Skip Navigation



American Journal of Epidemiology Advance Access published online on June 24, 2007
American Journal of Epidemiology, doi:10.1093/aje/kwm125
This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow Figures
Right arrow All Versions of this Article:
166/6/731    most recent
kwm125v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrowRequest Permissions
Right arrow Disclaimer
Google Scholar
Right arrow Articles by Maahs, D. M.
Right arrow Articles by Rewers, M.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Maahs, D. M.
Right arrow Articles by Rewers, M.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?

American Journal of Epidemiology Copyright © 2007 by the Johns Hopkins Bloomberg School of Public Health All rights reserved; printed in U.S.A.

Original Contribution

Determinants of Serum Adiponectin in Persons with and without Type 1 Diabetes

David M. Maahs1, Lorraine G. Ogden2, Janet K. Snell-Bergeon1, Gregory L. Kinney1, R. Paul Wadwa1, John E. Hokanson2, Dana Dabelea2, Adam Kretowski1, Robert H. Eckel3 and Marian Rewers1,2

1 Barbara Davis Center for Childhood Diabetes, University of Colorado Health Sciences Center, Aurora, CO
2 Department of Preventive Medicine and Biometrics, University of Colorado Health Sciences Center, Denver, CO
3 Department of Medicine, University of Colorado Health Sciences Center, Denver, CO

Correspondence to Dr. David M. Maahs, Barbara Davis Center for Childhood Diabetes, University of Colorado Health Sciences Center, P.O. Box 6511, Mail Stop A140, Aurora, CO 80045 (e-mail: David.Maahs{at}uchsc.edu).

Received for publication November 20, 2006. Accepted for publication March 13, 2007.


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 References
 
Low levels of adiponectin have been related to coronary heart disease, but adiponectin is higher in persons with type 1 diabetes who have an increased rate of coronary disease. In the Coronary Artery Calcification in Type 1 Diabetes Study (2000–2002), the authors investigated potential determinants of elevated adiponectin levels in persons with type 1 diabetes and whether a difference exists compared with nondiabetic persons. Serum adiponectin was measured in 1,393 persons (sex: 48% male; age: 38 (standard deviation: 9) years; diabetes duration: 23 (standard deviation: 9) years; 54% nondiabetic and 46% with type 1 diabetes). Determinants of log-transformed adiponectin levels were evaluated by multiple linear regression analysis with interaction terms to determine whether predictors of adiponectin levels differed by diabetes status. Adiponectin levels were higher in type 1 diabetic than nondiabetic persons (13.5 (standard deviation: 1.0) vs. 8.8 (standard deviation: 1.0) µg/ml; p < 0.0001), adjusting for age, gender, body mass index, and glomerular filtration rate. The final regression model explained 67% of the difference in adiponectin levels between type 1 diabetic and nondiabetic persons. The variables explaining this difference included high density lipoprotein cholesterol, albumin excretion rate, plasminogen activator inhibitor-1, and hemoglobin A1c level. Adiponectin is higher in type 1 diabetic than nondiabetic persons. Although some of the difference can be explained, further study is needed to better understand the relation between elevated adiponectin levels and patient outcomes, including coronary heart disease.

adiponectin; coronary disease; diabetes mellitus, type 1

Abbreviations: EGDR, estimated glucose disposal rate; GFRCG, glomerular filtration rate estimated by the Cockcroft-Gault formula; GFRMDRD, glomerular filtration rate estimated by the Modification of Diet in Renal Disease Equation; HDL, high density lipoprotein; PAI-1, plasminogen activator inhibitor-1


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 References
 
Adiponectin is a Mr 30,000 collagen-like protein synthesized by adipocytes that circulates in human plasma as approximately 0.01 percent of total plasma protein (1). Adiponectin is lower in males than females and in patients with type 2 diabetes, and it correlates negatively with intraabdominal fat, age, body mass index, insulin resistance and plasma insulin, triglyceride, glucose, and C-reactive protein levels (2–9). Conversely, adiponectin is positively correlated with high density lipoprotein (HDL) cholesterol (2) and is increased in persons with type 1 diabetes and in persons with decreasing renal function (10–16).

