D in the studies by Donovan and colleagues (31) and others (32,47) are likely related to the homogeneity of their study subjects with respect to disease severity and BMD. Subjects who participated in the study by Bachrach and co-workers (N=22) were classified as either osteopenic or osteoporotic (32), whereas subjects in the other 2 studies were all classified as having severe pulmonary disease and were awaiting lung transplantation. B M D studies on children have generally shown declining trends in B M D with increasing age and pulmonary disease severity (50), even within the first 18 years (29,50). It is clearly difficult to distinguish between progressive decline in pulmonary and overall clinical status in CF patients and aging as they are integrally related. It is more likely that declining B M D seen in studies involving children who have CF are related to increasing disease severity over time. Table 2 presents a cumulative summary of up to date published findings of low B M D in C F adult and children samples studied. There are a number of possible mechanisms by winch severity of pulmonary disease could affect BMD. One theory suggests that chronic infection and resulting respiratory acidosis with its associated increased production of inflammatory cytokines interferes with bone growth and bone mineral accretion (19,25,27,47). Alternatively it is proposed that lifestyle changes such as reduced activity level as patients 45 begin to feel more fatigued, as well as poor nutrition and possibly the increased use of corticosteroids with the progressive increase in pulmonary disease may adversely affect B M D (19). An additional theory to explain the significant bone loss seen relates to lifestyle or other events, which may have occurred during childhood preventing C F children from achieving peak bone mass during the critical pubertal years. We found C F patients classified as osteoporotic to also have been diagnosed as having C F later in life. This factor has implications for attainment of peak bone mass. The misfortune of being diagnosed later in life with C F in this group may have jeopardized pulmonary and nutritional status during childhood and had everlasting implications to adulthood pulmonary and B M D reserve. Later diagnosis would mean that pancreatic abnormalities may have been overlooked and underlying nutritional deficiencies neglected during the interim, consequently affecting (slowing or permanently stunting) growth during puberty and jeopardizing attainment of peak bone mass. Loss in B M D may have started in childhood in this group, or alternatively B M D loss may be greater compared to patients diagnosed within the first few years of life in relative terms because their attained peak bone mass is lower. In summary, as with other studies to date we were also unable to single out parameters, which would predict BMD status in adulthood for CF patients. Exercise capacity and BMI (together with age and gender) were the best predictors of BMD in the current study, but could only explain 35% of the variability in BMD. Pulmonary disease severity and age have been previously reported as predictors of BMD. Dietary intake is important but likely represents a parameter, which has greater implications on BMD in childhood. Age of diagnosis may also have implications for adulthood BMD as it may suggest delayed clinical inten>ention to stabilize the phenotypic characteristics of the disease. 4.2 EXERCISE AND BMD In our study patients reported their exercise activity patterns for 2-4 weeks (modality, duration, frequency and intensity) and also performed a V C V a x test. The categorical variable exercise status (whether subjects exercised or not) did not contribute to predicting BMD but their actual V 0 2 m a x values did. There were a few patients who were only involved in strength-training regimens, whereas the majority of patients who exercised were involved in some form of an aerobic activity program. The intensity, duration and frequency did differ among patients but due to our sample size we were unable to further categorize them and statistically look at differences based on the amount and intensity of exercise these patients were involved in. On a smaller sample (N=30) where we were able to retrospectively review the patients' childhood medical charts we were also able to ascertain their activity patterns during puberty, however we did not show any significant differences between childhood exercise or inactivity and adult BMD. A positive relationship has been shown between BMD and exercise in healthy normal children (51) and exercise has also been shown to play a role in further increasing B M D before the onset of puberty (52). 46 Studies on non-CF populations have shown higher B M D in those young adults who exercise (53). Localized bone hypertrophy is seen in athletes, which is sport specific. This has been elegantly shown by a number of investigators who have reported higher B M D in the dominant arm of young and old tennis players (54-56), higher B M D in the spine and femur of weightlifters who place greater strain or mechanical loading on those bones than endurance trained athletes and the least hypertrophy has been shown in swimmers (57,58). Intervention studies using exercise as a modality to augment B M D have shown modest increases in B M D or maintenance of BMD following aerobic weight-bearing activities such as running (46,59) and less change with low impact aerobic activities such as walking (60). There are conflicting results on the use of strength or resistance training to augment B M D with some studies showing an improvement or maintenance of B M D (58,61) and other studies showing no change in B M D following the intervention (62). Gains in B M D achieved through physical activity have also been shown to be short-lived; the gains are lost with cessation of the activity. Michel et al showed a 20% versus 3.8% decrease in spine B M D in 5 years in long-term runners (55-77 years) when they reduced their weekly training mileage by 20% compared to runners who maintained their training mileage during the study interval (63). It has also been suggested that individuals with higher bone mass to start with loose bone at a higher rate than those individuals with lower bone mass and consequently those who have been avid exercisers and decrease their activity are prone to greater decrements in B M D when their activity is reduced. Dalsky and colleagues showed a return to pre-training B M D levels in a group of postmenopausal women who participated in a 9-22 month exercise program followed by a detraining interval (59). Studies looking at CF patients' activity patterns are limited both in number and sample size but generally have not shown a significant contribution in eliciting higher B M D (27,34). Our study and Haworth and co-workers (19) have shown exercise to be associated with higher BMD. Bhudhikanok and co-workers showed a positive association between physical activity and change in B M D in their adolescent cohort but no association in their adult cohort (38). It is likely that the negative results seen in the former studies in most part are related to their smaller sample size and the more severe pulmonary disease characterizing their samples. It is also possible that differences in study findings are attributable to how each study defines exercise status or activity level. Although the importance of daily activity on B M D should not be dismissed in this population it is likely that other nutritional and clinical parameters may have a dependent influence on exercise activities in this group thus overshadowing the independent influence of daily activity and weight-bearing in maintaining and increasing BMD. 47 The American College of Sports Medicine (ACSM) in their position stand paper on osteoporosis and exercise state that while weight-bearing, activity is necessary for normal bone development and bone maintenance, exercise should not be substituted for hormone replacement therapy in menopausal women. The committee supports the prescription of an exercise regimen, which emphasizes strength, flexibility, coordination and cardiovascular fitness to maintain bone mass and reduce the risk of osteoporotic fractures by reducing the risk of falling (64). Prospective exercise intervention studies have shown little or no increase in BMD. Cross-sectional studies on the other hand have shown higher B M D in those who exercise and in athletes involved in weight-bearing activities compared to sedentary groups. A C S M emphasizes the need for longitudinal studies investigating B M D and exercise. It is unfortunate that in our study our normal and osteoporotic groups were small and we were unable to thus accumulate data on current exercise habits for these groups (with respect to mode of exercise, intensity, duration and frequency) and incorporate some of the findings into designing an exercise program for C F patients. Rose and Jay (65) emphasize that the exercise program for the CF