| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Clinical Institute of Medical and Chemical Laboratory Diagnostics (S.P., H.M., B.T.), Medical University Graz, A-8036 Graz, Austria; Ludwigshafen Risk and Cardiovascular Health Study nonprofit L.L.C. (B.W.,U.S.), D-79085 Freiburg, Germany; Division of Endocrinology (B.O.B.), Graduate School Molecular Diabetology and Endocrinology, Ulm University, D-89081 Ulm, Germany; Cardiology Group (B.R.W.), D-60594 Frankfurt Sachsenhausen, Germany; and Synlab Centre of Laboratory Diagnostics (W.M.), D-69031 Heidelberg, Germany
Address all correspondence and requests for reprints to: Professor Dr. Med. Winfried März, Synlab Centre of Laboratory Diagnostics, P.O. Box 10 47 80, D-69031 Heidelberg, Germany. E-mail: maerz{at}synlab.de.
 |
Abstract |
|---|
 Top Abstract IntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Objective: The objective of the study was to elucidate the relationship between adiponectin and mortality.
Design, Setting, and Participants: Adiponectin was determined in 2473 persons with and 673 persons without angiographic CAD. During a mean follow-up period of 5.45 yr, 427 persons with CAD and 55 persons without CAD died.
Main Outcome Measure: Hazard ratios for mortality according to adiponectin levels were measured.
Results: Adiponectin was positively related to female gender, age, low-density lipoprotein cholesterol, high-density lipoprotein cholesterol, homocysteine, and N-terminal pro-B-type natriuretic peptide. It was inversely related to glomerular filtration rate, body mass index, and triglycerides and was low in diabetes mellitus and CAD. An increase of 1 SD in adiponectin was associated with unadjusted and fully adjusted hazard ratios for death from any cause of 1.31 [95% confidence interval (CI) 1.20–1.42] and 1.22 (95% CI 1.12–1.34), and for death from cardiovascular causes of 1.32 (95% CI 1.19–1.45) and 1.23 (95% CI 1.11–1.37), respectively. In angiographic CAD, stable CAD, and unstable CAD, the predictive value of adiponectin was similar to that in the entire cohort, but it did not attain statistical significance in persons without angiographic CAD. Adiponectin was also positively related to the risk of death from noncardiovascular causes.
Conclusions: Despite the common view about adiponectin as a protective molecule in cardiovascular disease, high adiponectin independently predicts all-cause, cardiovascular, and noncardiovascular mortality in individuals with CAD.
|
Introduction |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
|
Subjects and Methods |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
We studied white participants of the Ludwigshafen Risk and Cardiovascular Health (LURIC) Study. LURIC is an ongoing prospective cohort study of white individuals investigating risk factors for CAD (21). Between June 1997 and January 2000, 3316 patients who all had undergone coronary angiography were included. Adiponectin serum levels were measured in randomly selected 3146 LURIC probands, comprising 95% of the LURIC cohort. The study was approved by the Institutional Review Board at the Ärztekammer Rheinland-Pfalz. Informed written consent was obtained from each of the participants. With the exception of acute coronary syndromes, the patients had to present in a stable clinical condition without major concomitant noncardiovascular disease.
CAD was assessed by angiography using the maximum luminal narrowing estimated by visual analysis. Clinically relevant CAD was defined as the occurrence of at least one stenosis 20% or greater in at least one of 15 coronary segments. Individuals with stenoses less than 20% were considered as controls. In 2874 of the 3146 persons included in this analysis, left ventricular (LV) function had been graded semiquantitatively by contrast ventriculography into normal or minimally, moderately, or severely impaired. Among these, the LV ejection fraction, calculated from the right anterior oblique view, was available in 1247 patients. In this subgroup, the semiquantitative assessment of LV function correlated well with the calculated LV ejection fraction (Spearman’s correlation coefficient = –0.834, P < 0.001), suggesting that the semiquantitative grading provides a reliable estimate of LV function.
