UBC Faculty Research and Publications

Vasopressin and terlipressin in adult vasodilatory shock: a systematic review and meta-analysis of nine… Serpa Neto, Ary; Nassar, Antônio P; Cardoso, Sérgio O; Manetta, José A; Pereira, Victor G; Espósito, Daniel C; Damasceno, Maria C; Russell, James A Aug 14, 2012

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

Item Metadata


52383-13054_2012_Article_931.pdf [ 689.15kB ]
JSON: 52383-1.0223256.json
JSON-LD: 52383-1.0223256-ld.json
RDF/XML (Pretty): 52383-1.0223256-rdf.xml
RDF/JSON: 52383-1.0223256-rdf.json
Turtle: 52383-1.0223256-turtle.txt
N-Triples: 52383-1.0223256-rdf-ntriples.txt
Original Record: 52383-1.0223256-source.json
Full Text

Full Text

RESEARCH Open AccessVasopressin and terlipressin in adult vasodilatoryshock: a systematic review and meta-analysis ofnine randomized controlled trialsAry Serpa Neto1*, Antônio P Nassar Júnior2, Sérgio O Cardoso1, José A Manetta1, Victor GM Pereira1,Daniel C Espósito1, Maria CT Damasceno1 and James A Russell3AbstractIntroduction: Catecholamines are the most used vasopressors in vasodilatory shock. However, the development ofadrenergic hyposensitivity and the subsequent loss of catecholamine pressor activity necessitate the search forother options. Our aim was to evaluate the effects of vasopressin and its analog terlipressin compared withcatecholamine infusion alone in vasodilatory shock.Methods: A systematic review and meta-analysis of publications between 1966 and 2011 was performed. TheMedline and CENTRAL databases were searched for studies on vasopressin and terlipressin in critically ill patients.The meta-analysis was limited to randomized controlled trials evaluating the use of vasopressin and/or terlipressincompared with catecholamine in adult patients with vasodilatory shock. The assessed outcomes were: overallsurvival, changes in the hemodynamic and biochemical variables, a decrease of catecholamine requirements, andadverse events.Results: Nine trials covering 998 participants were included. A meta-analysis using a fixed-effect model showed areduction in norepinephrine requirement among patients receiving terlipressin or vasopressin infusion comparedwith control (standardized mean difference, -1.58 (95% confidence interval, -1.73 to -1.44); P < 0.0001). Overall,vasopressin and terlipressin, as compared with norepinephrine, reduced mortality (relative risk (RR), 0.87 (0.77 to0.99); P = 0.04). Vasopressin compared with norepinephrine decreased mortality in adult patients (RR, 0.87 (0.76 to1.00); P = 0.05) and in patients with septic shock (42.5% vs. 49.2%, respectively; RR, 0.87 (0.75 to 1.00); P = 0.05;number needed to treat, 1 to 15). There was no difference in adverse events between the vasopressin and controlgroups (RR, 0.98 (0.65 to 1.47); P = 0.92).Conclusions: Vasopressin use in vasodilatory shock is safe, associated with reduced mortality, and facilitatesweaning of catecholamines. In patients with septic shock, use of vasopressin compared with norepinephrine mayalso decrease mortality.IntroductionThe mortality rate of patients with shock remains high[1]. Vasodilatory shock is characterized by low arterialblood pressure due to a significantly decreased systemicvascular resistance. The most frequent causes of thistype of shock are sepsis and post-cardiovascular surgeryrequiring cardiopulmonary bypass. However, massivevasodilatation can result from shock of any origin [1].Aggressive volume resuscitation is the mainstay ofinitial shock management, followed by vasoactive infu-sions when fluids do not restore adequate arterial pres-sure and tissue perfusion [2]. Currently, catecholaminesare the preferred vasopressor but the development ofadrenergic hyposensitivity with the loss of pressorresponsiveness makes finding other options necessary[3]. Additionally, catecholamines - especially dopamineand epinephrine - have significant adverse effects, such as* Correspondence: aryserpa@terra.com.br1Medical Intensive Care Unit, ABC Medical School (FMABC), Lauro GomesAvenue 1000, Santo André 09060-650, BrazilFull list of author information is available at the end of the articleSerpa Neto et al. Critical Care 2012, 16:R154http://ccforum.com/content/16/4/R154© 2012 Serpa Neto et al.; licensee BioMed Central Ltd. This is an open access article distributed under the terms of the CreativeCommons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, andreproduction in any medium, provided the original work is properly cited.decreased cardiac output and oxygen delivery, arrhyth-mia, and organ ischemia, especially at high doses.Catecholamines may even increase mortality rates [3].Vasopressin is a neurohypophyseal hormone withdiverse actions mediated by tissue-specific receptors.Low-dose vasopressin and its analog terlipressin haveemerged as promising therapies in vasodilatory shockfor several reasons. The rationale for using vasopressinand its analogs is the development of relative vasopres-sin deficiency in patients with vasodilatory shock andthe observation that exogenously administered vasopres-sin restores vascular tone, increases responsiveness toinfused catecholamines and raises blood