Copeptin and Arginine Vasopressin Concentrations in Critically Ill ...

0021-972X/06/$15.00/0 Printed in U.S.A.

The Journal of Clinical Endocrinology & Metabolism 91(11):4381– 4386 Copyright © 2006 by The Endocrine Society doi: 10.1210/jc.2005-2830

Copeptin and Arginine Vasopressin Concentrations in Critically Ill Patients Stefan Jochberger, Nils G. Morgenthaler, Viktoria D. Mayr, Gu¨nter Luckner, Volker Wenzel, Hanno Ulmer, Siegfried Schwarz, Walter R. Hasibeder, Barbara E. Friesenecker, and Martin W. Du¨nser Department of Anesthesiology (S.J., V.D.M., G.L., V.W., B.E.F.), Institute of Medical Biostatistics (H.U.), and Division of Experimental Pathophysiology and Immunology, Biocenter (S.S.), Innsbruck Medical University, A-6020 Innsbruck, Austria; Department of Research (N.G.M.), B.R.A.H.M.S. Aktiengesellschaft D-16761, Hennigsdorf, Germany; Department of Anesthesiology and Critical Care Medicine (W.R.H.), Krankenhaus der Barmherzigen Schwestern, A-4910 Ried im Innkreis, Austria; and Department of Intensive Care Medicine (M.W.D.), University Hospital of Bern, CH-3010 Bern, Switzerland Context: Determination of arginine vasopressin (AVP) concentrations may be helpful to guide therapy in critically ill patients. A new assay analyzing copeptin, a stable peptide derived from the AVP precursor, has been introduced. Objective: Our objective was to determine plasma copeptin concentrations. Design: We conducted a post hoc analysis of plasma samples and data from a prospective study. Setting: The setting was a 12-bed general and surgical intensive care unit (ICU) in a tertiary university teaching hospital. Patients: Our subjects were 70 healthy volunteers and 157 ICU patients with sepsis, with systemic inflammatory response syndrome (SIRS), and after cardiac surgery. Interventions: There were no interventions. Main Outcome Measures: Copeptin plasma concentrations, demographic data, AVP plasma concentrations, and a multiple organ

A

RGININE VASOPRESSIN (AVP) is a peptide hormone produced in the paraventricular nuclei of the hypothalamus and stored in neurosecretory vesicles of the posterior part of the pituitary gland. Released into the systemic circulation upon osmotic and hemodynamic stimuli, AVP exerts antidiuretic and strong vasopressor effects (1). As a matter of fact, AVP is one of the most important physiological stress and shock hormones (2). Accordingly, high AVP concentrations have been measured in different shock states in animals and humans (1– 4). Because comparably low AVP plasma concentrations were observed in septic shock patients, relative AVP deficiency has been hypothesized to contribute to the loss of vascular tone in vasodilatory shock patients (5). We and others have shown that exogenous infusion of AVP could reverse even

First Published Online August 29, 2006 Abbreviations: AVP, Arginine vasopressin; CI, confidence interval; ICU, intensive care unit; SIRS, systemic inflammatory response syndrome. JCEM is published monthly by The Endocrine Society (http://www. endo-society.org), the foremost professional society serving the endocrine community.

dysfunction syndrome score were documented 24 h after ICU admission. Results: AVP (P ⬍ 0.001) and copeptin (P ⬍ 0.001) concentrations were significantly higher in ICU patients than in controls. Patients after cardiac surgery had higher AVP (P ⫽ 0.003) and copeptin (P ⫽ 0.003) concentrations than patients with sepsis or SIRS. Independent of critical illness, copeptin and AVP correlated highly significantly with each other. Critically ill patients with sepsis and SIRS exhibited a significantly higher ratio of copeptin/AVP plasma concentrations than patients after cardiac surgery (P ⫽ 0.012). The American Society of Anesthesiologists’ classification (P ⫽ 0.046) and C-reactive protein concentrations (P ⫽ 0.006) were significantly correlated with the copeptin/AVP ratio. Conclusions: Plasma concentrations of copeptin and AVP in healthy volunteers and critically ill patients correlate significantly with each other. The ratio of copeptin/AVP plasma concentrations is increased in patients with sepsis and SIRS, suggesting that copeptin may overestimate AVP plasma concentrations in these patients. (J Clin Endocrinol Metab 91: 4381– 4386, 2006)