Diabetes mellitus was diagnosed if plasma glucose was greater than 1.25 g/liter in the fasting state or greater than 2.00 g/liter 2 h after the oral glucose load, respectively, or if individuals were receiving oral antidiabetics or insulin. Hypertension was diagnosed if the systolic and/or diastolic blood pressure exceeded 140 and/or 90 mm Hg or if there was a clinically significant history of hypertension. The definition of the National Cholesterol Education Program Adult Treatment Panel III for the metabolic syndrome was used.
Fasting blood collection was done before coronary angiography and measurements of adiponectin, lipoproteins, C-reactive protein (CRP), creatinine, fibrinogen, homocysteine, and N-terminal pro-B-type natriuretic peptide (NT-pro-BNP) were complete in 3146 individuals with coronary angiograms. Among these, 2473 persons had angiographic CAD, 1485 were in a clinically stable condition, whereas 988 presented within 7 d after onset of symptoms of unstable angina, non-ST-elevation myocardial infarction (NSTEMI; troponin T > 0.1 µg/liter), or ST-elevation myocardial infarction (STEMI; troponin T > 0.1 µg/liter).
Information on vital status was obtained from local person registries. No patients were lost to follow-up. Among the 3146 persons studied, 482 deaths (15.3%) had occurred during a median time of follow-up of 5.45 yr. Death certificates were obtained for 467 of the decedents (97%) and were missing for 15 decedents (3%) who were included in the total mortality analysis but excluded from the cardiovascular mortality analysis. Cardiovascular death included the following categories: sudden death, fatal myocardial infarction, death due to congestive heart failure, death immediately after intervention to treat CAD, fatal stroke, and other causes of death due to CAD.
Measurements of biochemical variables
The standard laboratory methods used have been described (21). Lipoproteins were separated by a combined ultracentrifugation-precipitation method. Sensitive CRP was measured by immunonephelometry (N High Sensitivity CRP; Dade Behring, Marburg, Germany). NT-pro-BNP was measured by electrochemiluminescence on an Elecsys 2010 (Roche Diagnostics, Mannheim, Germany). Adiponectin was determined in serum by ELISA (Biovendor Laboratory Medicine Inc., Brno, Czech Republic) according to the manufacturer’s instructions. Both the intra- and interassay coefficient of variation were less than 10%. Glomerular filtration rate (GFR) was calculated as GFR (millimeters per minute/1.73 m2) = 186·creatinine–1.154·age–0.203 and GFR (millimeters per minute/1.73 m2) = 142·creatinine–1.154·age–0.203 in males and females, respectively.
Statistical analysis
Adiponectin, triglycerides, and NT-pro-BNP were transformed logarithmically before being used in parametric procedures. We established quartiles of continuous variables according to the values in control subjects without angiographic CAD. Clinical and biochemical characteristics of CAD patients and controls are presented as percentages for categorical variables and means ± SDs or medians and 25th and 75th percentiles for continuous variables. Associations of categorical and continuous variables were analyzed by logistic regression and univariate ANOVA, respectively, with covariables as indicated (Table 1
). We studied the effects of gender, age, CAD, and cardiovascular risk factors on adiponectin using ANOVA models in which we included those factors not under examination as covariables (Table 2). To examine the relationship of adiponectin on total and cardiovascular mortality, we calculated hazard ratios and 95% confidence intervals (CIs) using the Cox proportional hazards model both according to quartiles and per increase of 1 SD (calculated from the logarithmically transformed values). Multivariable adjustment was carried out for gender, age, angiographic CAD, and cardiovascular risk factors as indicated in the legends of Tables 3 and 4. All statistical tests were two sided. P < 0.05 was considered statistically significant. The SPSS 11.0 statistical package (SPSS Inc., Chicago, IL) was used.