Adiponectin accumulates in damaged vascular walls (17) and beneficially modulates the endothelial inflammatory response to vascular injury (18, 19). In cross-sectional data, adiponectin has been hypothesized to be increased as a compensatory response in patients with type 1 diabetes who have microvascular complications (13–15). Adiponectin has also been associated with lipoprotein lipase activity (20) and inversely with plasma hepatic lipase activity (which may link adiponectin to HDL cholesterol) (21). Given adiponectin's relation with obesity and insulin resistance as well as other cardiovascular risk factors, it has been proposed as a link in the adipose-vascular axis (22) and may play a role in the development of atherosclerosis (23–25).

We have previously reported that low adiponectin levels predict progression of coronary artery calcification in a nested case-control study of young adults with and without type 1 diabetes (23). This relation, however, was significantly stronger in the nondiabetic persons and demonstrated a nonlinear association in persons with type 1 diabetes. More specifically, the odds ratio for coronary artery calcification progression decreased in each increasing quartile of adiponectin for nondiabetic persons but not for persons with type 1 diabetes in the highest quartile of adiponectin, suggesting a difference in the effects of adiponectin on coronary artery calcification progression between nondiabetic persons and persons with type 1 diabetes.

The reasons for elevated adiponectin in persons with type 1 diabetes and the paradox between the known antiatherogenic effects of adiponectin and the premature mortality from coronary artery disease in type 1 as compared with nondiabetic persons are unclear. Hypotheses to explain elevated adiponectin levels in persons with type 1 diabetes include a compensatory response to vascular injury (13, 14), decreased clearance due to renal insufficiency (26), effects of subcutaneous insulin treatment (27), and posttranslational modifications (glycosylation) (28, 29) that could differ in persons with type 1 diabetes. Previous investigations of determinants of adiponectin in type 1 diabetes have not included a nondiabetic control group. Therefore, the purpose of this investigation was 1) to confirm that adiponectin is higher in persons with type 1 diabetes than in nondiabetic persons in a large cohort and 2) to delineate the determinants of adiponectin (including renal function, obesity, and lipid-related measures) and whether a difference exists in the determinants of adiponectin levels between persons with type 1 diabetes and nondiabetic persons.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
 References
 
Study participants
The data presented in this report were collected as part of the baseline examination of 1,416 participants in the Coronary Artery Calcification in Type 1 Diabetes (CACTI) Study who were 19–56 years of age and included 652 men and women with type 1 diabetes and 764 nondiabetic controls (30). Patients with diabetes generally had been diagnosed when younger than 30 years of age and, among those who were 30 years or older at diagnosis, positive antibodies or a clinical course consistent with type 1 diabetes was present. All nondiabetic control subjects had never been diagnosed with diabetes including gestational diabetes and were generally spouses, friends, and neighbors of the cases (30). Serum adiponectin was measured in 1,393 persons on baseline samples (635 with type 1 diabetes and 758 nondiabetic controls).

Participants completed the baseline examination between March 2000 and April 2002, and a more detailed description of the study and baseline characteristics of this cohort has been published (31). Resting systolic blood pressure and fifth-phase diastolic blood pressure were measured three times while the participants were seated, and the second and the third measurements were averaged. Participants completed a standardized questionnaire including medical history and medication inventory as previously reported (31).

After an overnight fast, blood was collected and centrifuged, and separated plasma was stored at –70°C until assayed. Adiponectin was measured on stored samples in the Adult General Clinical Research Center core laboratory at the University of Colorado Health Sciences Center in Denver, Colorado, in duplicate, by use of a commercial radioimmunoassay procedure (Linco Research, Inc., St. Charles, Missouri). Stored samples from the participants' baseline study visit were diluted 1:500 prior to testing. Intraassay precision was 3.9 percent, and interassay precision is 8.5 percent. Results are reported in micrograms per milliliter, with a sensitivity cutoff of 1.0 µg/ml.

Total plasma cholesterol and triglyceride levels were measured by standard enzymatic methods, HDL cholesterol was separated with dextran sulfate, and low density lipoprotein cholesterol was calculated by the Friedewald formula. High performance liquid chromatography (Bio-Rad Laboratories, Inc., Hercules, California) was used to measure hemoglobin A1c. Plasma glucose was measured by use of the standard hexokinase method. Homocysteine was determined by the automated IMx procedure (Abbott Laboratories, Abbott Park, Illinois). C-reactive protein, plasminogen activator inhibitor-1 (PAI-1), and fibrinogen were measured in the laboratory of Dr. Russell Tracy at the University of Vermont (Burlington, Vermont). C-reactive protein was measured with a BN II nephelometer (Dade Behring, Deerfield, Illinois) utilizing a particle-enhanced immunonephelometric assay. PAI-1 was determined by a two-site enzyme-linked immunsorbent assay. Fibrinogen was measured in an automated clot-rate assay by use of a Star instrument (D-Star Instruments, Inc., Manassas, Virginia). Urinary albumin was measured by radioimmunoassay, and the albumin excretion rate was determined by radioimmunoassay, with the results of two timed overnight urine collections averaged.