advanced vasodilatory shock states unresponsive to standard catecholamine therapy (1, 2, 6). This observation has led to the hypothesis that substitution of low AVP plasma concentrations to levels observed in acute shock may restore cardiovascular stability (7). If this concept proves beneficial, it would be rational to use AVP plasma levels as a tool to guide AVP therapy. Furthermore, determination of AVP plasma concentrations would be helpful to diagnose pathologies of the vasopressinergic system in patients with perioperative hypotension and chronic angiotensin converting enzyme-inhibitor therapy (8), prolonged vasopressor drug requirements after severe disease, or in patients with neurological disease and electrolyte disorders, such as the syndrome of inadequate antiduiretic hormone secretion, diabetes insipidus, or cerebral salt waste syndrome (9, 10). However, currently serious concerns exist about the methodological reliability of laboratory assays determining plasma AVP concentrations. This is largely due to the fact that ex vivo AVP is unstable, largely attached to platelets, and rapidly cleared even at temperatures as low as ⫺20 C (11, 12). Although AVP cannot be sensitively detected by sandwich immunoassays, the alter-

4381

4382

J Clin Endocrinol Metab, November 2006, 91(11):4381– 4386

native RIAs require extensive and time-consuming preanalytical procedures, including radio labeling of AVP and peptide extraction of the samples (11, 13). Thus, despite its important physiological role, routine measurements of AVP are currently not available to answer urgent clinical questions in the intensive care setting. Recently, a new sandwich immunoluminometric assay for copeptin has been introduced (14, 15). Copeptin is a 39-amino acid glycopeptide of currently unknown physiological function. Together with neurophysin II and AVP, it makes up the pre-pro-vasopressin molecule (Fig. 1) (16, 17). In contrast to AVP, copeptin is highly stable ex vivo, even for several days at room temperature, and has recently been suggested to be a suitable indirect parameter to assess AVP plasma concentrations in patients with septic shock (14, 15). The present study evaluates the correlation between copeptin and AVP plasma concentrations in healthy volunteers and critically ill patients with sepsis, with systemic inflammatory response syndrome (SIRS), and after cardiac surgery, respectively. Patients and Methods This study is a post hoc analysis of plasma samples and data from a recent prospective trial examining AVP plasma concentrations in a mixed critically ill patient population (18). The study was approved by the institutional review board and the ethics committee of the Innsbruck Medical University. The original trial was performed from March 2003 until January 2004 in a 12-bed general and surgical intensive care unit (ICU) in a tertiary university teaching hospital. Written informed consent was obtained from the next relatives of all patients eligible for study enrollment.

Patients The control group of healthy volunteers consisted of blood donors. After written informed consent was obtained from all subjects, 3 ml of EDTA blood was sampled from each blood donor, and sex, age, and arterial blood pressure were documented. Patients from the original analysis with the diagnosis of either sepsis (19) or SIRS (19) or patients after cardiac surgery were included into the protocol. Patients were excluded if they were less than 19 yr old, pregnant, suffering from central nervous system pathology, discharged or had died within 24 h after admission to the ICU, or already receiving an exogenous AVP infusion.

Data collection in critically ill patients Twenty-four hours after admission to the ICU, 3 ml of EDTA blood was sampled for determination of AVP plasma concentrations. The blood was immediately centrifuged in the ICU, and the supernatant EDTA plasma samples were frozen at ⫺80 C. At the same time, a multiple organ dysfunction syndrome score (20) was documented. Moreover, demographic data, length of ICU stay, as well as patient outcome at discharge from the ICU were reported in every study patient.

Determination of AVP plasma concentrations After completion of patient recruitment, frozen EDTA plasma samples were transferred to the endocrinology diagnostic laboratory. For measurement of AVP, 0.5 ml EDTA plasma was extracted with 2 ml

FIG. 1. Structure of pre-pro-vasopressin.