pressure,thereby reducing the need for catecholamine use [4].Observational and randomized controlled studies invol-ving the use of vasopressin infusion in patients with vaso-dilatory shock have produced conflicting results. Our aimwas to summarize these studies using a systematic reviewof the literature and a meta-analysis of randomized con-trolled trials focused on vasopressin and its analog terli-pressin in adult patients with vasodilatory shock. We alsoevaluated vasopressin and terlipressin in studies of septicshock only.Materials and methodsSearch methods for identification of studiesStudies were identified using the Medline (1966 to 2011)and CENTRAL (1800 to 2011) databases using a sensitivesearch strategy combining medical subject headings andkeywords (see Additional file 1 for details). Abstractsfrom recent major conferences (American ThoracicSociety, European Society of Intensive Care Medicine,and Society of Critical Care Medicine) were searched foradditional relevant studies. All of the review articles andcross-referenced studies from the retrieved articles werescreened for pertinent information.Selection of studiesThis meta-analysis was limited to studies that dealt withthe role of vasopressin and/or terlipressin compared withcatecholamine infusion in the treatment of vasodilatoryshock in adult critically ill patients. Vasodilatory shockwas defined as hypotension due to peripheral vasodilata-tion as result of failure of the vascular smooth muscle toconstrict [2]. We included trials that satisfied the follow-ing inclusion criteria: randomized controlled trials com-paring adult critically ill patients who had vasodilatoryshock receiving treatment with vasopressin or terlipressincompared with patients not receiving such treatment;survival, biochemical and hemodynamic data; andpatients in the studies receiving vasoactive infusion in thecontrol arm and vasopressin plus vasoactive infusion inthe experimental arm. Studies were excluded if survivaloutcome or biochemical and/or hemodynamic data werenot provided, if pediatric patients were analyzed, or ifthey included patients without vasodilatory shock. Whenwe found duplicate reports of the same study in prelimin-ary abstracts and articles, we analyzed data from the mostcomplete dataset.Data extractionData were independently extracted from each report bythree authors, using a data-recording form developed forthis purpose. After extraction, data were reviewed andcompared by ASN. Disagreements between the twoextractors were solved by consensus among the investiga-tors. Whenever needed, additional information concern-ing a specific study was obtained by directly questioningthe principal investigator.Definition of endpointsThe primary endpoint was overall survival. The survivaltime was defined as the time from randomization untildeath from any cause, or was censored on the date of thelast follow-up assessment. Secondary endpoints included:change in hemodynamic variables; change in biochemicalvariables; decrease of catecholamine requirements; andassessment of adverse events.Statistical analysisThe effects of vasopressin and terlipressin on vasodilatoryshock outcomes and the adverse effects of these drugswere examined. We extracted data regarding the studydesign, patient characteristics, treatment duration, medi-cation, dosage, mean change for hemodynamic and bio-chemical variables, overall survival, and decreasedcatecholamine requirement. Changes in hemodynamicand biochemical variables were expressed as a percentageand were defined as the final value (after follow-up)minus the baseline value divided by the baseline value.The decreased catecholamine requirement was defined asany sustained decrease in catecholamine infusion overthe study follow-up period. Norepinephrine and dopa-mine infusions were reported as micrograms per minute,vasopressin infusion was reported as units per minute,and terlipressin infusion was reported as microgramsper hour. The conversion of weight-based variables(μ/kg/minute) to time-based variables (μg/minute orμg/hour) was made by multiplying the value by the meanweight, either provided by the author or estimated as70 kg. The conversion of bolus administration (mg) tocontinuous administration (μg/hour) was made by divid-ing the total daily dose by 24 and multiplying it by a con-version factor (mg to μg). All of the time variables weredescribed as hours, and the other variables are describedin the text.For the survival analysis and for the proportion ofpatients who experienced adverse events, a pooledSerpa Neto et al. Critical Care 2012, 16:R154http://ccforum.com/content/16/4/R154Page 2 of 10estimate of the relative risks (RRs) of the individual studieswas computed using a fixed-effect model, according toMantel and Haenszel, and these results graphically repre-sented using forest plot graphs. For continuous variables,the standardized mean difference, which consists of thedifference in means divided by the standard deviation, wasused. The homogeneity assumption was checked by a chi-squared test with the degrees of freedom equal to thenumber of analyzed studies minus one. A sensitivity analy-sis was performed