We measured height and weight and calculated body mass index. Minimum waist and maximum hip measurements were obtained in duplicate, and the results were averaged. Intraabdominal fat and subcutaneous fat were assessed by an abdominal computed tomography scan at the L4–L5 levels. The total intraabdominal fat volume and subcutaneous fat volume in cubic centimeters were measured with AccuAnalyzer software from AccuImage (San Francisco, California).

Insulin resistance was approximated as the inverse of the estimated glucose disposal rate (EGDR), calculated according to the formula: EGDR = 24.31 – 12.22 x (waist/hip ratio) – 3.29 x (hypertension) – 0.568 x (hemoglobin A1c). The equation was derived from hyperinsulinemic euglycemic clamps performed in 24 type 1 diabetic participants in the Pittsburgh Epidemiology of Diabetes Complications Study (32). The glomerular filtration rate was estimated by both the Cockcroft-Gault formula (GFRCG) (33) and the Modification of Diet in Renal Disease Equation (GFRMDRD) (34). Duration of diabetes was determined by patient self-report. Current and former smoking status was obtained by questionnaire.

Statistical analysis
Data are presented as arithmetic means and standard deviations for continuous variables (geometric means and ranges for log-transformed variables) and percentages for categorical variables (table 1). Log adiponectin levels were compared between diabetic and nondiabetic persons, including adjustment for age, gender, measures of obesity (body mass index, waist/hip ratio, or computed tomography-measured visceral fat), and renal function (serum creatinine, albumin excretion rate, GFRCG, GFRMDRD), all reported determinants of adiponectin levels.


View this table:
[in this window]
[in a new window]

  		TABLE 1. Baseline characteristics of subjects with adiponectin measurements (n = 1,393), stratified by diabetes type, Coronary Artery Calcification in Type 1 Diabetes Study, 2000–2002

 
Correlation coefficients between log adiponectin and each of the covariates were calculated for all participants and then stratified by diabetes status. Fisher's z transformations were used to test for significant differences between correlation coefficients for type 1 versus nondiabetic participants in univariate analysis. Covariates with a correlation coefficient p value of less than 0.05 for all participants combined (table 2) were entered into a backward elimination linear regression model. Interaction terms between diabetes status and each of the covariates were added to this reduced model to investigate whether differences existed in the determinants of log adiponectin by diabetes status in multivariate analysis. Covariates associated with log adiponectin from stratified analyses on either diabetic or nondiabetic participants were then considered for entry into the model with their corresponding interaction terms to produce a final model. Variance inflation factors were calculated to assess the degree to which independent effects of the covariates could be estimated with a level of 10 as an a priori threshold for exclusion from the model selection process. (Waist circumference was the only variable excluded for this reason.) GFRCG and GFRMDRD were not considered in the initial models since they were highly correlated with the covariates used in their estimation. The relation between adiponectin and each of the glomerular filtration rate measures was thus estimated in a subsequent analysis, after removing gender, serum creatinine, waist/hip ratio, computed tomography visceral fat measurements, and albumin excretion rate from the model. Standardized ß coefficients for each covariate and their respective p values were calculated (table 3). Determinants of log adiponectin were further investigated, stratifying by albuminuria status (yes/no) and glomerular filtration rate (<60 and <30 ml/minute per 1.73 m2) (35). Finally, to determine the percentage of the difference in adiponectin levels between diabetic and nondiabetic persons that could be explained by the variables included in our final model, we added these variables sequentially to assess the percentage of change in the parameter estimate for the difference in adiponectin levels between type 1 and nondiabetic persons (table 4).