patient should be personalized and take into consideration the patient's level of motivation, cardiorespiratory fitness, muscular strength and degree of postural abnormalities (address breathing mechanics, posture, chest mobility), but must also be flexible to take into account unavoidable setbacks which occur with illness (increased congestion, reduced pulmonary function and exercise tolerance during pulmonary infections). The hip and spine are common sites of reduced B M D and activities prescribed should provide a stimulus to the muscles and bones in these sites. Weight-bearing activities such as running and rowing will provide the greatest stimulus for increasing BMD. Heart rate (HR) is an easy parameter for the CF patient to monitor and use to set their exercise training intensity. Exercise at heart rates above 60% of maximal HR (or age predicted maximal HR) should provide an adequate training stimulus. For C F patients who desaturate during exercise, the exercise intensity should be limited to HR's where oxygen saturation does not fall below 90%, or alternatively they should use supplemental oxygen to maintain oxygen saturation above 90%. Non-weight-bearing activities such as cycling and swimming have not been shown to increase B M D (64). While walking is a weight-bearing activity, faster walking to increase the weight-bearing stimulus is recommended for stimulating bone formation (64). Weight lifting has also been suggested as an exercise modality for increasing B M D , although literature results are conflicting. A minimum frequency of exercise of three times per week with a minimum 20-minute duration is suggested (64,65). An exercise regimen incorporating an aerobic activity such as running, rowing, brisk walking and a strength-training component is recommended. Motivation is the key for beginning an exercise program but also into integrating it into one's lifestyle. Compliance to exercise is not only important in increasing or preventing further losses in BMD but also in maintaining any gains in BMD. Variety in exercise routines and flexibility of the exercise regimen to adapt to the CF patient's changing clinical condition is thus important. 48 In summary, exercise that involves weight bearing and mechanical loading of muscle and bone can increase or maintain BMD. The contribution of regular exercise in maintaining normal BMD in CF patients is but one factor of many, which impact BMD in this population. Progression of pulmonary disease severity will impact the frequency, intensity and duration of exercise performed by a patient as well as the mode of exercise and likely minimize the possible contribution of exercise to augmenting or maintaining BMD. Consequently, pulmonary disease severity is ultimately linked to the level of involvement in exercise in this population. It is difficult to assess the impact of long-term exercise on BMD in CF patients and assess what is an adequate level of exercise to maintain normal BMD due to lack of literature data. Our study does show that those CF patients who are in better physical condition have higher BMD. The relative contributions of exercise on maintaining or increasing BMD are dependent on a patient's long-term commitment to an exercise regimen. V02max, a measure of fitness level was one of our most important predictors of low BMD. 4.3 GROWTH AND REPRODUCTIVE FUNCTION We showed delayed onset of puberty in males and females based on Tanner staging to be associated with reduced BMD. Our results support earlier findings (32,38,66). Baroncelli and co-workers, however, showed similar values for biochemical markers and BMD among CF patients with and without pubertal delay, suggesting that lagging sexual maturation may not be the cause of reduced BMD (37). Specifically within our female cohort, we found age of menarche and puberty on average to be delayed in those CF females who had low BMD in adulthood. Fifty-five percent of females in our study reported amenorrhea or irregular menses but we did not find significantly reduced BMD when female CF patients were grouped based on normal or abnormal gonadal function. Others have reported low BMD in CF females with irregular menses or delayed onset of menarche (31,32,34). Bhudhikanok and co-workers as well showed that these females also had low to normal estradiol and gonadotropin levels indicative of hypogonadism (34). Bachrach and colleagues (1994) additionally showed that these patients with gonadal dysfunction also weighted less, had more severe pulmonary disease and reported less weight-bearing activity (32). Bhudhikanok and co-workers showed that hormone replacement therapy in those patients who have gonadal dysfunction may counteract the negative effects on BMD (38). Although we also showed a similar trend for BMI, weight, pulmonary disease severity, as well as lower total fat mass, these differences were not significant for our sample. At this time no definitive answer can be drawn as it is based on small subsamples and the conclusions are based on descriptive results. Non-CF literature does suggest that delayed puberty and menarcheal age are associated with low BMD (38,67,68). 49 What CF literature has shown to date is that growth as measured by height and weight are diminished in children who have CF. Increases in height and weight have been shown to be strongly associated with increases in BMD in non-CF children (52,69). Inverse associations between height, weight and BMI have been shown with BMD in children and young adults who have CF (29,32,34,38). Baroncelli and co-workers showed reduced height in prepubertal, pubertal and young CF adults in comparison to age, sex and BMI matched healthy controls (37). Bhudhikanok and co-workers showed a trend towards lower percent gains in BMD over time in adolescent CF subjects with pubertal delay (38). This has been partly attributed to suboptimal nutrition during the critical developmental years (39) and suggested to be due to lagging sexual maturation and delayed achievement of peak bone mass (32). Disease severity is likely the main contributor to delayed onset of puberty. CF children who have mild disease are more likely to have an undisrupted growth spurt and Salamoni and associates showed such patients to display similar increases to bone mineral content to age, gender and height matched healthy controls (36). A number of studies have shown disease severity to have a negative impact on physical growth in children who have CF, and to adversely affect nutritional status (38,39,70,71). Gains in BMD over time have been shown in adult CF patients who have experienced pubertal delay and have lower age adjusted BMD compared to CF adults with normal development; hormone replacement or corticosteroid therapy were excluded as contributing variables (38). Poor management of patients' gastrointestinal disease during childhood is likely to contribute to low BMD in adulthood. Feranchak and associates found even in children diagnosed early in life to have CF still to show deficiencies in nutrient intake due to poor compliance to supplement and pancreatic enzyme intake (72). Consequently, we may see increases in BMD in young adults who have CF and were behind in their physical development because of improved compliance in adulthood. In summary, we found delayed onset of puberty to be associated with poorer BMD status. Pulmonary and pancreatic disease severity as well as insufficient nutrient intake are likely important contributors lo low BMD status. 4.4 N U T R I T I O N A L In the present study we attempted to look at diet as a possible predictor of BMD. Current dietary practices are unlikely to be representative of childhood nutritional patterns during the critical period of bone growth, however current nutritional practices could contribute to current bone mineral status. Pancreatic enzyme supplementation both current and during childhood is one factor which impacts skeletal growth during childhood and thus peak bone mass as well as current BMD status. In an attempt to examine the impact of dietary intake on BMD, patients completed a 3-day dietary recall, which included listing calcium and vitamin D supplementation and pancreatic enzyme usage. Current vitamin D and calcium intake seemed to be deficient even in terms of the requirements for a healthy non-CF population. Recommended intakes for vitamin D and calcium are higher for CF patients than for the healthy 50 population because of their pancreatic disease and consequent fat malabsorption. Although there are no established guidelines for levels of calcium and vitamin D intake, which are considered adequate for this group of patients, it has been recognized and is recommended that CF patients increase their intakes of vitamin D and calcium to prevent bone loss during adulthood. Caloric intake, vitamin D and calcium intake, BS-APase and albumin failed to predict BMD. Since these values represent current nutritional status and do not provide us with a picture of the patient's nutritional intake during the