|
|
|
|
|
Results |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Compared with individuals without CAD, patients with angiographic CAD were significantly older; current or past smoking, type 2 diabetes mellitus, hypertension, and cerebrovascular and peripheral artery disease were more prevalent. Fifty-three percent of the CAD patients had a history of myocardial infarction. The CAD patients had higher systolic blood pressure, higher fasting glucose, higher homocysteine, higher triglycerides, and high-density lipoprotein cholesterol (HDL-C). Low-density lipoprotein cholesterol (LDL-C) was slightly lower in CAD patients than controls, even after accounting for the use of lipid-lowering drugs. CRP and fibrinogen were higher in CAD patients than controls, a finding that in part relates to the presence of patients with acute coronary syndromes in the CAD group. CAD patients had significantly higher NT-pro-BNP and were more likely to have impaired LV function. Body mass index, diastolic blood pressure, and GFR were similar in patients and controls (Table 1
). Analysis of subjects with CAD revealed that at the fourth adiponectin quartile, the prevalence of patients with myocardial infarction and three-vessel disease was even significantly lower, compared with the first adiponectin quartile (P < 0.05 for both), and that there was no significant difference in the distribution of classes of stable angina according to the Canadian Cardiovascular Society angina score (data not shown).
Association of adiponectin with cardiovascular risk factors and CAD status
We examined the effect of gender, age, angiographic CAD, and risk factors on adiponectin in a general linear model in which we included those factors not under examination as covariables (Table 2). Adiponectin was significantly higher in women than men and increased with age; it was lower in patients with diabetes mellitus than nondiabetic subjects. Patients with CAD had significantly lower adiponectin concentrations than those without CAD, the values being lowest in patients presenting with unstable CAD. Adiponectin increased with increasing quartiles of HDL-C and LDL-C and decreased at high triglycerides. GFR was negatively and homocysteine was positively related to adiponectin. Hypertension, smoking status, and fibrinogen were not significantly associated with adiponectin. Adiponectin slightly, but significantly, decreased as CRP increased. High concentrations of NT-pro-BNP were also accompanied by high concentrations of adiponectin. Overall, the Pearson’s correlation coefficient between logarithmically transformed NT-pro-BNP and adiponectin was 0.235 (P < 0.001). In persons with normal or minimally, moderately, or severely impaired LV function, the respective correlation coefficients were 0.227 (n = 2033, P < 0.001), 0.326 (n = 310, P < 0.001), 0.364 (n = 345, P < 0.001), and 0.406 (n = 186, P < 0.001). Multiple linear regression using forward selection of independent variables revealed that the four most important predictors of the adiponectin concentration were HDL-C greater than NT-pro-BNP greater than gender greater than triglycerides.
Adiponectin and mortality from all causes
Compared with subjects in the lowest quartile of adiponectin, the unadjusted hazard ratios for death at adiponectin concentrations in the second through fourth quartile were 1.15 (95% CI 0.90–1.47), 1.34 (95% CI 1.04–1.73), and 2.07 (95% CI 1.63–2.63), respectively (model 1, Table 3). Inclusion of age and gender as covariables only slightly modified these estimates (model 2, Table 3). Adiponectin retained its prognostic importance after adjusting for the CAD status at presentation (no CAD, stable CAD, or unstable CAD) and risk factors (model 3, Table 3). Inclusion of NT-pro-BNP into the model slightly decreased hazard ratios, but adiponectin in the upper quartile stayed significantly associated with death from any cause (model 4, Table 3). An increase of 1 SD in adiponectin was associated with unadjusted (model 1, Table 3) and fully adjusted (model 4, Table 3) hazard ratios for death from any cause of 1.31 (95% CI 1.20–1.42) and 1.22 (95% CI 1.12–1.34), respectively. Analyses stratified by gender did not materially change our findings (men: hazard ratio 1.20, 95% CI 1.09–1.33; women: hazard ratio 1.29 95% CI 1.04–1.59, fully adjusted models).
Among the 2473 subjects with angiographic CAD, 427 (17.2%) died during follow-up. In this subgroup, hazard ratios for death were slightly higher than those obtained in the entire study sample (models 1–4, Table 3).
Among the subjects with angiographic CAD, 1485 subjects had stable CAD. Nine hundred eighty-eight patients presented with unstable CAD (unstable angina, NSTEMI, or STEMI). In these subgroups of patients, 267 (18.0%) and 160 deaths (16.2%), respectively, occurred. In both of them, we found consistent and robust associations of adiponectin with mortality from all causes, with a tendency toward higher hazard ratios in patients with unstable CAD (Table 3, models 1–4).