by recalculating the pooled RR esti-mates for different study subgroups based on the relevantclinical features. This analysis serves to show whether theoverall result would be affected by a change in the meta-analysis selection criteria [5]. An estimate of the potentialpublication bias was carried out by plotting the single-study RR on a log scale against the respective standarderror (funnel plot). For multiple comparisons, we used theBonferroni correction method. Inter-rater reliability wasdetermined by comparing the number of studies searchedby Author 1 versus the number searched by Author 2 ineach stage of the search by the kappa coefficient [5].All variables were tested for normality using theKolgomorov-Smirnov test. The parametric variableswere described as the means and standard deviations,and the nonparametric variables were described as themedians and interquartile ranges. All of the analyseswere made using Review Manager version 5.1 (Copenha-gen: The Nordic Cochrane Center, The Cochrane Colla-boration, 2011) and Statistical Package for the SocialSciences version 16.0 (SPSS Inc., Chicago IL, USA). Forall analyses, P < 0.05 was considered significant. Forpublication bias, P < 0.1 was considered significant.ResultsLiterature searchThe search strategy retrieved 105 unique citations. Ofthese citations, 84 were excluded after the first screeningbased on the abstracts or titles, leaving 21 articles for afull-text review (Figure 1). In this review, 12 articles wereexcluded for the following reasons: outcome of interestnot described (n = 5); study design not appropriate (n =4); pediatric patients (n = 2); and evaluated vasopressin inboth arms (n = 1). Finally, nine articles (998 participants)were included in the meta-analysis [6-14]. For all of thecomparisons of inter-rater reliability in each stage of thesearch, the kappa coefficient ranged from 0.87 to 0.93.Figure 1 Literature search strategy. AVP, vasopressin.Serpa Neto et al. Critical Care 2012, 16:R154http://ccforum.com/content/16/4/R154Page 3 of 10Study characteristicsThe characteristics of the nine selected studies are shownin Table 1. Six studies evaluated vasopressin as a thera-peutic approach, two evaluated terlipressin and one studyevaluated both. In seven studies the disease responsiblefor vasodilatory shock was septic shock exclusively, inone study the shock occurred after post-left ventricularassist device, and in the last study the shock occurredpost cardiotomy. The mean dose of the drugs across thestudies was 38.72 ± 40.14 μg/minute for norepinephrine,0.055 ± 0.027 U/minute for vasopressin, and 59.03 ±47.59 μg/hour for terlipressin. The mean age was 61.91 ±6.29 in entire group, and the median follow-up time was24 (4.00 to 48.0) hours. The assessment of the studyquality is exposed in Table S1 in Additional file 2.Vasopressin and terlipressin in vasodilatory shock:hemodynamic and biochemical effectsTable S2 in Additional file 2 shows the changes in thehemodynamic variables of the vasopressin, terlipressin,and control patients in each study and in a combinedanalysis. Compared with the control group, the vasopres-sin group showed a significant reduction in heart rate(-12.8 ± 6.14% vs. -1.62 ± 6.86%; P = 0.023), and a non-significant increases in central venous pressure (+19.0 ±1.41% vs. +10.66 ± 9.29%; P = 0.083), stroke volumeindex (+14.0 ± 8.54% vs. +1.00 ± 2.94%; P = 0.05), andleft ventricular stroke work index (+65.00 ± 5.65% vs.29.0 ± 26.54%; P = 0.064). When compared with terli-pressin group, the vasopressin group showed a nonsigni-ficant smaller reduction in cardiac index (-4.00 ± 9.25%vs. -16.33 ± 4.04%; P = 0.070). Compared with the con-trol group, the terlipressin group showed a significantreduction in heart rate (-17.3 ± 7.09% vs. -1.62 ± 6.86%;P = 0.011) and oxygen delivery index (-16.0 ± 2.64% vs.-2.0 ± 6.28%; P = 0.036), and a borderline significantreduction in the oxygen consumption index (-11.33 ±6.02% vs. -1.00 ± 5.65%; P = 0.071).Table S3 in Additional file 2 shows the changes in thelaboratorial variables of the vasopressin, terlipressin, andcontrol patients in each study and in the combined analy-sis. There were no differences among the groups accord-ing to the variables evaluated. In the analyses of thestandardized mean difference, we found a significant dif-ference between the terlipressin and control groups inthe cardiac index (-0.44 (-0.87 to -0.02); P = 0.04) andoxygen delivery index (-0.79 (-1.23 to -0.36); P = 0.0004),and a tendency toward a reduction in the gastric PaCO2gap difference (-0.47 (-0.96 to 0.01); P = 0.06) (Figures S1to S5 in Additional file 3).There was a significant reduction in the norepinephr-ine requirement among the patients receiving either aterlipressin or a vasopressin infusion. In the analyses ofthe standardized mean difference, there is a significantdifference between the vasopressin and control patients(-1.56 (-1.71 to -1.41); P < 0.0001) and between theterlipressin and control patients (-1.97 (-2.62 to -1.32);P < 0.0001) with respect to the norepinephrine dosages(Figure 2).Vasopressin and terlipressin in vasodilatory shock:mortality and adverse eventsThe