Jochberger et al. • Copeptin and Vasopressin in Ill Patients

ethanol and centrifuged, the plasma fraction was frozen in liquid N2, and the ethanol fraction was evaporated. The dry residue was then reconstituted in 0.5 ml assay buffer, of which 0.3 ml extract was assayed using a RIA (DRG Diagnostics, Marburg, Germany) (21). The AVP assay standard calibration curve ranged from 0.5 to 60 pmol/liter, with a detection limit of 0.5 pmol/liter. In case of test results lying significantly out of the clinically expected range, tests were repeated and results reconfirmed. All test results less than or equal to 0.5 pmol/liter (lower limit of the standard calibration curve of the AVP assay) were reimputed based on the age and sex of the covariate, using a multiple imputation procedure (SOLAS 02.10, version 1999; Statistical Solutions Ltd., Saugus, MA) to avoid inclusion of potentially nonquantitative test results (applicable only in the healthy control group).

Determination of copeptin plasma concentrations From the original EDTA plasma samples, approximately 250 –500 ␮l was left over and subsequently used for determination of copeptin plasma concentrations. Samples were measured in a blinded way in an external laboratory in Berlin, Germany. The sandwich immunoluminometric assay used to evaluate copeptin levels has recently been described in detail (14). Briefly, 50 ␮l of EDTA plasma was incubated with antibodies diluted in 200 ␮l of standard assay buffer under agitation (170 – 300 rpm) at room temperature (18 –24 C) for 2 h. The polyclonal antibodies used were directed against the amino acid sequence 132–164 of pre-pro-vasopressin. The test tubes were then washed four times with 1 ml LUMI test wash solution (B.R.A.H.M.S. AG; Hennigsdorf, Germany), and bound chemoluminescence was measured for 1 sec per tube with a LB952T luminometer (Berthold; Wildbad, Germany). The analytical detection limit of the assay is 1.7 pmol/liter; its interlaboratory coefficient of variation is less than 20% for values higher than 2.25 pmol/liter.

Study end points The primary end point of the study was to evaluate copeptin plasma concentrations and the correlation between copeptin and AVP plasma concentrations in healthy volunteers and critically ill patients with sepsis, with SIRS, or after cardiac surgery, respectively. The secondary end point of this study was to evaluate possible differences in the correlation of copeptin and AVP plasma concentrations between critically ill patients with sepsis, with SIRS, and after cardiac surgery.

Statistical analysis For statistical analysis, the SPSS software program (version 12.0.1; SPSS Inc., Chicago, IL) was used. One-sample Kolmogorov-Smirnov tests were performed separately for different subpopulations to check for normal distribution of study variables. Normality assumption was approximately fulfilled in all variables except for copeptin and AVP plasma concentrations. However, after ln-transformation, both variables were approximately normally distributed, with homogeneity of variances (Levene’s test) between subgroups, and thus fulfilled requirements for the application of parametric statistical methods. To determine associations between copeptin and AVP plasma concentrations in healthy volunteers and critically ill patients, a correlation and linear regression analysis was performed. To examine differences in the correlation of the two peptides among critically ill patients, a ratio of copeptin/AVP plasma concentrations was calculated. Differences in mean plasma AVP and copeptin concentrations between healthy volunteers and subgroups of critically ill patients, as well as the ratio of copeptin/AVP plasma concentrations between subgroups of critically ill patients were analyzed using a one-way ANOVA. Because of multiple testing between subgroups of critically ill patients, a Bonferroni correction was applied for the comparisons of the one-way ANOVA. Therefore, P values less than 0.017 were considered to indicate statistical significance. To search for possible causative factors for differences in the means of copeptin/AVP ratios in critically ill patients, a linear stepwise regression model was established. Predictors were selected from the following variables, which were available from the original data set: demographic (age, sex, body mass index, preexistent diseases), clinical [American Society of Anesthesiologists’ classification as a dichotomized