Only 55 deaths (8.2%) occurred among the 673 subjects with coronary stenoses less than 20%. There was a tendency toward an increased risk of death at high adiponectin (unadjusted hazard ratio of death of 1.22 per 1 SD, 95% CI 0.93–1.59). This association, however, did not reach statistical significance and was not robust against adjustment for confounding variables (fully adjusted hazard ratio of 0.99 per 1 SD, 95% CI 0.75–1.31).
Adiponectin and mortality from cardiovascular causes
Because death certificates were not available from 15 deceased persons, the analysis for cardiovascular mortality included a total of 3131 individuals. Among these, 326 (10.4%) died from cardiovascular causes. Hazard ratios for death from cardiovascular causes according to adiponectin were similar to those obtained for mortality from all causes in all models and across all subgroups examined (Table 4). An increase of 1 SD in adiponectin generated unadjusted and fully adjusted hazard ratios for death from cardiovascular causes of 1.32 (95% CI 1.19–1.45) and 1.23 (95% CI 1.11–1.37), respectively. Again, consistent results were obtained when we analyzed both genders separately (men: hazard ratio 1.23, 95% CI 1.09–1.39; women: hazard ratio 1.34 95% CI 1.03–1.73, fully adjusted models).
Among the 673 subjects with coronary stenoses less than 20%, 36 deaths from cardiovascular causes (5.3%) occurred. We again found a nonsignificant positive association with adiponectin, which disappeared after adjustment for confounding variables (unadjusted hazard ratio of cardiovascular death of 1.37 per 1 SD, 95% CI 0.99–1.89; fully adjusted hazard ratio of 1.07 per 1 SD, 95% CI 0.76–1.50).
Adiponectin and mortality from causes other than cardiovascular
Among 3131 persons with death certificates available, 141 (4.6%) died from noncardiovascular causes, 30 (1%) from fatal infection, 53 from cancer (1.7%), and 58 (1.9%) from miscellaneous diseases. Adiponectin was also associated with death from noncardiovascular causes, unadjusted and fully adjusted hazard ratios being 1.29 (95% CI 1.11–1.51) and 1.21 (95% CI 1.02–1.44), respectively, per increase by 1 SD.
Adiponectin, LV function, and mortality
Adiponectin and NT-pro-BNP significantly correlated, but adiponectin remained predictive of total and cardiovascular mortality after adjusting for NT-pro-BNP (Table 3, model 4). To further substantiate that adiponectin provided prognostic information independent from LV function, we generated nested strata of individuals according to LV function and quartiles of adiponectin (Fig. 1). In each of the subgroups defined by different LV function, adiponectin in the fourth quartile was associated with an approximate doubling of the hazard ratios for total and cardiovascular death, compared with the first quartile. The highest hazard ratios were encountered in the layer with severely impaired LV function and the highest concentration of adiponectin (Fig. 1).
|
|
Discussion |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Over the last few years, several cross-sectional and in vitro studies supported the notion that adiponectin would protect against vascular diseases (7, 8, 9, 10, 11, 12, 13, 14, 15, 22). However, this opinion has been challenged by two recent prospective studies showing no association between adiponectin and the incidence of cardiovascular disease (16, 17). To the best of our knowledge, there are currently only three other prospective studies published addressing the relationship between adiponectin and mortality, and neither of these has investigated a large number of extensively phenotyped subjects. In the first one, including subjects with end-stage renal disease (n = 227), adiponectin plasma levels did not predict all-cause mortality, but there was a slight, yet significant association between low adiponectin and cardiovascular events (23). The second study was performed in subjects with first-ever ischemic stroke (n = 160) who were examined for adiponectin within 24 h after this event (24). Whereas there was no significant association of adiponectin with coronary heart disease at baseline, subjects who had died during the 5-yr follow-up period had significantly reduced adiponectin levels at baseline when compared with survivors.