association of vasopressin with a decreased dose ofnorepinephrine infusion reduced mortality in adultpatients (RR, 0.87 (0.76 to 1.00); P = 0.05). The terlipres-sin infusion did not influence mortality (RR, 0.88 (0.62 to1.25); P = 0.47), and in the combined analysis terlipressin+ vasopressin resulted in a decreased mortality (RR, 0.87(0.77 to 0.99); P = 0.04) (Figure 3). With respect toadverse events, vasopressin and control patients wereanalyzed because no study that evaluated the effects ofterlipressin described adverse events (see Table S4 inAdditional file 2). There was no difference in adverseevents between the vasopressin and control groups (RR,0.98 (0.65 to 1.47); P = 0.92) (Figure S6 in Additional file3). There is no correlation between the magnitude of thereduction of norepinephrine during follow-up and therisk ratio for mortality (r = 0.143; P = 0.652).To explore the study heterogeneity, stratified analyseswere performed across a number of key study character-istics and clinical factors. The summary of the analysis isshown in Table 2, and the complete analysis is shown inTable S5 in Additional file 2. In the survival analysis, wefound a reduction in mortality with vasopressin patientswith septic shock (P = 0.05) (Figure 3). The numberneeded to treat for this condition was 1 to 15. This effecton mortality disappeared after analysis without the studyby Russell and colleagues [10]. The reduction in norepi-nephrine requirement was more significant in the dou-ble-blinded studies, in patients with septic shock, andwith smaller doses of vasopressin and terlipressin. Whenwe analyzed the data without the study by Russell andcolleagues [10], the standardized mean difference foundwas smaller but still significant (Table S5 in Additionalfile 2).When analyzing only the double-blinded studies andthe patients with septic shock, vasopressin and terlipres-sin were associated with an increase in cardiac indexcompared with norepinephrine alone. However, thesefindings were more significant in the studies with ashorter follow-up and a smaller number of patients.Higher doses of terlipressin (>40 μg/hour) were asso-ciated with decreases in cardiac index, oxygen deliveryindex, oxygen consumption index, and gastric PaCO2 gapdifference. Further, higher doses of vasopressin (>0.05 U/minute) were associated with an increased gastric PaCO2gap difference. Finally, bolus infusion of terlipressin wasassociated with a significant reduction of the oxygenSerpa Neto et al. Critical Care 2012, 16:R154http://ccforum.com/content/16/4/R154Page 4 of 10Table 1 Characteristics of the studies included in the systematic reviewStudy Arms n Age (years) Design Disease Dosage Prognostic index Time (hours) MAP objective (mmHg)Albanèse and colleagues [6] N 10 65 RCT Septic shock 119 μg/minute 29 (APACHE II) 6 65 ± 10TP 10 66 OL 50 μg/hour 28 (APACHE II) 6Dünser and colleagues [7] N 24 68 RCT Vasodilatory shock 58.8 μg/minute 49.7 (SAPS II) 48 >70N+AV 24 68 OL (PS in 40%) 0.06 U/minute 51.6 (SAPS II) 48Morelli and colleagues [8] N 15 64 RCT Septic shock 15 μg/minute 58 (SAPS II) 48 70 ± 5N+TP 15 67 OL 110.5 μg/hour 62 (SAPS II) 48N+AV 15 66 0.03 U/min 60 (SAPS II) 48Morelli and colleagues [9] N 20 67 RCT Septic shock 84 μg/minute 59 (SAPS II) 4 70 ± 5N+TP 19 66 OL (N >0.9 μg/kg/minute) 16.6 μg/hour 60 (SAPS II) 4Russell and colleagues [10] N 382 62 RCT Septic shock 15 μg/minute 27.1 (APACHE II) 672 65 ± 10N+AV 397 59 DB (N >5 μg/minute) 0.03 U/minute 27 (APACHE II) 672Argenziano and colleagues [11] N+P 5 52 RCT Vasodilatory shock 19.7 μg/minute - 0.25 >70N+AV 5 52 DB Post-LVAD 0.1 U/minute 0.25Patel and colleagues [12] N 11 68 RCT Septic shock 17 μg/minute 24 (APACHE II) 4 Physician decisionN+AV 13 68 DB (high doses of D) 0.06 U/minute 22 (APACHE II) 4Malay and colleagues [13] N+P 5 56 RCT Septic shock 12 μg/minute 26 (APACHE II) 24 >70N+AV 5 53 DB 0.04 U/minute 27 (APACHE II) 24Lauzier and colleagues [14] N 10 58 RCT OL Septic shock 28.1 μg/minute 23.5 (APACHE II) 48 >70N+AV 13 51 (<12 hours of shock) 0.09 U/minute 22. (APACHE II) 48Total C 482 62.23 ± 5.70 - - 38.72 ± 40.14 - 24 (4 to 48)AV 472 59.60 ± 7.68 0.05 ± 0.02 36 (24 to 48) -TP 44 66.33 ± 0.57 59.03 ± 47.59 6 (4 to 48)Data presented as mean ± standard deviation or median (interquartile range). AV, arginine vasopressin; C, control; D, drug that the patients were already receiving at baseline; DB, double-blind; LVAD, left ventricularassist device; MAP, mean arterial pressure; N, norepinephrine; OL, open-label; P, placebo; PS, post-cardiotomy shock; RCT, randomized controlled trial; TP, terlipressin.SerpaNetoetal.CriticalCare2012,16:R154http://ccforum.com/content/16/4/R154Page5of10delivery index and a borderline significant reductionof the cardiac index. These findings were not foundwith the continuous infusion. The Grading of Recom-mendations Assessment, Development and Evaluation(GRADE) evidence profile for the impact of vasopressinfor the treatment of vasodilatory shock from this sys-tematic review and meta-analysis of randomized con-trolled trials is shown in Table S6 in Additional file 2.Figure 2 Standardized mean difference for norepinephrine reduction