indicator variable (categories 1–3 vs. 4 and 5), preoperative ejection fraction, therapeutic intervention severity and simplified acute physiology score within 24 h after ICU admission (22), multiple organ dysfunction syndrome score count, length of ICU stay, ICU outcome], hemodynamic (heart rate; systolic arterial blood pressure; mean arterial blood pressure; central venous pressure; mean pulmonary arterial pressure; pulmonary capillary wedge pressure; cardiac index; stroke volume index; systemic vascular resistance index; as well as norepinephrine, milrinone, and epinephrine requirements at the time of blood sampling), and laboratory data (platelet count, plasma concentrations of sodium, creatinine, troponin I, C-reactive protein, and arterial lactate at the time of blood sampling). Based on univariate comparisons and a stepwise variable selection procedure based on the probability of the F-statistics only, covariates that showed a significant effect on the AVP/copeptin ratio were included into the final model to exclude interactions of covariates. All data are given as mean values ⫾ sd, if not indicated otherwise.

Results

The study protocol included 157 critically ill patients (sepsis, n ⫽ 25; SIRS, n ⫽ 36; patients after cardiac surgery, n ⫽ 96), as well as 70 healthy volunteers. Test results of eight healthy volunteers were excluded because of test results at or below the lower detection limit of the AVP assay. Table 1 displays demographic and clinical data of the respective study groups. Patients with SIRS (hemorrhagic shock, n ⫽ 4; acute respiratory distress syndrome, n ⫽ 6; after abdominal surgery, n ⫽ 19; after cardiopulmonary resuscitation, n ⫽ 4; acute pancreatitis, n ⫽ 2; acute renal failure, n ⫽ 1) were significantly younger than patients with sepsis and after cardiac surgery (P ⫽ 0.001). Patients with sepsis had a significantly higher degree of multiple organ dysfunction than patients with SIRS or after cardiac surgery (P ⬍ 0.001). There were significant differences in mean copeptin and AVP plasma concentrations among groups. Mean copeptin plasma concentrations were significantly higher in critically ill patients (87 ⫾ 77 pmol/liter) when compared with healthy controls (6 ⫾ 3 pmol/liter) (P ⬍ 0.001). Copeptin plasma concentrations in patients after cardiac surgery (101 ⫾ 82 pmol/liter) were significantly higher than in patients with sepsis (52 ⫾ 30 pmol/liter) and SIRS (88 ⫾ 89 pmol/liter) (P ⫽ 0.003). Similarly, AVP concentrations in critically ill patients (12.8 ⫾ 22 pmol/liter) were significantly higher than in healthy controls (1.1 ⫾ 0.4 pmol/liter) (P ⬍ 0.001); patients after cardiac surgery (17.8 ⫾ 27.9 pmol/liter) had higher AVP plasma concentrations than patients with sepsis (5.2 ⫾ 3.5 pmol/liter) and SIRS (8.1 ⫾ 8.3 pmol/liter) (P ⫽ 0.003). Independent of the presence of critical illness or the diagnoses of sepsis, of SIRS, or status after cardiac surgery,

4383

plasma concentrations of copeptin and AVP were significantly correlated with each other (Fig. 2). A ratio of copeptin/AVP plasma concentrations of 5.9 ⫾ 2.9 [confidence interval (CI) 95%, 5.2– 6.5] was observed in healthy volunteers. Critically ill patients with sepsis (12 ⫾ 7.4; CI 95%, 9.6 –14.4) and with SIRS (16.7 ⫾ 21.4; CI 95%, 9.4 –23.9) exhibited a significantly higher ratio of copeptin/ AVP plasma concentrations than patients after cardiac surgery (9.5 ⫾ 6.2; CI 95%, 8.2–10.8; P ⫽ 0.012). Out of all demographic, clinical, hemodynamic, and laboratory parameters, only the American Society of Anesthesiologists’ classification (categories 1–3 vs. 4 and 5; standardized regression coefficient, 0.176; P ⫽ 0.046) and plasma concentrations of the C-reactive protein (standardized regression coefficient, 0.241; P ⫽ 0.006; Fig. 3) were significantly correlated with the ratio of copeptin/AVP plasma concentrations. Discussion