pivotal skeletal growth years the low and paradoxical intercorrelations are not surprising. Bachrach and co-workers (32) have reported similar findings. However in their study, they suggested that patients with less severe pulmonary disease consumed larger doses of vitamin D supplements. We found our patients with more severe pulmonary disease to consume higher quantities of vitamin D and calcium, consequently we found inverse associations for BMD and calcium and vitamin D intakes suggesting higher intake of calcium and vitamin D in those with declining BMD. But these differences between studies may be simply related to the clinical practices of each CF clinic. Specifically, it is the practice of our clinic and dietician to prescribe a program of supplementation with multivitamins and minerals, which includes vitamin D (i.e., 800 IU) and calcium (i.e., 500mg) supplementation as well in some cases liquid diet supplementation, in those with more severe pulmonary disease who also seem to be undernourished or underweight. This practice potentially explains the discrepancy in intakes seen across groups. Patient supplementation was not biased by BMD status as the 3-day dietary recalls were completed prior to the dietician's knowledge of the patient's BMD status. Patients in our study did meet their estimated dietary requirements (Figure 2), but again as mentioned above, this only gives us a snapshot of current nutritional practices. Guidelines for the nutritional management of CF patients recommend that CF patients who are also pancreatic insufficient supplement their diet with 2 multivitamins per day (thus providing 800IU of vitamin D) to ensure an adequate intake of fat-soluble vitamins. Equations have been developed to estimate energy requirements and predict energy expenditure based on pancreatic sufficiency, pulmonary disease severity and activity level (43). In our study patients' daily caloric intake did meet estimated requirements (Figure 2) and there were no significant differences between BMD groups. We did find in comparing daily caloric intake with estimated daily energy expenditure (Table 8, kcal-DEE) that those subjects with low BMD were consuming on average approximately 400 calories over DEE, however these subjects also had lower BMI. This would seem to suggest that their pancreatic disease was not adequately managed. In our study there were 7 patients (in addition to our liver transplanted patient on total perienteral nutrition) who were prescribed liquid food supplementation due to their inability to meet their energy demands through conventional food intake in order to augment their dietary intake and meet estimated daily intake (Table 10). Haworth and colleagues also showed an adequate caloric intake in their 51 cohort, however even with an adequate caloric intake and vitamin D supplementation by 900 IU per day, 53% of their subjects showed 25-OHD levels indicative of vitamin D deficiency (19). Other studies where supplementation with vitamin D has also been prescribed (400-800 IU/day) have shown 25-OHD levels characteristic of vitamin D deficiency in some of their adult CF patients (20,22,27,31,32,34,38). . Not all studies, however, have shown low levels of vitamin D metabolites. Henderson and Madsen did not show a relationship between calcium intake and lumbar and femoral BMD in their CF children (29). They reported normal values for 25-OHD and 1,25-di-OHD in all but one patient and additionally were unable to show an association between serum levels of 25-OHD, 1,25-di-OHD, calcium, phosphorus and BMD values. Bachrach et al (32) showed similar serum levels of calcium, phosphorus, PTH and 1,25-di-OHD in the majority of their sample to values expressed in Henderson and Madsen (29), as have other studies (27,36). Bhudhikanok and co-workers conversely showed greatly reduced levels of 25-OHD, but normal 1,25-di-OHD levels in 14 of the 28 patients measured (34). Similar findings were reported in their follow-up study (38). Because vitamin D is formed in the skin, the low levels of vitamin D metabolites (i.e., 25-OHD and 1,25-OHD) seen in CF patients might be related to inadequate exposure to sunlight. Reiter and colleagues measured sun exposure, dietary vitamin D intake and serum calcium and vitamin D metabolites in young CF patients (12-25 years of age) compared to healthy age-matched controls (73). They showed a direct linear relationship in CF patients between hours of sunlight exposure and 25-OHD (r=0.49) (with an estimated daily dietary vitamin D intake of 1000 IU and serum calcium levels within normal range (i.e., 2.2-2.6 ininoll"1)). The 25-OHD and 1,25-OHD levels in this study, although they were higher in the controls were still within the normal range in the CF cohort. Most of their 20 patients were classified as osteopenic. We also had patients record the number of hours tiiey spend outdoors on those days when they completed the dietary recall, and except for a few patients in the osteoporotic group who were also characterized with severe pulmonary disease, the majority of patients spent at least an average of 2 hours per day outside. Baroncelli and co-workers showed increased bone resorption in prepubertal CF patients, decreased bone formation rate and increased bone resorption rate in pubertal CF patients and increased bone resorption in young CF adults compared to healthy age, sex and BMI matched controls (37). CF patients were on a prescribed diet, which included a daily caloric intake that was 150% of the recommended dietary allowances, vitamin D (800 IU/day) and calcium (728-1150 mg/day based on age) supplementation. A marker of bone resorption, which we investigated in our study, was bone specific alkaline phosphatase; we did not find any differences between BMD groups or any significant associations with BMD or current dietary practices. Haworth and co-workers, in their large study did show a significant negative association between BMD and serum levels of BS-APase, c-reactive protein, osteocalcin, parathyroid hormone and total APase suggesting increased bone turnover in those patients with low BMD (19). The authors also 52 suggested a possible association of bone turnover with CFTR genotype. CF patients who were homozygous for the AF508 mutation had higher levels of some bone turnover markers (i.e., BS-APase and urinary deoxypyridinoline crosslinks) compared to heterozygotes for the AF508 mutation.. Significant direct moderate associations between BMD and BMI were shown in this study and have also been shown by others (19,26,27,31,32,35). In their study looking at change in BMD over 1.5 years Bhudhikanok and co-workers showed a positive association of change in BMI (and weight) with change in BMD in adult CF patients (38). We found fat and lean muscle mass to be positively associated with BMD. Similar associations have been reported among non-CF populations as well (74). The high correlation we found between total fat mass and BMI and their significant associations with BMD suggest a possible independent role of pancreatic disease severity and resulting malnutrition on BMD. The independent role of pancreatic disease severity and pulmonary disease severity on BMD has been previously suggested by Grey and associates (27). As discussed earlier we found CF patients who were osteoporotic to be initially diagnosed as having CF much later in life. Feranchak and colleagues prospectively evaluated fat-soluble vitamin status longitudinally in 127 infants who were identified as having CF within 6 weeks postpartum (72). Deficiency of one or more fat-soluble vitamins was present in about half of the study population at initial screening and despite fat-soluable vitamins and pancreatic enzyme supplementation, 10%-57% of the subjects continued to show deficiency. Patients diagnosed later in life with CF would therefore be at risk for chronic nutritional deficiencies and such a negative energy and nutrient balance could prove to be suboptimal for growth. Ultimately this nutritional deficit could compromise peak bone mass attained in childhood and lead to a higher bone turnover rate during the pre-diagnosis period. It is also recommended that CF patients be prescribed calcium supplements, however the level of supplementation is still arbitrary. Aris and associates showed lower absorption of calcium in CF patients versus healthy controls, with absorption improving with pancreatic enzyme intake (75). Calcium intake for our cohort was above recommended nutritional intake for the general population and intakes were similar or greater than values reported by Haworth and co-workers (19). Bhudhikanok and associates showed a direct association of calcium intake in adult CF patients with changes in BMD (38). In order to definitively answer the question of whether CF patients receive adequate calcium, vitamin D and overall total caloric intake and investigate the association with BMD, a prospective longitudinal study is needed which would follow CF children from adolescence to adulthood