between vasopressin/terlipressin and control. Vasopressin (AVP)and terlipressin (TP) was associated with a significantly standardized mean difference with control (P < 0.0001 for both comparisons). CI,confidence interval; SD, standard deviation.Figure 3 Meta-analysis of overall survival for vasopressin, norepinephrine and terlipressin combined analyses. Meta-analysis of overallsurvival for vasopressin (AVP) + norepinephrine (NE) or terlipressin (TP) + NE in vasodilatory shock, and for AVP + NE in septic shock. CI,confidence interval; M-H, Mantel and Haenszel.Serpa Neto et al. Critical Care 2012, 16:R154http://ccforum.com/content/16/4/R154Page 6 of 10Table 2 Summary of stratified analyses of pooled relative risks and standardized mean differenceStratified analysis Trials n Vasopressin P value Heterogeneity Terlipressin P value HeterogeneityMortalityDiseaseSeptic shock 4 989 0.87 (0.75 to 1.00) 0.05 0.76 0.88 (0.62 to 1.25) 0.47 0.61Nonseptic shock 2 58 0.95 (0.65 to 1.37) 0.77 0.46 - - -Norepinephrine reductionDesignDouble-blind 3 813 -1.75 (-1.91 to -1.59) <0.0001 <0.0001 - - -Open-label 5 170 -0.45 (-0.85 to -0.06) 0.03 0.70 -1.97 (-2.62 to -1.32) <0.0001 <0.0001DiseaseSeptic shock 6 925 -1.64 (-1.79 to -1.48) <0.0001 <0.0001 -1.97 (-2.62 to -1.32) <0.0001 <0.0001Nonseptic shock 2 48 -0.66 (-1.22 to -0.11) 0.02 0.03 - - -Vasopressin dosage≤0.05 U/minute 2 809 -1.66 (-1.83 to -1.50) <0.0001 0.005 - - ->0.05 U/minute 4 105 -0.77 (-1.22 to -0.32) 0.0008 <0.0001 - - -Terlipressin dosage≤40 μg/hour 1 39 - - - -4.63 (-5.88 to -3.38) <0.0001 ->40 μg/hour 1 30 - - - -0.98 (-1.74 to -0.22) 0.01 -Cardiac indexDesignDouble-blind 3 46 0.94 (0.29 to 1.58) 0.004 0.45 - - -Open-label 6 190 0.03 (-0.36 to 0.42) 0.89 0.80 -0.44 (-0.87 to -0.02) 0.04 0.54DiseaseSeptic shock 7 176 0.44 (-0.00 to 0.88) 0.05 0.10 -0.44 (-0.87 to -0.02) 0.04 0.54Nonseptic shock 2 58 0.05 (-0.47 to 0.56) 0.86 0.70 - - -Terlipressin dosage≤40 μg/hour 1 39 - - - -0.29 (-0.92 to 0.35) 0.38 ->40 μg/hour 2 50 - - - -0.57 (-1.14 to -0.00) 0.05 0.37Follow-up≤24 hours 5 103 0.94 (0.29 to 1.58) 0.004 0.45 -0.48 (-1.00 to 0.04) 0.07 0.28>24 hours 4 131 0.03 (-0.36 to 0.42) 0.89 0.80 -0.37 (-1.09 to 0.35) 0.31 -Patients≤25 5 87 0.52 (0.02 to 1.03) 0.04 0.12 -0.91 (-1.84 to 0.02) 0.06 -25 to 50 4 147 0.08 (-0.36 to 0.53) 0.72 0.65 -0.32 (-0.80 to 0.15) 0.18 0.86Oxygen deliveryTerlipressin dosage≤40 μg/hour 1 39 - - - -0.75 (-1.40 to -0.10) 0.02 ->40 μg/hour 2 50 - - - -0.83 (-1.42 to -0.25) 0.005 0.38Oxygen consumptionTerlipressin dosage≤40 μg/hour 1 39 - - - -0.13 (-0.76 to 0.50) 0.69 ->40 μg/hour 2 50 - - - -0.52 (-1.08 to 0.05) 0.07 0.64Gastric PaCO2 gapVasopressin dosage≤0.05 U/minute 1 30 -0.38 (-1.11 to 0.34) 0.30 - - - ->0.05 U/minute 3 95 0.43 (0.01 to 0.84) 0.04 0.09 - - -Terlipressin dosage≤40 μg/hour 1 39 - - - -0.15 (-0.78 to 0.48) 0.64 ->40 μg/hour 1 30 - - - -0.94 (-1.70 to -0.18) 0.02 -Data presented as relative risks (95% confidence interval) for mortality; and as the standardized mean difference (95% confidence interval for cardiac index,norepinephrine reduction, oxygen delivery, oxygen consumption, and gastric PaCO2 gap.Serpa Neto et al. Critical Care 2012, 16:R154http://ccforum.com/content/16/4/R154Page 7 of 10DiscussionThis systematic review suggests that the combination ofvasopressin with norepinephrine in vasodilatory shock issafe, is associated with a reduction in patient mortality,and facilitates weaning of catecholamines, avoiding thelatter’s potential adverse events. Vasopressin and terli-pressin did not decrease the cardiac or oxygen deliveryindices. All of these changes are more significant inpatients with septic shock. The stratified analysis ofvasopressin combined with norepinephrine in patientswith septic shock was associated with a borderline sig-nificant increase in hospital survival when comparedwith norepinephrine alone.Vasodilatory shock pathogenesis is multifactorial.Increased nitric oxide, consequent to the activation ofinducible nitric oxide synthase, is a major contributor tovasodilatation, acting both directly and via cyclic guano-sine monophosphate to lower intracellular calcium levels,to decrease myosin light chain phosphorylation, and toactivate calcium-sensitive (KCa) and adenosine tripho-sphate-sensitive (KATP) K+ channels [1,15]. An increasingnumber of studies have consistently found that patientswho have vasodilatory shock and are evaluated in theICU setting have very low plasma levels of vasopressin[16]. Hence, a relative deficiency of vasopressin may alsobe crucial to the altered functional status of vascularsmooth muscle.The clinical use of vasopressin has followed observationssuggesting that exogenous administration of vasopressinduring shock increases systemic blood pressure [17].Blood pressure restoration does not necessarily improveoutcome, however, if the increased blood pressure isaccompanied by a worsening of cardiac performance andby decreased cardiac output and oxygen delivery [18,19].High doses of vasopressin are associated with decreasedcardiac output, oxygen delivery and consumption, andwith increased gastric PaCO2 gap difference [20]. High-dose vasopressin is