Copeptin plasma concentrations were significantly higher in critically ill patients, compared with healthy volunteers. Cardiac surgery patients had higher copeptin plasma concentrations than critically ill patients with sepsis and with SIRS. In healthy volunteers as well as in critically ill patients, plasma concentrations of AVP and copeptin correlate significantly with each other. Although the ratio of copeptin/ AVP plasma concentrations was comparable in healthy volunteers and patients after cardiac surgery, it was significantly increased in critically ill patients with sepsis and with SIRS. The significant correlation between copeptin and AVP plasma concentrations in healthy blood donors, as well as in critically ill patients with sepsis, with SIRS, or after cardiac surgery indicates that both assays measured the result of the same physiological process, namely the secretion of the equimolar proteolytic products of pre-pro-vasopressin from the neurohypophysis (11, 12). Similarly, a recent study has reported significantly increased plasma copeptin concentrations in patients with septic shock (15), a state in which AVP plasma concentrations have been observed to be significantly elevated, too (2, 4, 18). Interestingly, the correlation between AVP and copeptin measurements was lowest in the healthy blood donors group. The reason for this unexpected observation may be the fact that, in contrast to the copeptin assay, the lower detection limit of the AVP assay was well within the range of AVP plasma concentrations in the healthy blood donors

TABLE 1. Demographic and clinical data of healthy volunteers and the three subgroups of critically ill patients

n Age (yr) Male sex (n/%) Systolic blood pressure (mm Hg) Serum sodium concentration (mmol/liter) MODS score (pts) ICU length of stay (d) ICU mortality (n/%)

Volunteers

Sepsis

SIRS

Post CS

P valuea

70 41 ⫾ 12 40/57 NA NA NA NA NA

25 60 ⫾ 12 16/64 103 ⫾ 13 144 ⫾ 7 6.9 ⫾ 3 12 ⫾ 13 2/8

36 55 ⫾ 21 26/72.2 111 ⫾ 23 142 ⫾ 5 4.6 ⫾ 2.7 14 ⫾ 11 2/5.6

96 65 ⫾ 10 50/52.1 112 ⫾ 25 141 ⫾ 5 4.1 ⫾ 2.1 9⫾8 11/11.5

0.001 0.098 0.135 0.058 0.001 0.057 0.497

CS, Cardiac surgery; MODS, multiple organ dysfunction syndrome; NA, not applicable. Data are given as mean values ⫾ SD, unless described otherwise. a Differences between patients with sepsis, with SIRS, and after cardiac surgery.

4384



FIG. 2. Correlations between copeptin and AVP plasma concentrations in healthy volunteers (A), critically ill patients with sepsis (B), critically ill patients with SIRS (C), and patients after cardiac surgery (D).

group. Thus, approximately 12% of test results had to be excluded because of potentially being nonquantitative measurements. In critically ill patients, on the other hand, AVP plasma concentrations were significantly higher and, in none of the patients, below the lower detection limit of the AVP assay. Based on these results, determination of copeptin plasma concentrations in critically ill patients deserves to be evaluated as an indirect laboratory parameter to assess the vasopressinergic system in future studies. In view of the rapid

availability of test results, copeptin analysis may be appropriate to answer urgent clinical questions. For the critical care clinician, this could be particularly helpful in patients where knowledge of endogenous AVP concentrations is pivotal for therapy (10), such as in patients with prolonged perioperative hypotension and ongoing vasopressor drug requirements (8) or patients with neurological disorders and electrolyte disturbances (9). However, the copeptin/AVP plasma concentration ratio was significantly different among patients after cardiac



4385

FIG. 3. Correlation between the copeptin/ AVP ratio and plasma concentrations of C-reactive protein in critically ill patients.