and periodically document their dietary intake patterns. Lastly, as we stated in tire previous section disease severity has a negative impact on physical growth and can ultimately affect nutritional status. Henderson and associates showed CF children (mean age=11.9 53 yrs; SD=4.0 yrs) to have body fat values, which were 18% lower than age, sex and race matched normal controls (39). Salamoni and co-workers studied 14 well nourished with fairly mild to moderate pulmonary disease severity (mean S-K score=76) CF children ranging in age from 7-20 years (mean age 12.2) (36). They were matched with similar gender, age and pubertal staging healthy classmates and the authors did not find any differences in bone mineral content and body composition (fat and lean muscle mass) between the 2 groups. Adequate dietary intake, which would also suggest on average a commensurate intake of vitamins and minerals has been associated in non-CF studies with normal BMD, body composition, onset of puberty and attainment of peak bone mass (52). Bachrach and co-workers have also suggested that CF patients surviving into their fourth decade likely had less severe pulmonary disease during childhood and are able to achieve greater peak bone mass or alternatively show a slower rate of decline in bone mass (32). In summary, dietary intake during childhood greatly influences BMD status in adulthood. CF patients who require pancreatic enzymes, are underweight and have low total body fat mass are more likely to have reduced BMD. Late diagnosis of CF can leave patients undernourished but also malnourished during their growing years and as a consequence have lower BMD in adulthood. Dietary intake parameters examined during adulthood fail to predict BMD status. Plasma levels of bone formation and turnover markers suggest the presence of disturbances in bone mineralization, which start in childhood. There is evidence from the present study and previous studies to suggest that inadequate dietary intake and subsequent weight gain during those pivotal adolescent years are directly responsible for final peak bone mass attained and the pattern of bone loss seen during adulthood. Calcium and vitamin D supplementation are necessary to stop or reduce bone loss. CFTR genotype may play a role in bone turnover. Plasma levels of bone turnover markers, vitamin D and calcium levels may not predict BMD status in adulthood. 4.5 VERTEBRAL FRACTURES, KYPHOSIS AND ASSOCIATION WITH B A C K PAIN We measured the prevalence of vertebral fractures (T8-T12). A vertebra was considered fractured if the ratio of the anterior or middle vertebral body height to the posterior vertebral body height was 0.80 or less. Patients in our study did not show vertebral fractures. The average values of vertebrae 8 through 12 were compared between bone mineral density groupings and we found no significant differences between groups. We did find a low but significant association between vertebral wedging and pulmonary disease severity, but no association with increasing age (Table 17). Grey and co-workers reported 5 out of die 16 CF patients whom they tested to exhibit vertebral wedging of at least one thoracic vertebra and 4 of those patients were classified as osteopenic (27). Vertebral fractures may go unnoticed was shown by Aris and co-workers (47). They showed an increased prevalence of vertebral fractures following a review of their patients' chest radiographs. These patients were older and were characterized as having more severe 54 pulmonary disease. Ross and co-workers described a significantly higher incidence of kyphosis in those patients with more pronounced vertebral wedging and showed direct relationships between increasing kyphosis and vertebral wedging with increasing severity of pulmonary disease in their CF sample (76). They also additionally showed kyphosis to be more pronounced in older CF patients. We found no significant relationships between T8-T12 vertebral fractures and low BMD and similarly between degree of kyphosis and low BMD (Table 17). Degree of kyphosis was measured indirectly in our study using a surveyor's flexicurve. Values on a sample of healthy normal controls of similar age, stature and gender distribution was obtained for comparison to determine normal values. A higher prevalence of apparent kyphosis was seen in our CF patients (i.e., 36%) suggesting an exaggerated curvature of the spine compared to the healthy controls which all exhibited values within the normal range. We did not find increasing degree of kyphosis with declining BMD. Henderson and Specter (28) and some of the early studies investigating kyphosis in CF patients have reported increased prevalence of kyphosis in those patients with more severe pulmonary disease (77), whereas others have been unable to demonstrate such a relationship (78,79). We showed a moderate association with pulmonary disease severity, and no association widi increasing age which has been previously also suggested (28,45,76). Henderson and Madsen looked at the relationship by comparing CF children BMD values with excessive thoracic kyphosis to those with normal kyphosis, and although they argued that there was a trend for lower BMD values in those categorized with excessive kyphosis their findings were not statistically significant (28). Complaints of lower back pain are not necessarily related to low BMD of the spine or to overall low BMD was our study finding. As can be seen in table 18 even patients with normal BMD complained of back pain. Due to the small sample sizes in our normal and osteoporotic groups we are only able to descriptively present our results. Most patients who complained of back pain reported that they experienced back pain daily or weekly and the pain was of mild to moderate intensity localized in the lumbar and/or thoracic spine. Patients also described that their back pain interfered with their ability to effectively cough and perform chest physiotherapy, exercise and in a smaller percentage interfered with breathing and work performance. Back pain in CF patients has previously been investigated by Ross and co-workers (76), who showed that the majority of their CF cohort complained of back pain and it was similarly of a chronic nature, of mild to moderate intensity, localized in the thoracic and lumbar spine and interfered with similar activities that is coughing, exercise, breathing and chest physiotherapy. Although this study did not look at BMD, they did perform a number of standard postural examination tests evaluating range of motion of upper body and trunk, and muscle testing of the shoulders, chest, abdomen and back as described by the American Academy of Othropaedic Surgeons (1965). They showed back pain in CF patients to be associated with 55 reduced shoulder, abdominal, trunk and chest mobility and suggested that muscle weakness and reduced flexibility likely contribute to the flexed spinal posture characteristic in this population. This posture is suggested to be partially established by this population's chronic labored breathing, increased coughing (76,77,79) and hyperinflation (26) and not necessarily because of structural thoracic kyphosis. Henderson and Specter (1994), suggest that hyperinflation may contribute to excessive compression of the anterior portion of vertebral bodies causing a growth abnormality, which ultimately leads to the wedge-shaped deformity of the vertebrae and kyphosis. In summary, kyphosis contributes to height loss and to disability. Although there is evidence of increasing kyphosis and vertebral wedging with increasing severity of pulmonary disease in CF patients, these two conditions do not seem to be strongly associated with low BMD. The weak associations may simply be a statistical issue related to the small number of CF patients characterized in each study as normal or osteoporotic. The kyphotic posture can be related to vertebral wedging but does not ha\>e to be and has been shown to be simply related in some CF patients to decreased lung volume, their posture due to chronic labored breathing and coughing, and muscle weakness. Whether this posture together with abnormal bone mineralization actually contributes to vertebral compression fractures and kyphosis is unclear. Back pain is a common complaint by CF patients and is reported to interfere with their ability to effectively cough and perform chest physiotherapy and exercise, however current findings seem to suggest that back pain does not have to be associated with low BMD, but can be simply associated with muscle weakness and poor flexibility in this population. 