thus not indicated as an alternative toother vasopressors for the treatment of vasodilatory shock.One important limitation to address is that the effectsdetected by our paper might be related to the decreasingdoses of catecholamines instead of the fixed dose of addi-tional vasopressor (although we did not find a relation-ship between reduction of norepinephrine and mortality).There is a current trend for decatecholaminization ofpatients, because catecholamines have deleterious effectson immune function, thrombogenicity and metabolicefficiency, on stimulating bacterial growth, and on caus-ing myocardial injury [21]. In fact, higher mortality wasnoted in those patients where higher mean blood pres-sure values were generated using a progressively highercatecholamine dose [21].We also found that vasopressin significantly reducedthe heart rate in patients with vasodilatory shockwithout changes in cardiac output. This is an importantfinding because it could prevent the development and/or progression of the myocardial dysfunction associatedwith septic shock and the tachycardia-induced cardio-myopathy [22]. Recent studies also suggest that diastolicdysfunction is a common finding in patients with septicshock and is a major predictor of mortality in thesepatients [23]. As we already know, reducing the heartrate in patients with diastolic dysfunction can achieveadequate ventricular filling.We found differences between the regimen used foradministration of terlipressin (bolus versus continuous).Previous studies have suggested that intermittent bolusinjection may be associated with significant adverse effects,including excessive microregional and systemic vasocon-striction, as well as decreases in cardiac output and oxygendelivery [24]. Conversely, recent studies provide evidencethat continuous infusion of low-dose terlipressin exertsbeneficial hemodynamic effects with reduced side effectsas compared with traditional bolus injection. In general,continuous low-dose terlipressin was associated withimproved parameters of myocardial contractility and renalfunction as well as less vascular leakage compared withbolus injection [24].The largest randomized controlled trial of vasopressininfusion in septic shock showed no benefit of vasopressinversus norepinephrine [10]. However, this study hassome interesting points to be discussed. First, it wasfound in the vasopressin-treated patients with less severeshock that the 25.8% relative reduction in 28-day mortal-ity compared with norepinephrine was both striking andsignificant. This result is consistent with the evidence ofa better synergistic effect of low-dose vasopressin inisolated arteries with conditions mimicking less severeseptic shock [25]. Secondly, the dose of vasopressin cho-sen is lower than that generally used. Thirdly, the mortal-ity in the control group is lower than the 60% anticipatedin the sample size calculation, which could make thestudy underpowered to detect a significant difference inthe outcome. Finally, the mean norepinephrine dose atrandomization was considered too low by some studies[26]. However, in our stratified analysis no relationshipbetween the baseline dose of norepinephrine and out-come was found (data not shown).Polito and colleagues recently published a meta-analy-sis of vasopressin in vasodilatory shock with importantdifferences from our study [27]. First, we analyzed onlyadult patients because pediatric shock has a much lowermortality than adult shock and this may contaminate theoverall results. Also, we performed a more robust electro-nic search of references that resulted in the addiction ofone study [11] that was missed by Polito and colleagues.Finally, we evaluated a higher number of variables, con-ducted an analysis based on the standardized meanSerpa Neto et al. Critical Care 2012, 16:R154http://ccforum.com/content/16/4/R154Page 8 of 10difference of reduction in norepinephrine requirement,and performed a more robust sensitivity analysis.These reported findings should be viewed within thecontext of the limitations of this study and research inthe field. Although our literature search procedureswere extensive, other trials may have appeared or maynot have been published, and publication bias is there-fore possible, which could overestimate the efficacy ofthese treatments. The assessment of adverse effects waslimited to those studies in which adverse events wereexplicitly reported. The changes in hemodynamic andbiochemical variables were calculated, and the standarddeviations were probably biased because the correlationwithin patients is not reported. In addition, another lim-itation of our meta-analysis is that it is dominated bythe largest randomized controlled trial from Russell andcolleagues [10]. Finally, assumptions used to calculatethe drug dosage may have influenced the results.ConclusionsThe present meta-analysis has demonstrated the benefitof the association of vasopressin and terlipressin inreducing norepinephrine requirements in patients withvasodilatory shock, particularly in patients with septicshock. Our results show that vasopressin treatment isnot associated with decreased cardiac output or oxygendelivery and consumption, even in higher doses. How-ever, the pooled analyses showed that higher doses ofterlipressin were associated with worsening of thesevariables. Vasopressin significantly reduces mortality ingeneral patients, and specifically in patients with septicshock.Key messages• Vasopressin and terlipressin reduce the norepinephr-ine requirements in patients with vasodilatory shock.