The third study performed by Kistorp et al. (25) analyzed 195 patients with chronic heart failure (CHF). Adiponectin was positively related to both increased mortality and NT-pro-BNP, an established marker for CHF, suggesting that in subjects with CHF, high adiponectin concentrations may be a marker of a wasting process in subjects with increased risk of death (25). Our study confirms and largely extends these findings. Multivariate regression revealed that NT-pro-BNP was ranked among the most important predictors of adiponectin levels. The correlation coefficient between adiponectin and NT-pro-BNP was 0.24 in all patients, and it was 0.41 in patients with severely impaired LV function, which is very similar to the correlation coefficient of 0.47 reported by Kistorp et al. (25). Remarkably, adiponectin was associated with total and cardiovascular mortality, independent of NT-pro-BNP and in subjects with normal LV function, suggesting that the link between high adiponectin and mortality is not confined to patients suffering from CHF. Hence, mechanisms different from a general wasting process may relate the adiponectin concentration to cardiovascular mortality.
One such linking mechanism may involve renal function. ANOVA revealed that the calculated GFR was a strong inverse predictor of adiponectin. Even though the synthesis of adiponectin appears decreased in renal disease (26), the clearance of adiponectin by the kidney may have a strong influence on its concentration in the steady state (27). Hence, high adiponectin may merely reflect impaired renal function, which is a cardiovascular risk factor on its own (28). Considering putative vasculoprotective associations of adiponectin with markers of endothelial cell activation/injury, it was also speculated that the elevation of adiponectin in renal failure may be a counterregulatory response aimed at mitigating the endothelial damage in patients with renal dysfunction (29). Furthermore, adiponectin increased in parallel to homocysteine in the current study. Homocysteine is a strong marker of cardiovascular risk (30) and is also eliminated by the kidney (31). Thus, the relationship between adiponectin on the one hand and GFR and homocysteine on the other hand may in part account for the increased mortality at high adiponectin concentrations. It should be noted, however, that the association between adiponectin and mortality remained stable after adjustment for these variables, and it remained unaffected by other potentially confounding cardiovascular risk factors linked to high adiponectin such as age or HDL-C. Thus, yet-unknown reasons may underlie the association between adiponectin and the risk of death, a consideration that is supported by the finding that adiponectin was also positively related to death from causes other than cardiovascular.
Apart from this, we also want to note that we are aware that subjects with acute coronary syndromes are known to have changes in several parameters, including inflammatory biomarkers or LDL. However, the fact that we observed similar associations between adiponectin and mortality in subgroup analysis of subjects with stable and unstable CAD argues against a strong influence on our results. Furthermore, it should also be noted that the results of an increased mortality at the fourth adiponectin quartile cannot be attributed to more severe CAD in these subjects because the prevalence of patients with myocardial infarction and three-vessel disease was even significantly lower, compared with the first adiponectin quartile, and there was no significant difference in the distribution of classes of stable angina according to the Canadian Cardiovascular Society angina score.
Can our findings be brought in line with the current contention that adiponectin acts as an antiatherosclerotic molecule? Adiponectin has been shown to accumulate in the injured vessel wall and protect against atherosclerosis in apolipoprotein E-deficient mice (32, 33). Adiponectin is considered to play a beneficial role as a scaffold of newly formed collagen in myocardial remodeling after ischemic injury (34), and its supplementation in mice was shown to attenuate cardiac hypertrophy in response to pressure overload (35). Interestingly, in type 2 diabetes, the transcardiac extraction of adiponectin has been found impaired, measured as the difference between adiponectin in the aortic root and the coronary sinus (36). Such examples of impaired function of adiponectin have been termed adiponectin resistance and may in part be due to dysfunction or down-regulation of adiponectin receptors, as recently observed in obesity (37). In keeping with this suggestion, one could consider our results as the consequence of an active counterregulatory up-regulation of this adipokine in subjects at high risk. The hypothesis that aging and vascular disease may raise adiponectin has been suggested recently (38, 39, 40).
Considering that adiponectin was a predictor of mortality in our study, it may be viewed as paradoxical that, compared with controls, we found lower adiponectin levels but higher mortality rates in subjects with CAD. Toward this it should be noted that the observed associations between adiponectin and mortality cannot prove causality. Therefore, the concept of adiponectin as an antiatherosclerotic molecule that predicts mortality due to a counterregulatory elevation in high-risk subjects fits well with our study results. Furthermore, the data about mortality in our controls should be interpreted with caution because, compared with subjects with CAD, they are derived from a relatively low number of deaths. Thus, further studies with more cases are warranted to provide more reliable results about the association of adiponectin and mortality in subjects free of CAD.