4.6 NONVERTEBRAL FRACTURES Due to the small number of patients in the normal and osteoporotic groups, the prevalence of nonvertebral fractures is only presented descriptively. It would seem that patients with normal or osteopenic BMD are more prone to nonvertebral fractures. Patients in these 2 groups were also more likely to rate themselves as leading an active lifestyle. And we did find that these patients also have higher aerobic capacity, substantiating their subjective rating of activity level. These 2 groups predominately had less severe pulmonary disease and thus are likely to participate in increased trauma-producing activities compared to the osteoporotic group, which seems to exhibit more severe pulmonary disease and likely spend more of their time at home resting. Others have drawn similar conclusions in their studies (32,34,38). These authors reported 12 fractures in 9 of the 71 CF patients (children and adults combined) who participated in their study (32). In their 1998 follow-up study, they showed fractures to be more commonly presented in those patients who were on long-term steroid therapy, had low BMD at baseline and showed further losses in BMD at follow-up and displayed short stature in adulthood (38). Aris and co-workers, who looked at an older cohort with more severe pulmonary disease reported a higher fracture rate (37 out of 70 had a history of fractures which occurred as early as 6 years of age) and additionally found unreported 56 fractures of ribs and vertebrae from review of these patients' chest radiographs (47). A higher fracture rate was shown in the pubertal CF group compared to their prepubertal and young adult CF groups and healdiy controls in the study by the Baroncelli group (37). The prevalence of fractures in their CF patients was 10.2% compared to 2.8% in their control group. Additionally in the CF groups the fractures were classified as the result of minimal trauma compared to major trauma in the control groups. In summary, non-vertebral fracture rate is higher in CF patients and fractures in this group are more commonly related to minor trauma, which would otherwise occur under major trauma conditions in the non-CF population. CF patients with low BMD in childhood, who show further BMD losses during puberty and short stature in adulthood are more likely to experience fractures. We are more likely to see CF patients with less severe pulmonary disease report fractures during both childhood and adulthood and this can be attributed to the more active lifestyle these patients can afford compared to the CF patient with more severe pulmonary disease which due to their pulmonary disease may lead a more sedentary lifestyle. 4.7 STEROID USE Corticosteroid use in puberty and adulthood were not shown to contribute to low BMD. Corticosteroid use, duration and dose have not been shown as well in previous studies to impact BMD (29,32,35,37). Bhudhikanok and colleagues reported the highest use of corticosteroids (approximately 50% of their sample, mean duration of steroid use=4.9 years), but they were still unable to show any predictive power of steroid use and dose on BMD, even though they also showed a high prevalence of low BMD (see table 2) (34). In a follow-up study (conducted 1.5 years later) Bhudhikanok and colleagues reassessed their cohort and showed either no change or lower BMD in 65% of the CF children studied (38). They showed CF adults and children receiving regular corticosteroid therapy were more likely to show BMD loss at follow-up and fractures were more common in those subjects on corticosteroids (38). Usually in most studies the percentage of the study population receiving corticosteroids has been low, the retrospective nature of many studies and the possible association of corticosteroid use with otiier clinical parameters may be factors, which potentially mask the independent contribution of corticosteroid use (29,32). Studies in patients who have asthma have shown oral corticosteroid use and in some studies even inhaled corticosteroid use to adversely affect BMD (80,81). Shane and associates also found an inverse association between steroid use and BMD in COPD patients but not for CF patients in their sample of lung transplantation candidates with end-stage pulmonary disease (35). In summary, short-term intermittent treatment with inhaled or oral corticosteroids does not seem to affect BMD in adulthood. Patients with long-term exposure were more likely to show low BMD, more severe pulmonary disease, pubertal delay, smaller frame size and higher fracture incidence. There is some 57 debate as to whether the affects of corticosteroid use may be masked by other clinical abnormalities in this group. It is also likely that a minimal association is shown in the literature between corticosteroid use and BMD in CF patients, while a strong association is commonly shown in patients who have asthma and this is related to differences in dose and duration of use. 58 C H A P T E R V C O N C L U S I O N 5.0 C O N C L U D I N G C O M M E N T S Osteoporosis is caused by a failure to accumulate BMD at a normal rate and / or an accelerated loss of bone mineral. BMD is reduced in adult patients who have CF and is related to markers of pulmonary disease severity, nutritional deficits, poor exercise tolerance. Osteoporosis is also a known common complication of organ transplantation and it is Uierefore important to assist CF patients to maintain normal BMD values so as to reduce pre and post-transplant fracture rate. There is a consensus for supplementation with fat-soluble vitamin D and calcium intake to ensure adequate serum levels to prevent or slow down bone turnover. As maintenance of BMD is a life-long process, it is important to make certain that CF patients receive adequate caloric and nutrient intake during the critical growth years in order to achieve peak bone mass. Adequate pancreatic enzyme supplementation in those patients who are diagnosed as pancreatic insufficient is also imperative to ensure maximal absorption of vitamin D and other nutrients and therefore prevent malnutrition. Early diagnosis of CF will also guarantee that malabsorption problems are aggressively addressed during childhood. Descriptive results support literature findings that onset of puberty may be delayed and CF patients with delayed onset of puberty are more likely to have low BMD. Exercise should be encouraged not only during adultiiood but also in childhood as a mode of enhancing BMD. Aggressive treatment of pulmonary infections will reduce the rate of clinical deterioration and possibly the negative impact of inflammatory cytokines on BMD. Longitudinal studies investigating BMD, diet, exercise, growth and attainment of peak bone mass and corticosteroid use need to be initiated to definitively determine the impact of childhood practices on adult BMD. 59 6.0 B I B L I O G R A P H Y 1. Wallace, C. S., M. Hall, and R. J. Kulin. 1993. Pharmacologic management of cystic fibrosis. Clinical Pharmacy 12:657-674. 2. Tsui, L. C , M. Buchwald, D. Barker, J. C. Braman, R. Knowlton, J. W. Schumm, H. Eiberg, J. Mohr, D. Kennedy, N. Plavsic, and et al. 1985. Cystic fibrosis locus defined by a genetically linked polymorphic DNA marker. Science 230:1054-1057. 3. Rommens, J. M., M. C. Iannuzzi, B. Kerem, M. L. Drumm, G. Melmer, M. Dean, R. Rozmahel, J. L. Cole, D. Kennedy, N. Hidaka, and et al. 1989. Identification of the cystic fibrosis gene: chromosome walking and jumping. Science 245:1059-1065. 4. Burke, W., M. L. Aitken, S. H. Chen, and C. R. Scott. 1992. Variable severity of pulmonary disease in adults with identical cystic fibrosis mutations. Chest 102:506-509. 5. Stamp, T. C , and D. M. Geddes. 1993. Osteoporosis and cystic fibrosis [editorial; comment]. Thorax 48:585-586. 6. Aris, R. M. , I. P. Neuringer, M. A. Weiner, T. M. Egan, and D. Ontjes. 1996. Severe osteoporosis before and after lung transplantation. Chest 109:1176-1183. 7. Tanner, J. M. 1990. Foetus into man: physical growth from conception to maturity. Havard University Press, Cambridge, Massachusetts. 8. Aloia, J. F. 1994. The gain and loss of bone in the human life cycle. [Review] [112 refs]. Advances in Nutritional Research 9:1-33. 9. Lam, C , M. Kattan, A. Collins, and J. Kleinerman. 1983. Long-term sequelae of bronchiolitis induced by nitrogen dioxide in hamsters. Am. Rev. Respir. Dis. 128:1020-1023. 10. Davis, P. B., M. Drumm, and M. W. Konstan. 1996. Cystic fibrosis. [Review]. Am. J. Respir. Crit. Care Med. 154:1229-1256. 11. Sturgess, J., and J. Imrie. 1982. Quantitative evaluation of the development of tracheal submucosal glands in infants with cystic fibrosis and control infants. American Journal of Pathology 106:303-311. 12. Reid, L., and R. d. Haller. 1967. The bronchial mucous glands-their hypertrophy and change in intracellular mucus. Bibliotheca Paediatrica 86:195-199. 13. Zuelzer, W. W., and W. A. Newton. 1949. The pathogenesis of fibrocystic disease of the pancreas. A study of 36 cases with special reference to pulmonary lesions. Pediatrics 4:53-69. 14. Oppenheimer, E. H , and J. R. Esterly. 1975. Pathology of cystic fibrosis review of the literature and comparison with 146 autopsied cases. Perspectives in Pediatric Pathology 2:241-278. 15. Bedrossian, C. W., S. D. Greenberg, D. B. Singer, J. J. Hansen, andH. S. Rosenberg. 1976. The lung in cystic fibrosis. A quantitative study including prevalence of pathologic findings among different age groups. Human Pathology 7:195-204. 