• Vasopressin is not associated with decreased cardiacoutput or oxygen delivery and consumption.• Vasopressin significantly reduces mortality inpatients with septic shock.Additional materialAdditional file 1: Detailed search methods for identification ofstudies.Additional file 2: Table S1 presenting the assessment of studyqualities. Table S2 presenting the change in hemodynamic variables((final - baseline value/baseline value) × 100%). Table S3 presenting thechange in biochemical variables ((final - baseline value/baseline value) ×100%). Table S4 presenting adverse events. Table S5 presenting stratifiedanalyses of pooled relative risks and standardized mean difference. TableS6 presenting Grading of Recommendations Assessment, Developmentand Evaluation (GRADE) evidence profile for impact of vasopressin orterlipressin for vasodilatory shock from systematic review and meta-analysis of randomized controlled trials.Additional file 3: Figure S1 showing the standardized meandifference of the cardiac index. Figure S2 showing the standardizedmean difference of the oxygen delivery index (DO2i). Figure S3 showingthe standardized mean difference of the oxygen consumption index(VO2i). Figure S4 showing the standardized mean difference of arteriallactate. Figure S5 showing the standardized mean difference of thegastric PaCO2 gap (Pr-aCO2). Figure S6 showing risk for adverse events.AbbreviationsICU: intensive care unit; PaCO2: partial pressure of carbon dioxide; RR: relativerisk.Author details1Medical Intensive Care Unit, ABC Medical School (FMABC), Lauro GomesAvenue 1000, Santo André 09060-650, Brazil. 2Department of Intensive CareMedicine, São Camilo Hospital, Pompeia Avenue 1178, São Paulo 05024-000,Brazil. 3Heart and Lung Institute, Division of Critical Care Medicine, St Paul’sHospital, 1081 Burrard Street, Vancouver, BC V6Z 1Y6, Canada.Authors’ contributionsASN participated in the concept and design of the study, data acquisition,statistical analysis and interpretation, drafted the manuscript, and revised themanuscript for important intellectual content. APNJr participated in the datainterpretation, drafted the manuscript, and revised the manuscript forimportant intellectual content. SOC participated in the data acquisition andinterpretation, drafted the manuscript, and revised the manuscript forimportant intellectual content. JAM participated in the data interpretation,drafted the manuscript, and revised the manuscript for important intellectualcontent. VGMP participated in the data acquisition, and interpretation,drafted the manuscript, and revised the manuscript for important intellectualcontent.DCE participated in the concept and design of the study, data acquisition,and interpretation, drafted the manuscript, and revised the manuscript forimportant intellectual content. MCTD participated in the data interpretation,drafted the manuscript, and revised the manuscript for important intellectualcontent. JAR participated in the data interpretation, drafted the manuscript,and revised the manuscript for important intellectual content. All authorsread and approved the final version of the manuscript.Competing interestsJAR reports receiving consulting fees from Ferring Pharmaceuticals. Therewas no financial support for the manuscript. JAR reports holding stock inthe University of British Columbia, which has submitted a patent related tothe use of vasopressin in septic shock. JAR reports holding stock in SiriusGenomics Inc., which has submitted patents owned by the University ofBritish Columbia and licensed to Sirius Genomics that are related to thegenetics of sepsis and its treatment. JAR reports receiving consulting feesfrom Astra Zeneca, from BioCritica, and from Sirius Genomics Inc. JARreports receiving grant support from Sirius Genomics, FerringPharmaceuticals, Astra Zeneca and Eli Lilly. JAR has received speakinghonoraria from Pfizer and Eli Lilly. The remaining authors declare that theyhave no competing interests.Received: 29 April 2012 Revised: 26 May 2012Accepted: 14 August 2012 Published: 14 August 2012References1. Landry DW, Oliver JA: The pathogenesis of vasodilatory shock. N Engl JMed 2001, 345:588-595.2. Dellinger RP, Levy MM, Carlet JM, Bion J, Parker MM, Jaeschke R, Reinhart K,Angus DC, Brun-Buisson C, Beale R, Calandra T, Dhainaut JF, Gerlach H,Harvey M, Marini JJ, Marshall J, Ranieri M, Ramsay G, Sevransky J,Thompson BT, Townsend S, Vender JS, Zimmerman JL, Vincent JL,International Surviving Sepsis Campaign Guidelines Committee; AmericanAssociation of Critical-Care Nurses; American College of Chest Physicians;American College of Emergency Physicians; Canadian Critical Care Society;European Society of Clinical Microbiology and Infectious Diseases, et al:Serpa Neto et al. Critical