surgery and critically ill patients with either sepsis or SIRS. In the linear regression model, the copeptin/AVP ratio was significantly correlated with the American Society of Anesthesiologists’ classification and plasma concentrations of the C-reactive protein. These results indicate that both the preoperative risk assessment and, in particular, the severity of systemic inflammation contribute to a higher copeptin/AVP ratio. Accordingly, these results are consistent with the clinical observation that both sepsis and SIRS are defined by the presence of systemic inflammation (19). In contrast, only a minority of patients after cardiac surgery in this study population (25 of 95, 26%) developed SIRS during the first 24 h after surgery. Therefore, for clinical interpretation of copeptin plasma concentrations in critically ill patients, this implies that copeptin tends to overestimate AVP plasma concentrations in critically ill patients with sepsis and SIRS, whereas a comparable ratio of copeptin and AVP plasma concentrations can be found in postoperative cardiac surgery patients as in healthy volunteers. When interpreting the results of the copeptin/AVP ratio of this study, an important limitation needs to be considered. Because a certain part of test samples in the healthy control group had to be excluded because they were lying below the lower detection limit of the AVP assay, the ratio of copeptin/ AVP plasma concentrations in healthy volunteers may not be appropriate to serve as a standardized ratio. Therefore, no statistical comparisons of the ratio of copeptin/AVP plasma concentrations have been made in this study.

In the 1970s, there have already been attempts to measure another part of the pre-pro-vasopressin molecule, namely neurophysin II. Robertson and colleagues (23, 24) have first described a RIA to determine human neurophysin II. Using this assay in healthy, sitting, normally hydrated subjects of both sexes, plasma neurophysin concentrations were found to be 73 ⫾ 5 pg/ml (mean ⫾ sem, n ⫽ 20) (25). Although showing a highly significant correlation with AVP plasma concentrations in small numbers of selected patients (26, 27), neurophysin testing in patients with small-cell lung cancer has not gained wide clinical acceptance. In conclusion, plasma concentrations of copeptin and AVP in healthy volunteers and critically ill patients correlate significantly with each other. The ratio of plasma copeptin and AVP concentrations is increased in patients with sepsis and with SIRS, suggesting that copeptin may overestimate AVP plasma concentrations in these patients. Acknowledgments We thank Mrs. Irene Gaggl and the team of the Laboratory of Molecular Endocrinology at the Institute of Pathophysiology at Innsbruck Medical University for analyzing the blood samples. We also thank the nursing staff of the surgical ICU of Innsbruck Medical University. Received December 29, 2005. Accepted August 23, 2006. Address all correspondence and requests for reprints to: Stefan Jochberger, M.D., Department of Anesthesiology, Innsbruck Medical University, Anichstrasse 35, 6020 Innsbruck, Austria. E-mail: [email protected].

4386



References 1. Du¨nser MW, Wenzel V, Mayr AJ, Hasibeder WR 2003 Management of vasodilatory shock: defining the role of arginine vasopressin. Drugs 63:237–256 2. Mutlu GM, Factor P 2004 Role of vasopressin in the management of septic shock. Intensive Care Med 30:1276 –1291 3. Wilson MF, Brackett DJ, Tompkins P, Benjamin B, Archer LT, Hinshaw LB 1981 Elevated plasma vasopressin concentrations during endotoxin and E. coli shock. Adv Shock Res 6:15–26 4. Sharshar T, Blanchard A, Paillard M, Raphael JC, Gajdos P, Annane D 2003 Circulating vasopressin levels in septic shock. Crit Care Med 31:1752–1758 5. Landry DW, Levin HR, Gallant EM, Ashton Jr RC, Seo S, D’Alessandro D, Oz MC, Oliver JA 1997 Vasopressin deficiency contributes to the vasodilation of septic shock. Circulation 95:1122–1125 6. Luckner G, Du¨nser MW, Jochberger S, Mayr VD, Wenzel V, Ulmer H, Schmid S, Knotzer H, Pajk W, Hasibeder WR, Mayr AJ, Friesenecker BE 2005 Arginine vasopressin in 316 patients with advanced vasodilatory shock. Crit Care Med 33:2659 –2666 7. Oliver JA, Landry DW 2003 Patient with a sudden drop in blood pressure. Crit Care Med 31:326 –327 8. Morales DL, Garrido MJ, Madigan JD, Helman DN, Faber J, Williams MR, Landry DW, Oz MC 2003 A double-blind randomized trial: prophylactic vasopressin reduces hypotension after cardiopulmonary bypass. Ann Thorac Surg 75:926 –930 9. Palmer BF 2003 Hyponatremia in patients with central nervous system disease: SIADH versus CSW. Trends Endocrinol Metab 14:182–187 10. Robertson GL 1994 The use of vasopressin assays in physiology and pathophysiology. Semin Nephrol 14:368 –383 11. Robertson GL, Mahr EA, Athar S, Sinha T 1973 Development and clinical application of a new method for the radioimmunoassay of arginine vasopressin in human plasma. J Clin Invest 52:2340 –2352 12. Preibisz JJ, Sealey JE, Laragh JH, Cody RJ, Weksler BB 1983 Plasma and platelet vasopressin in essential hypertension and congestive heart failure. Hypertension 5:129 –138 13. Kluge M, Riedl S, Erhart-Hofmann B, Hartmann J, Waldhauser F 1999 Improved extraction procedure and IRA for determination of arginine 8-vasopressin in plasma: role of premeasurement sample treatment and reference values in children. Clin Chem 45:98 –103 14. Morgenthaler NG, Struck J, Alonso C, Bergmann A 2006 Assay for the