16. Manolagas, S. C. 1995. Role of cytokines in bone resorption. [Review] [45 refs]. Bone 17:63S-67S. 60 17. Manolagas, S. C , and R. L. Jilka. 1995. Bone marrow, cytokines, and bone remodeling. Emerging insights into the pathophysiology of osteoporosis. [Review] [63 refs]. New England Journal of Medicine 332:305-311. 18. Evans, C. H., S. C. Watkins, and M. Stefanovic-Racic. 1996. Nitric oxide and cartilage metabolism. [Review] [61 refs]. Methods in Enzymology 269:75-88. 19. Haworth, C. S., P. L. Selby, A. K. Webb, M. E. Dodd, H. Musson, L. N. R. Mc, G. Economou, A. W. Horrocks, A. J. Freemont, E. B. Mawer, and J. E. Adams. 1999. Low bone mineral density in adults with cystic fibrosis. Thorax 54:961-967. 20. Hanly, J. G., M. J. McKenna, C. Quigley, R. Freaney, F. P. Muldowney, and M. X. FitzGerald. 1985. Hypovitaminosis D and response to supplementation in older patients witii cystic fibrosis. Quarterly Journal of Medicine 56:377-385. 21. Solomons, N. W., J. B. Wagonfeld, C. Rieger, R. A. Jacob, M. Bolt, J. V. Horst, R. Rothberg, and H. Sandstead. 1981. Some biochemical indices of nutrition in treated cystic fibrosis patients. American Journal of Clinical Nutrition 34:462-474. 22. Hahn, T. J., A. E. Squires, L. R. Halstead, and D. B. Strominger. 1979. Reduced serum 25-hydroxyvitamin D concentration and disordered mineral metabolism in patients with cystic fibrosis. Journal of Pediatrics 94:38-42. 23. Genant, H. K., K. Faulkner, and C.-C. Gluer. 1991. Measurement of bone mineral density: Current Status. The american Journal of medicine 9L49S-53S. 24. Mundy, G. R. 1999. Bone remodelling and its disorders. Login Brothers Book Company, London. 25. Mischler, E. H , P. J. Chesney, R. W. Chesney, and R. B. Mazess. 1979. Demineralization in cystic fibrosis detected by direct photon absorptiometry. Am. J. Dis. Child. 133:632-635. 26. Gibbens, D. T., V. Gilsanz, M. I. Boechat, D. Dufer, M. E. Carlson, and C. I. Wang. 1988. Osteoporosis in cystic fibrosis. Journal of Pediatrics 113:295-300. 27. Grey, A. B., R. W. Ames, R. D. Matthews, and I. R. Reid. 1993. Bone mineral density and body composition in adult patients with cystic fibrosis. Thorax 48:589-593. 28. Henderson, R. C , and B. B. Specter. 1994. Kyphosis and fractures in children and young adults with cystic fibrosis. Journal of Pediatrics 125:208-212. 29. Henderson, R. C , and C. D. Madsen. 1996. Bone density in children and adolescents with cystic fibrosis. Journal of Pediatrics 128:28-34. 30. Haworth, C. S., P. L. Selby, A. K. Webb, and J. E. Adams. 1998. Osteoporosis in adults with cystic fibrosis. Journal of the Royal Society of Medicine 91:14-18. 31. Donovan, D. S., Jr., A. Papadopoulos, R. B. Staron, V. Addesso, L. Schulman, C. McGregor, F. Cosman, R. L. Lindsay, and E. Shane. 1998. Bone mass and vitamin D deficiency in adults with advanced cystic fibrosis lung disease. Am. J. Respir. Crit. Care Med. 157:1892-1899. 61 32. Bachrach, L. K., C. W. Loutit, and R. B. Moss. 1994. Osteopenia in adults with cystic fibrosis. Am. J. Med. 96:27-34. 33. Flores, H. L., B. Zemel, V. A. Stallings, and T. F. Scanlin. 1995. Bone mineral density in adolescent females with cystic fibrosis. (Abstract). Pediatric Pulmonology 12:S261. 34. Bhudhikanok, G. S., J. Lim, R. Marcus, A. Harkins, R. B. Moss, and L. K. Bachrach. 1996. Correlates of osteopenia in patients with cystic fibrosis. Pediatrics 97:103-111. 35. Shane, E., S. J. Silverberg, D. Donovan, A. Papadopoulos, R. B. Staron, V. Addesso, B. Jorgesen, C. McGregor, and L. Schulman. 1996. Osteoporosis in lung transplantation candidates with end-stage pulmonary disease, yim. J. Med. 101:262-269. 36. Salamoni, F., M. Roulet, F. Gudinchet, M. Pilet, D. Thiebaud, and P. Burckhardt. 1996. Bone mineral content in cystic fibrosis patients: correlation with fat-free mass. Arch. Dis. Child. 74:314-318. 37. Baroncelli, G. I., F. De Luca, G. Magazzu, T. Arrigo, C. Sferlazzas, C. Catena, S. Bertelloni, and G. Saggese. 1997. Bone demineralization in cystic fibrosis: evidence of imbalance between bone formation and degradation. Pediatric Research 41:397-403. 38. Bhudhikanok, G. S., M. C. Wang, R. Marcus, A. Harkins, R. B. Moss, and L. K. Bachrach. 1998. Bone acquisition and loss in children and adults with cystic fibrosis: a longitudinal study. Journal of Pediatrics 133:18-27. 39. Henderson, R. C , and C. D. Madsen. 1999. Bone mineral content and body composition in children and young adults with cystic fibrosis. Pediatric Pulmonology 27:80-84. 40. Anonymous. 1995. Standardization of Spirometry, 1994 update. American Thoracic Society. American Journal of Respiratory and Critical Care Medicine 152:1107-1136. 41. Shwachman, H., and L. Kulczycki. 1958. Long-term study of one hundred five patients with cystic fibrosis, yim. J. Dis. Child. 96:6-15. 42. Brasfield, D., G. Hicks, S. Soong, and R. E. Tiller. 1979. The chest roentgenogram in cystic fibrosis: a new scoring system. Pediatrics 63:24-29. 43. Ramsey, B. W., P. M. Farrell, and P. Pencharz. 1992. Nutritional assessment and management in cystic fibrosis: a consensus report. The Consensus Committee. American Journal of Clinical Nutrition 55:108-116. 44. Libennan, U. A., S. R. Weiss, J. Broil, H. W. Minne, H. Quan, N. H. Bell, J. Rodriguez-Portales, R. W. Downs, Jr., J. Dequeker, and M. Favus. 1995. Effect of oral alendronate on bone mineral density and the incidence of fractures in postmenopausal osteoporosis. The Alendronate Phase III Osteoporosis Treatment Study Group [see comments]. New England Journal of Medicine 333:1437-1443. 45. Milne, J. S., and I. J. Lauder. 1976. The relationship of kyphosis to the shape of vertebral bodies. Ann. Hum. Biol. 3:173-179. 62 46. Chow, R. K., and J. E. Harrison. 1987. Relationship of kyphosis to physical fitness and bone mass on post-menopausal women. American Journal of Physical Medicine 66:219-227. 47. Aris, R. M , J. B. Renner, A. D. Winders, H. E. Buell, D. B. Riggs, G. E. Lester, and D. A. Ontjes. 1998. Increased rate of fractures and severe kyphosis, sequelae of living into adulthood with cystic fibrosis. Ann. Intern. Med. 128:186-193. 48. Ott, S. M , and M. L. Aitken. 1998. Osteoporosis in patients with cystic fibrosis. Clinics in Chest Medicine 19:555-567. 49. Cummings, S. R., D. M. Black, M. C. Nevitt, W. Browner, J. Cauley, K. Ensrud, H. K. Genant, L. Palermo, J. Scott, and T. M. Vogt. 1993. Bone density at various sites for prediction of hip fractures. The Study of Osteoporotic Fractures Research Group [see comments]. Lancet 341:72-75. 50. Laursen, E. M., C. Molgaard, K. F. Michaelsen, C. Koch, and J. Muller. 1999. Bone mineral status in 134 patients with cystic fibrosis. Arch. Dis. Child. 81:235-240. 51. Kroger, H., A. Kotaniemi, P. Vainio, and E. Alhava. 1992. Bone densitometry of the spine and femur in children by dual-energy x-ray absorptiometry [published erratum appears in Bone Miner 1992 Jun;17(3):429]. Bone & Mineral 17:75-85. 52. Slemenda, C. W., T. K. Reister, S. L. Hui, J. Z. Miller, J. C. Christian, and C. C. Johnston, Jr. 1994. Influences on skeletal mineralization in children and adolescents: evidence for varying effects of sexual maturation and physical activity. Journal of Pediatrics 125:201-207. 53. Cooper, C , M. Cawley, A. Bhalla, P. Egger, F. Ring, L. Morton, and D. Barker. 1995. Childhood growth, physical activity, and peak bone mass in women. J. Bone Miner. Res. 10:940-947. 54. Huddleston, A. L., D. Rockwell, D. N. Kulund, and R. B. Harrison. 1980. Bone mass in lifetime tennis athletes. JAMA 244:1107-1109. 55. Jones, H. H., J. D. Priest, W. C. Hayes, C. C. Tichenor, and D. A. Nagel. 1977. Humeral hypertrophy in response to exercise. Journal of Bone & Joint Surgery - American Volume 59:204-208. 56. Montoye, H. J., R. Gayle, and M. Higgins. 1980. Smoking habits, alcohol consumption and maximal oxygen uptake. Medicine & Science in Sports & Exercise 12:316-321. 57. Heinrich, C. H., S. B. Going, R. W. Pamenter, C. D. Perry, T. W. Boyden, and T. G. Lohman. 1990. Bone mineral content of cyclically menstruating female resistance and endurance trained athletes. Medicine & Science in Sports & Exercise 22:558-563. 58. Nilsson, B. E., and N. E. Westlin. 1971. Bone density in athletes. Clinical Orthopaedics & Related Research 11'.179-182. 59. Dalsky, G. P., K. S. Stocke, A. A. Ehsani, E. Slatopolsky, W. C. Lee, and S. J. Birge, Jr. 1988. Weight-bearing exercise training and lumbar bone mineral content in postmenopausal women. Ann. Intern. Med. 108:824-828. 63 60. Hatori, M , A. Hasegawa, H. Adachi, A. Shinozaki, R. Hayashi, H. Okano, H. Mizunuma, and K. Murata. 1993. The effects of walking at the anaerobic threshold level on vertebral bone loss in postmenopausal women. Calcif. Tissue Int. 52:411-414. 61. Pruitt, L. A., R. D. Jackson, R. L. Bartels, and H. J. Lehnhard. 1992. Weight-training effects on bone mineral density in early postmenopausal women. J. Bone Miner. Res. 7:179-185. 62. Sinaki, M., H. W. Wahner, K. P. Offord, and S. F. Hodgson. 