Care 2012, 16:R154http://ccforum.com/content/16/4/R154Page 9 of 10Surviving Sepsis Campaign: international guidelines for management ofsevere sepsis and septic shock: 2008. Crit Care Med 2008, 36:296-327.3. Mullner M, Urbanek B, Havel C, Losert H, Waechter F, Gamper G:Vasopressors for shock. Cochrane Database Syst Rev 2004, 3:CD003709.4. Holmes CL, Walley KR: Vasopressin in the ICU. Curr Opin Crit Care 2004,10:442-448.5. Sutton AJ, Higgins JP: Recent developments in meta-analysis. Stat Med2008, 27:625-650.6. Albanèse J, Leone M, Delmas A, Martin C: Terlipressin or norepinephrinein hyperdynamic septic shock: a prospective, randomized study. Crit CareMed 2005, 33:1897-1902.7. Dünser MW, Mayr AJ, Ulmer H, Knotzer H, Sumann G, Pajk W,Friesenecker B, Hasibeder WR: Arginine vasopressin in advancedvasodilatory shock. A prospective, randomized, controlled study.Circulation 2003, 107:2313-2319.8. Morelli A, Ertmer C, Rehberg S, Lange M, Orecchioni A, Cecchini V,Bachetoni A, D’Alessandro M, Van Aken H, Pietropaoli P, Westphal M:Continuous terlipressin versus vasopressin infusion in septic shock(TERLIVAP): a randomized, controlled pilot study. Crit Care 2009, 13:R130-R143.9. Morelli A, Ertmer C, Lange M, Dünser M, Rehberg S, Van Aken H,Pietropaoli P, Westphal M: Effects of short-term simultaneous infusion ofdobutamine and terlipressin in patients with septic shock: theDOBUPRESS study. Br J Anaesth 2008, 100:494-503.10. Russell JA, Walley KR, Singer J, Gordon AC, Hébert PC, Cooper DJ,Holmes CL, Mehta S, Granton JT, Storms MM, Cook DJ, Presneill JJ, Ayers D,VASST Investigators: Vasopressin versus norepinephrine infusion inpatients with septic shock. N Engl J Med 2008, 358:877-887.11. Argenziano M, Choudhri AF, Oz MC, Rose EA, Smith CR, Landry DW: Aprospective randomized trial of arginine vasopressin in the treatment ofvasodilatory shock after left ventricular assist device placement.Circulation 1997, 96:286-290.12. Patel BM, Chittock DR, Russell JA, Walley KR: Beneficial effects of short-term vasopressin infusion during severe septic shock. Anesthesiology2002, 96:576-582.13. Malay MB, Ashton RC, Landry DW, Townsend RN: Low-dose vasopressin inthe treatment of vasodilatory septic shock. J Trauma 1999, 47:699-703.14. Lauzier F, Lévy B, Lamarre B, Lesur O: Vasopressin or norepinephrine inearly hyperdynamic septic shock: a randomized clinical trial. IntensiveCare Med 2006, 32:1782-1789.15. Barrett LK, Singer M, Clapp LH: Vasopressin: mechanisms of action on thevasculature in health and in septic shock. Crit Care Med 2007, 35:33-40.16. Russell JA: Vasopressin in vasodilatory and septic shock. Curr Opin CritCare 2007, 13:383-391.17. Landry DW, Levin HR, Gallant EM, Seo S, D’Alessandro D, Oz MC, Oliver JA:Vasopressin pressor hypersensitivity in vasodilatory septic shock. CritCare Med 1997, 25:1279-1282.18. Avontuur JA, Tutein Nolthenius RP, Buijk SL, Kanhai KJ, Bruining HA: Effectof L-NAME, an inhibitor of nitric oxide synthesis, on cardiopulmonaryfunction in human septic shock. Chest 1998, 113:1640-1646.19. Cobb JP: Use of nitric oxide synthase inhibitors to treat septic shock: thelight has changed from yellow to red. Crit Care Med 1999, 27:855-856.20. Klinzing S, Simon M, Reinhart K, Bredle DL, Meier-Hellmann A: High-dosevasopressin is not superior to norepinephrine in septic shock. Crit CareMed 2003, 31:2646-2650.21. Singer M, Matthay MA: Clinical review: Thinking outside the box - aniconoclastic view of current practice. Crit Care 2011, 15:225-234.22. Richard C: Stress-related cardiomyopathies. Ann Intensive Care 2011,1:39-46.23. Landesberg G, Gilon D, Meroz Y, Georgieva M, Levin PD, Goodman S,Avidan A, Beeri R, Weissman C, Jaffe AS, Sprung CL: Diastolic dysfunctionand mortality in severe sepsis and septic shock. Eur Heart J 2012,33:895-903.24. Lange M, Ertmer C, Rehberg S, Morelli A, Köhler G, Kampmeier TG, VanAken H, Westphal M: Effects of two different dosing regimens ofterlipressin on organ functions in ovine endotoxemia. Inflamm Res 2011,60:429-437.25. Leone M, Boyle WA: Decreased vasopressin responsiveness invasodilatory septic shock-like conditions. Crit Care Med 2006,34:1126-1130.26. Dünser MW, Hasibeder WR, Wenzel V: Vasopressin in septic shock. N Engl JMed 2008, 358:2736.27. Polito A, Parisini E, Ricci Z, Picardo S, Annane D: Vasopressin for treatmentof vasodilatory shock: an ESICM systematic review and meta-analysis.Intensive Care Med 2012, 38:9-19.doi:10.1186/cc11469Cite this article as: Serpa Neto et al.: Vasopressin and terlipressin inadult vasodilatory shock: a systematic review and meta-analysis of ninerandomized controlled trials. Critical Care 2012 16:R154.Submit your next manuscript to BioMed Centraland take full advantage of: • Convenient online submission• Thorough peer review• No space constraints or color figure charges• Immediate publication on acceptance• Inclusion in PubMed, CAS, Scopus and Google Scholar• Research which is freely available for redistributionSubmit your manuscript at www.biomedcentral.com/submitSerpa Neto et al. Critical Care 2012, 16:R154http://ccforum.com/content/16/4/R154Page 10 of 10


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