It should be considered that although many beneficial features have been attributed to adiponectin, our results raise the possibility that adiponectin also has undesirable effects, warranting careful consideration of therapeutic approaches that aim to enhance the secretion or action of this molecule (22, 37). Toward this, the observed associations of adiponectin with renal function and heart failure should be further elucidated to clarify whether adiponectin is only a marker of these diseases or is also pathophysiologically involved in the development of renal or cardiac dysfunction. Notably, the influence of adiponectin on the myocardium is further supported by the fact that the carriers of the single nucleotide polymorphism +276 G/G were recently found with significantly higher left ventricular mass (41).
A limitation of this study is that we enrolled individuals in whom coronary angiography was clinically indicated. This population is at intermediate to high risk of future cardiovascular events, and our findings may not apply to persons at lower risk. Another limitation of our study is that we determined cause of death simply by death certificates. Furthermore, we analyzed adiponectin without accounting for the occurrence of different isoforms that may have distinct biological functions (42).
In summary, the current study is the largest one available to address the relationship between adiponectin and death. At least among the patients with angiographic CAD, adiponectin turned out to be a surprisingly robust predictor of all-cause, cardiovascular, and noncardiovascular mortality, respectively, independent from established and emerging risk factors.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
Disclosure statement: The authors have nothing to disclose.
First Published Online August 15, 2006
Abbreviations: CAD, Coronary artery disease; CHF, chronic heart failure; CI, confidence interval; CRP, C-reactive protein; GFR, glomerular filtration rate; HDL-C, high-density lipoprotein cholesterol; LDL-C, low-density lipoprotein cholesterol; LURIC, Ludwigshafen Risk and Cardiovascular Health; LV, left ventricular; NSTEMI, non-ST-elevation myocardial infarction; NT-pro-BNP, N-terminal pro-B-type natriuretic peptide; STEMI, ST-elevation myocardial infarction.
Received April 19, 2006.
Accepted August 8, 2006.
|
References |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
and decreases adiponectin gene expression in adipose tissue. A study in monozygotic twins. J Clin Endocrinol Metab 91:2776–2781This article has been cited by other articles:
 |
|
 |
|
  M. Karmazyn, D. M. Purdham, V. Rajapurohitam, and A. Zeidan Signalling mechanisms underlying the metabolic and other effects of adipokines on the heart Cardiovasc Res, July 15, 2008; 79(2): 279 - 286. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. M. Dekker, T. Funahashi, G. Nijpels, S. Pilz, C. D. A. Stehouwer, M. B. Snijder, L. M. Bouter, Y. Matsuzawa, I. Shimomura, and R. J. Heine Prognostic Value of Adiponectin for Cardiovascular Disease and Mortality J. Clin. Endocrinol. Metab., April 1, 2008; 93(4): 1489 - 1496. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Rao, L. Li, H. Tighiouart, B. L. Jaber, B. J. G. Pereira, V. S. Balakrishnan, and the HEMO Study Group Plasma adiponectin levels and clinical outcomes among haemodialysis patients Nephrol. Dial. Transplant., March 11, 2008; (2008) gfn070v1. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. Schnabel, C. M. Messow, E. Lubos, C. Espinola-Klein, H. J. Rupprecht, C. Bickel, C. Sinning, S. Tzikas, T. Keller, S. Genth-Zotz, et al. Association of adiponectin with adverse outcome in coronary artery disease patients: results from the AtheroGene study Eur. Heart J., March 1, 2008; 29(5): 649 - 657. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. G. Wannamethee, P. H. Whincup, L. Lennon, and N. Sattar Circulating Adiponectin Levels and Mortality in Elderly Men With and Without Cardiovascular Disease and Heart Failure Arch Intern Med, July 23, 2007; 167(14): 1510 - 1517. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