1989. Efficacy of nonloading exercises in prevention of vertebral bone loss in postmenopausal women: a controlled trial. Mayo Clinic Proceedings 64:762-769. 63. Michel, B. A., N. E. Lane, A. Bjorkengren, D. A. Bloch, and J. F. Fries. 1992. Impact of running on lumbar bone density: a 5-year longitudinal study. Journal of Rheumatology 19:1759-1763. 64. American College of Sports Medicine Position Stand on Osteoporosis and Exercise. 1995. ACSM position stand on osteoporosis and exercise. Medicine and Science in Sports and Exercise 27:i-vii. 65. Rose, J., and S. Jay. 1986. A comprehensive exercise program for persons with cystic fibrosis. Journal of pediatric nursing 1:323-334. 66. Shaw, N., C. Bedford, D. Heaf, H. Carty, and J. Dutton. 1995. Osteopenia in adults with cystic fibrosis [letter; comment]. Am. J. Med. 99:690-692. 67. Bachrach, L. K., D. K. Katzman, I. F. Litt, D. Guido, and R. Marcus. 1991. Recovery from osteopenia in adolescent girls with anorexia nervosa. Journal of Clinical Endocrinology & Metabolism 72:602-606. 68. Kreipe, R. E., B. H. Churchill, and J. Strauss. 1989. Long-term outcome of adolescents with anorexia nervosa. Am. J. Dis. Child. 143:1322-1327. 69. Katzman, D. K., L. K. Bachrach, D. R. Carter, and R. Marcus. 1991. Clinical and anthropometric correlates of bone mineral acquisition in healthy adolescent girls. Journal of Clinical Endocrinology & Metabolism 73:1332-1339. 70. Huang, N. N., D. V. Schidlow, T. H. Szatrowski, J. Palmer, L. R. Laraya-Cuasay, W. Yeung, K. Hardy, L. Quitell, and S. Fiel. 1987. Clinical features, survival rate, and prognostic factors in young adults with cystic fibrosis. Am. J. Med. 82:871-879. 71. Kraemer, R., A. Rudeberg, B. Hadorn, and E. Rossi. 1978. Relative underweight in cystic fibrosis and its prognostic value. A eta Paediatr. Scand. 67:33-37. 72. Feranchak, A. P., M. K. Sontag, J. S. Wagener, K. B. Hammond, F. J. Accurso, and R. J. Sokol. 1999. Prospective, long-term study of fat-soluble vitamin status in children with cystic fibrosis identified by newborn screen. Journal of Pediatrics 135:601-610. 73. Reiter, E. O., S. M. Brugman, J. W. Pike, M. Pitt, S. Dokoh, M. R. Haussler, R. S. Gerstle, and L. M. Taussig. 1985. Vitamin D metabolites in adolescents and young adults with cystic fibrosis: effects of sun and season. Journal of Pediatrics 106:21-26. 64 74. Michaelsson, K., R. Bergstrom, H. Mallmin, L. Holmberg, A. Wolk, and S. Ljunghall. 1996. Screening for osteopenia and osteoporosis: selection by body composition. Osteoporosis International 6:120-126. 75. Aris, R. M , G. E. Lester, S. Dingman, and D. A. Ontjes. 1999. Altered calcium homeostasis in adults with cystic fibrosis. Osteoporosis International 10:102-108. 76. Ross, J., J. Gamble, A. Schultz, and N. Lewiston. 1987. Back pain and spinal deformity in cystic fibrosis. Am. J. Dis. Child. 141:1313-1316. 77. Denton, J. R., R. Tietjen, and P. F. Gaerlan. 1981. Thoracic kyphosis in cystic fibrosis. Clinical Orthopaedics & Related Research 1\-1A. 78. Logvinoff, M. M., G. T. Fon, L. M. Taussig, and M. J. Pitt. 1984. Kyphosis and pulmonary function in cystic fibrosis. Clinical Pediatrics 23:389-392. 79. Erkkila, J. C , W. J. Warwick, and D. S. Bradford. 1978. Spine deformities and cystic fibrosis. Clinical Orthopaedics & Related Research 146-150. 80. Adinoff, A. D., and J. R. Hollister. 1983. Steroid-induced fractures and bone loss in patients with asthma. New England Journal of Medicine 309:265-268. 81. Packe, G. E. , J. G. Douglas, A. F. McDonald, S. P. Robins, and D. M. Reid. 1992. Bone density in asthmatic patients taking high dose inhaled beclomethasone dipropionate and intermittent systemic corticosteroids. Thorax 47:414-417. 7.1 A P P E N D I X A Sample of questionnaires used in the study • Exercise activity questionnaire / diary • Reproductive history questionnaire • Bone fracture questionnaire • Back pain questionnaire Exercise activity qucstionnai 67 SHEET #: D A I L Y A C T I V I T Y P L A N N E R You are not expected to exercise daily or feel you need to for this study. We would like you to simply record any exercise you may perform. Please record the activity you performed (run, walk, cycle, aerobics, tennis, etc.), the duration of you exercise session (in minutes), and the intensity of you exercise session. For Intensity of Exercise column you will measure and record your heart-rate at 2-3 points during you exercise session by palpating your carotid artery. Your will measure heart-rate for a 10 second duration and input the average value from your 2-3 readings into the column. Leave Formula Calculation column and Total rows blank. If you have performed no exercise for the week leave that week's columns and rows blank. If you have any questions call Daisy at (604) 682-2344 Local 2680. Period WeekS Day 1-7 Activity Duration in minutes Exercise intensity heart-rate in 10 sec Fonnula Calculation L E A V E B L A N K W K 1 1 Mon run 35 30 2 Tues aerobics cycle 60 20 27 35 3 Wed 4 Thu 5 Fri golf 90 20 6 7 Sun downhill sking cross-country sking 60 90 27 33 Total WK2 1 Mon . 2 Tues aerobics swimming 60 40 26 25 3 Tues downhill sking 60 28 4 Wed downhill sking 60 27 5 Fri cycle / run / weights 15/ 25/ 45 2 5 / 3 0 / do not need heart-rate value for weights 6 7 Sun running hiking 20 75 33 22 Total WTO 1 2 Tues kite flying 120 15 3 4 Thu weights cycle 45 20 13 29 5 Thu swim 20 18 6 7 1 Total i . « i i u = u*y u i W C C K o i exercise in Day column, a you need to use more uian one row for you exercise regimen for one day you maj do so, only remember to note the day of week in the Day column. You are not expected to exercise daily, or to exercise at all. We only would like you to document the physical activities you do perform. Take one heart-rate per 15-20 minutes of continuous activity. Count the first heart-beat as zero. ID Code: 68 SHEET it: DAILY ACTIVITY PLANNER You arc nol expected to exercise daily or feci you need to for llu's study. Wc would like you to simply record any exercise you may perforin. Please record the activity you performed (run, walk, cycle, aerobics, tennis, etc.), the duration (in minutes) and Uic intensity of your exercise session. For Exercise Intensity column you will measure and record your heart-rate at 2-3 points during you exercise session by palpating your carotid artery' (once for every 15-20 min of exercise). Your will measure lican-rate for a 10 second duration and input ihc average value from your 2-3 readings into llie column. Record day of the week in Day column. Leave Formula Calculation column and Total rows blank. If you ha\<e performed no exercise for the week lea\>e that week's columns and rows blank If you have any questions call Daisy at (604) 6822344 local 2680. Period Week// Day 1-7 Activity Duration in minutes Exercise intensity heart-rate in 10 sec. Formula Calculation L E A V E BLANK WK1 1 2 3 4 5 6 7 Total * WK2 1 2 3 4 5 6 7 ^ ^ i ^ ^ ™ S M : ^ ^ p ^ ^ $ ; ; Total ID Code: 69 SHEET #: DAILY ACTIVITY PLANNER Period Week// Day 1-7 Activity Duration in minutes Exercise intensity heart-rate in 10 sec. Formula Calculation , L E A V E B L A N K WK3 1 2 3 4 5 6 7 Total WK4 1 2 3 4 5 6 7 Total 70 Reproductive history questionnaire (Developed by Canadian Multicentre Osteoporosis Study (CAMOS)) In this section I vould like to ask you questions tha: mil help us understand /tow uomen reive to bone structure. We ask evenonedUse questions^ 5 1 ' Before menopause, have you ever gone 3 months or more without a menstrual perk (not including pregnancy or dudng breastfeeding) • Yes • No Go to 5.2 What was the longest single period of time without a menstrual flow? If you count all the periods you have raissed throughout your menstruating years, how many months would.that be/ (this question asks for the cumulative time) * 2 * Have your menstrual periods stopped for more than one year? (No period one year or more after last menstruation) • Yes • No L+ At what age? :—iryears 5.3 Have you had your uterus removed (hysterectomy)! • Yes • No At what age? _ years L 5.4* Have you ever had one or both ovaries removed? • Yes, one ovary removed at what age? • . rt • Yes, both ovaries removed • at what age? — (f evades were removed oa separate occasions, write the age ctvMchthe secona O Yes, do not know how many at what age? • No * Sec note* la tninuil Respondcn i i.u. A 72 you o r d i d you ever take estrogen for menopause s r j o u m - ^ ^ • N o G o to 5.6 • Y e s , current ly • Y e s , but not now I What type(s)? (Interviewers to show Ogen*. Premarin* pills, colors and doses and Estraderm*. Estracotnb*- patches, sizes and doses) O P i l l • Patch • Injection How many times/year? How many years? Q Vaginal cream How frequently? * Sec note* In m*nu*l Respondent I.D. 8 73 Do you or did you ever take Provera* • Yes, currently • Yes, but not now I What type(s)? {Interviewers to , for menopause o r j o x i ^ ^ • No Go to 5.7 show Provera* pills, colors and doses) • Pi l l Pi l lN0 • Injection Number of days/mouth. Age started Age stopped Total number of months taken How many times/year? How many years? Have you ever used birth ;control-pills or oral-contraceptives. D Yes"" • • Ho. Z f 0 £ ^ At what age did you start? * For approximately how long did you use birth control plus7 (ifpenviLi nun. — . natural/surgical menopause) years (approximately) .years months Are you still using birth control pills? • Yes • No L-* At what age did you stop using birth control pills? years Go to 5.9 * Sec notes fa tntnuxl (If not using birth control pills, not menopausal, have not had both ovaries removed) Can you tell by the way you feel that your period is coming? • Yes, every month • Yes, most months • Yes, less than half the time • Yes, one or twice a year • Never If YES, to any of the above: What signs or symptoms indicate to you that your period is coming? • menstrual cramps or aching back or legs • bloating, fluid retention • increased appetite (in general or for sweet, salty or spicy foods) • moodiness (frustration, irritability, sadness) • breast tenderness in the front or the nipple Q breast tenderness up under the arm or on the outer sides of the breast • breast swelling • headaches (migraine or tension) • acne / pimples / blemishes : Q other : - I . .. / 5.9 How many times have you been P«gnantf If 0 : Go to 5.12 n o w many U U K » U * T " JX/- ~— * ~ . .