View this table:
[in this window]
[in a new window]

  		TABLE 2. Correlation coefficients and p values for log adiponectin, stratified by diabetes status, Coronary Artery Calcification in Type 1 Diabetes Study, 2000–2002

 

View this table:
[in this window]
[in a new window]

  		TABLE 3. Determinants of log adiponectin (standardized beta-coefficient and p value) in multiple linear regression, with all variables simultaneously in the model, Coronary Artery Calcification in Type 1 Diabetes Study, 2000–2002

 

View this table:
[in this window]
[in a new window]

  		TABLE 4. Difference in mean log adiponectin levels between type 1 diabetes mellitus and non-diabetes mellitus participants, with sequential adjustment for predictors of log adiponectin, Coronary Artery Calcification in Type 1 Diabetes Study, 2000–2002

 
Human subjects
The study protocol was reviewed and approved by the Colorado Combined Institutional Review Board, and informed consent was obtained from all participants prior to enrollment.


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
DISCUSSION
 References
 
Baseline characteristics of the 1,393 persons who had serum adiponectin measured were determined (table 1). As expected, adiponectin levels and numerous covariates differed between the diabetic and nondiabetic persons, including potential determinants of adiponectin such as age, race, glycemia, waist/hip ratio, visceral fat, insulin resistance, blood pressure, lipids, C-reactive protein, PAI-1, renal-related measures, homocysteine, alcohol intake, and current smoking.

Adiponectin was higher in diabetic persons than nondiabetic persons adjusted for age, gender, and body mass index (geometric mean: 13.5 (standard deviation: 1.0) vs. 8.8 (standard deviation: 1.0) µg/ml; p < 0.0001). This relation did not change with further adjustment for serum creatinine, GFRCG, GFRMDRD, HDL cholesterol, waist/hip ratio, or computed tomography-measured visceral fat (data not shown).

Correlation coefficients between log adiponectin and covariates, stratified by diabetes status with a diabetes interaction term, were determined (table 2). Of note, C-reactive protein and serum creatinine had significant correlations in both diabetic and nondiabetic persons, but in opposite directions. The strongest univariate predictors of log adiponectin were measurements of central adiposity (r = –0.35 to –0.42) and HDL cholesterol (r = 0.41). The strength of the relation did not vary by diabetes status for several variables highly correlated with log adiponectin (gender, obesity-related measurements, HDL cholesterol). Variables with significantly different correlation in diabetic versus nondiabetic persons included age, blood pressure, triglycerides, total cholesterol, C-reactive protein, PAI-1, fibrinogen, glycemia, insulin resistance, renal function, and race.

Correlates of log adiponectin levels were evaluated with multiple linear regression analysis with a backward elimination approach (table 3). Interaction terms were included in the final model to determine whether predictors of adiponectin levels differed by diabetes status. In both type 1 and nondiabetic persons, gender (male) and central adiposity measures were inversely correlated with adiponectin, while HDL cholesterol and homocysteine were positively correlated with adiponectin levels. Serum creatinine (borderline) had an inverse relation to adiponectin in nondiabetic persons but was positive in type 1 persons. Daily insulin dose (as a proxy for insulin resistance) was inversely related to adiponectin in type 1 persons. Variables positively predictive only in diabetic persons included hemoglobin A1c, fibrinogen, and albumin excretion rate, with triglycerides inversely related. For nondiabetic persons, PAI-1 and non-Hispanic White race were significantly inversely associated with adiponectin. However, only fibrinogen and serum creatinine had a statistically significant difference in predicting adiponectin levels between diabetic and nondiabetic persons in this analysis.

Next, the relation of renal function to adiponectin was further explored. Given that gender, renal function, and obesity-related measurements are part of glomerular filtration rate estimation equations, gender, serum creatinine, waist/hip ratio, and computed tomography visceral fat measurements were removed from the final model to determine whether the glomerular filtration rate (both GRFCG and GFRMDRD) was associated with log adiponectin. Likewise, given the high correlation between the albumin excretion rate and the glomerular filtration rate, the albumin excretion rate was also removed. GFRCG was a highly significant correlate of log adiponectin in diabetic persons (r = –0.17; p < 0.0001), and it was also significant in nondiabetic persons (r = –0.08; p = 0.03). Likewise, a similar analysis was performed for GFRMDRD, which was significant for diabetic (r = –0.14; p = 0.0001) but not for nondiabetic (r = –0.05; p = 0.22) persons. Although few nondiabetic persons had impaired renal function resulting in decreased power to detect differences by these strata, no significant differences were found after stratifying by albuminuria status (yes/no) and by glomerular filtration rate (<60 and <30 ml/min per 1.73 m2) (Web figures 1–4). (This information is illustrated in four supplementary figures; each is referred to as "Web figure" in the text and is posted on the Journal's website (http://aje.oxfordjournals.org/).)