15. 16. 17. 18.

19.

20. 21. 22. 23. 24. 25. 26.

27.

measurement of copeptin, a stable peptide derived from the precursor of vasopressin. Clin Chem 52:112–119 Struck J, Morgenthaler NG, Bergmann A 2005 Copeptin, a stable peptide derived from the vasopressin precursor, is elevated in serum of sepsis patients. Peptides 26:2500 –2504 De Bree FM, Burbach JP 1998 Structure-function relationships of the vasopressin prohormone domains. Cell Mol Neurobiol 18:173–191 Holwerda DA 1972 A glycopeptide from the posterior lobe of pig pituitaries. I. Isolation and characterization. Eur J Biochem 28:334 –339 Jochberger S, Mayr VD, Luckner G, Wenzel V, Ulmer H, Schmid S, Knotzer H, Pajk W, Hasibeder WR, Friesenecker BE, Mayr AJ, Du¨nser MW 2006 Serum vasopressin concentrations in critically ill patients. Crit Care Med 34:293–299 Levy MM, Fink MP, Marshall JC, Abraham E, Angus D, Cook D, Cohen J, Opal SM, Vincent JL, Ramsay G; SCCM/ESICM/ACCPT/ATS/SIS 2003 2001 SCCM/ESICM/ACCP/ATS/SIS International Sepsis Definitions Conference. Crit Care Med 31:1250 –1256 Lefering R, Goris RJ, van Nieuwenhoven EJ, Neugebauer E 2002 Revision of the multiple organ failure score. Langenbecks Arch Surg 387:14 –20 Beardwell CG 1971 Radioimmunoassay of arginine vasopressin in human plasma. J Clin Endocrinol Metab 33:254 –260 Le Gall JR, Lemeshow S, Saulnier F 1993 A new Simplified Acute Physiology Score (SAPS II) based on a European/North American multicenter study. JAMA 270:2957–2963 Robertson AG 1975 Isolation, assay, and secretion of individual human neurophysins. J Clin Invest 55:360 –367 Seif SM, Huellmantel AB, Platia MP, Haluszczak C, Robinson AG 1977 Isolation, radioimmunoassay and physiologic secretion of rat neurophysins. Endocrinology 100:1317–1326 North WG, LaRochelle Jr FT, Melton J, Mills RC 1980 Isolation and partial characterization of two human neurophysins: their use in the development of specific radioimmunoassays. J Clin Endocrinol Metab 51:884 – 891 Legros JJ, Geenen V, Carvelli T, Martens H, Andre M, Corhay JL, Radermecker M, Zangerle PF, Sassolas G, Gharib C 1990 Neurophysins as markers of vasopressin and oxytocin release. A study in carcinoma of the lung. Horm Res 34:151–155 Pullan PT, Clappison BH, Johnston CI 1979 Plasma vasopressin and human neurophysins in physiological and pathological states associated with changes in vasopressin secretion. J Clin Endocrinol Metab 49:580 –587

JCEM is published monthly by The Endocrine Society (http://www.endo-society.org), the foremost professional society serving the endocrine community.