« (pTegnaTc} confirmed by a physician or pregnancy test) 5.10 * How many of these pregnancies resulted - if 0 Go to 5.12 in at least one live birth? — j — . (Count twins and triplets as 1) Age at 1st birth? years 5.11 Did you breast feed any of your children? • Yes • No . . U For how many months total _ . m o n t h s (U. adding up the months with each child) 5.12 -How old were you when you had your first menstrual period? years 12 * Sec note* In mtntttl 75 13* a) Did you have regular • Yes • No periods once they began? ^ Go to 5.14 b) If you had irregular periods, did they become regular? • Yes • No -+ Go to 5.14 L At what age years c) Have your periods been made regular by medication? • Yes • No L At what age years 14 On average, how often did you have menstrual periods when you were in your 20's and 30's? • 20 days or less • 21-25 days • 26-30 days • 31-36 days • 37 or more days • do not know Bone fracture questionnaire (Developed by Canadian Multicentre Osteoporosis Study (CAMOS)) [ C a M o s 77 Canadian Multicentre Osteoporosis Study RESPONDENT I.D. # Etude Canadienne Mutticentrique sur fosteoporose _ . F o l l o w - u p Q u e s t i o n n a i r e This questionnaire was desired to find out how you have been ^  „•„,.„ (date of last questionnaire). Please indicate your answerb^ttingachM 1. a) Have you broken one or more bones in the past year? Q Yes Q No —>• Go to question # 3 • b) - How many times have you fractured a bone, in the last year? 2. a) . Which bone(s) were broken the last time that you had a fracture, in tie last year? Q Back Q ForearmV Wrist • Hip • • Pelvis • Q Ribs . • Other [ ; b) For the most recent incident, how did it happen? Q Fell out of bed or offa chair Q Fell climbing a chair or ladder Q Fell on stairs Q Motor vehicle accident Q Sporting injury (Le. skiing, playing hockey, cycling running or jogging, etc.) Q Slipped or tripped in home (on carpet, wet floor, getting in/out of bath, etc) Q Slipped or tripped and fell outside me home other than sporting (on ice, on the curb, etc) Q Heavy Object fell or struck body causing the fracture Q Bone(s) broke with no fall or injury • Other - Specify: ; 78 •3. :a) b) Have you had any hospital admissions in tire' past year? • Yes • No I For what reason? (Check all that apply) Q Heart disease Q Pregnancy Q Breast cancer Q Cancer of the uterus Q Other cancer (specify): Q Removal of the uterus Q Removal of ovaries • . Other surgery (specify): . Q - Other hospital admission (specify), •_ A. Have you taken anybf these medicationsm t^he pastyear? (Gzrc/e all.that apply) • Yes • No Alendronate Clodronate Estrogen Aredia Deca-Durabolin Etidronate Bonefos Didrocal Fluoric Calcimar Didronel Fosamax Calcftonin Durabolin Nandrolone CMcitriol Estracomb Ogen C.E.S. Estraderm Ostac Climacteron Estrace Ostoforte Comments: Pamidronate Premarin Progesterone Prometrium Provera Rocaltrol Tamoxifen Vivelle DATE / / Day Month Year PLEASE RETURN THE QUESTIONNAIRES IN THE POSTAGE PAID ENVELOPE PROVIDED. THANK YOU! ( 79 8 90 Back pain questionnaire ID Code: Back Discomfort and Pain Questionnaire 1. Have you experienced back pain or back discomfort more than once over the past 12 months? a) Yes b) No c) Not sure Before you proceed to the questions below, please read the definition of back pain below. Definition of Back Pain: Recurring pain in the upper, mid, or lower back of al least one year's duration basing an onset not associated widi traumatic injury such as a car accident, a significant fall etc. 2. Has die occurrence of back pain in die past 12 months been associated widi injuries incurred from a car accident, or from some otiier form of major fall during tiiis time interval? • a) Yes b) No c) Not sure 3. If you are a woman and have been pregnant, did die onset of back pain occur during pregnancy and is still present post-pregnancy? a) Yes b) No c) Not sure If you responded N O to tlte above question 2 and 3(ifyou are a female participant), please proceed to question 4 and complete the remainder of the questionnaire. If you responded YES, or N O T SURE to the above question 2 and 3 (ifyou are a female participant), you do not have to complete the remainder of the questionaire. 4. Using die body diagram on the next page please circle the area or areas of your back where you feel pain. 5. How would you rate the severity of your back pain: a) Very mild b) Mild c) Moderate d) Severe e) Very severe 6. Is the onset of your back pain associated with (CIRCLE ALL THAT APPLY): a) a certain body position, please note die main position or maximum 2 positions: b) coughing c) respiratory illness d) physical stress (i.e., activity/exercise): Note main activity (maximum 2 activities): e) 0 emotional stress odier, please note^  i i 92/ 7. Does your pain interfere with any of the following activities (CIRCLE ALL THAT APPLY): a) ability to effectively cough b) breathing c) exercise d) work, note type of work; . e) chest physiotherapy f) other, please specify: =. g) other, please specify: i 8. If for question 7 you circled chest physiotherapy, please note which specific techniques the back pain interferes in you performing them: a) percussion b) PEP, flutter c) autogenic drainage, active cycle breathing d) postural drainage 9. How often do you experience back pain. Do you experience back pain: a) daily b) weekly c) monthly d) yearly e) not sure 10. When you experience back pain, how long does an episode last? a) less than 1 hour b) more than 1 hour, but less than 24 hours c) a day (i.e., 24 hours) d) several consecutive days, but less than a week e) a week f) over a week g) over a month 11. If you have sought therapy for your back pain, we would like to know if you have sought care from any of the following specialists. If you have not sought help, please proceed to next question. (Please circle all that apply): k) Physician-back care specialist 1) Physiotherapy m) Chiropractor n) Massage therapist o) other, please specify: ^ 12. Please rate your daily activity level. Place V in appropriate box. 4-"very active Thank-you for completing the questionnaire 7.2 APPENDIX B Study data on liver transplantation patient. 95 Table 27. Data from liver transplantation patient. The patient is on total enteral nutrition. Descriptive l i i B i i i Age of CF diagnosis lillllllill] l l l l l l l l l l iifiliillllllli Brasfield %prcdFEV, BMD status 20.6 At birth Male 75.0 20.0 78.0 Osteopenic BMD T-lumbar T-femoral Lumbar (gem"2) Femur (g cm" mWSmlM Total Total BMC i i i i i i i i i i i T-compositc -2.11 -1.20 0.76 0.87 0.77 1844 -1.65 Nutritional-physical Height l i l l l i l l Weight Iflllljjljl BMI I B l f l l l Fat mass (g) Lean mass u;> 1.59 50.0 19.8 8215.0 34234.0 Nutritional-dietary Calcium intake iflfllllllt Calcium supplcmen I (mg) Total calcium (mg) Vit. D intake l i l l l l l l l supplcmen IlBllIIlI Total Vit. i i i i i i i i i i i 511.0 0 511.0 32108.0 0 32108.0 Nutrilional-dictaiy Kcal (cal) Kcal from Ensure1M DEE (cal) DER (cal) BS-APasc l l l l l l l j j l Albumin 3411.0 3411.0 2454.8 2685.8 59.0 34.0 Exercise Exercise status-adulthood Exercise status-cluldhood Subjective rating of activity level ifiiiiii:iiii (ml kg"1 min' l i i i i i l l i i i l l i i i i i l l l i l l l i l l l l l ! ! ; |||t§§l|||il (Imin1) % of prcd iHl l l l l - l IIIIIIIIIII Sedentary Active Sedentary 21.4 1.07 2.41 44.2 Kyphosis, Back pain & nonvertebral fractures Kyphosis i l l l t l i l l i T8-T12 Back pain present Nonvertebral fractures 4.59 0.96 No None Puberty Age of onset (yrs) 15.8 Steroid use Childhoo l i l l l i l l Type used Adulthoo d Type used Yes Oral Yes Oral 


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



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"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items