Finally, adjustment for all of the variables in the final multiple linear regression model except for hemoglobin A1c explained 40.7 percent of the unadjusted difference in adiponectin levels between diabetic and nondiabetic persons (table 4). We first compared adiponectin levels adjusted for age and sex and then sequentially added variables from our final model in order of the percentage of the difference explained. HDL cholesterol (15.1 percent), albumin excretion rate (9.3 percent), PAI-1 (9.0 percent), and race (6.4 percent) explained 39.8 percent of the difference in adiponectin in addition to age and sex. The additional variables in our final model explained only an additional 0.9 percent of the difference. Finally, as there is little overlap in hemoglobin A1c levels between the two groups, hemoglobin A1c was added to the final model and explained a further 26.3 percent of the difference; together, these variables explained a total of 67 percent of the difference.


   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
References
 
Our data confirm previous reports that adiponectin is higher in persons with type 1 diabetes than in nondiabetic persons, even after controlling for renal function, obesity-related measurements, and HDL cholesterol. The main new finding reported is that we explain 67 percent of the difference in cross-sectional adiponectin levels between type 1 diabetic and nondiabetic persons. Although elevated adiponectin has been previously reported in type 1 diabetes (10–16), this is the largest type 1 diabetes cohort to date and has an appropriate nondiabetic control group. We confirm univariate correlations between adiponectin and gender, obesity-related measures, and renal function in both type 1 diabetic and nondiabetic persons. We then systematically analyzed whether various covariates explain the difference in adiponectin levels between type 1 and nondiabetic persons. Variables in our final multiple linear regression model explained 41 percent of the difference in adiponectin between type 1 and nondiabetic persons. The addition of hemoglobin A1c increased the percentage of difference in adiponectin explained by the model to 67 percent.

Previous publications of cross-sectional data have reported increased adiponectin levels in type 1 persons with nephropathy (16). In cross-sectional data, increases in adiponectin as a response to vascular insults in persons with type 1 diabetes have been suggested (13–16). Although one study had an external control group, differences in adiponectin by diabetes status beyond simple correlation coefficients were not examined (15). In another, the control group confirmed that adiponectin is higher in persons with type 1 diabetes, although further analysis to explore the etiologies of the differences was not performed (14). The one paper that explored the mechanism for increased adiponectin levels in persons with type 1 diabetes did not have a control group to elucidate differences by diabetes status (13). Therefore, our results extend previous studies by using statistical modeling to explain the differences in cross-sectional determinants of adiponectin in a large cohort of persons with type 1 diabetes and nondiabetic controls.

A key determinant of adiponectin levels is intraabdominal fat (9), and differences in fat distribution and function between diabetic and nondiabetic persons could be a potential explanation of the difference in adiponectin levels. It has been hypothesized that visceral fat contributes preferentially to circulating adiponectin levels and that expanded visceral adipose tissue containing large adipocytes produces less adiponectin, resulting in lower adiponectin levels in persons with greater visceral adiposity (27). Although visceral fat was decreased in type 1 as compared with nondiabetic persons, its contribution to explaining the difference in adiponectin levels was statistically significant in our final model, but of limited magnitude.

Exogenous insulin delivery in persons with type 1 diabetes could play a role in elevated adiponectin levels, but our data did not support this. Of note, in rats with chronic poorly controlled or untreated insulin-deficient diabetes, plasma adiponectin levels are lower than in controls and increased with insulin treatment or insulin gene therapy (36). In contrast, a nonstatistically significant decrease in plasma adiponectin following a hyperinsulinemic clamp has been reported (including persons with type 1 diabetes) (11), and insulin dose was inversely related to adiponectin in our data. It is possible that the chronic insulin treatment of type 1 diabetes resulting in prolonged elevations of peripheral insulin concentrations induces a degree of adipose hyperplasia with more small adipocytes that produce greater amounts of adiponectin. The effect of insulin treatment on adiponectin in persons with type 1 diabetes requires further study.

Decreased elimination of adiponectin due to impaired renal function may also explain increased adiponectin levels in persons with type 1 diabetes. Adiponectin levels in end-stage renal disease patients with type 1 diabetes are increased compared with those with type 2 diabetes or no diabetes (12), and adiponectin levels decrease following successful kidney transplantation (37). Our data support the importance of renal function to adiponectin levels and a curious difference in serum creatinine's relation to adiponectin between persons with type 1 diabetes and nondiabetic persons. In our data, serum creatinine levels have less variability in nondiabetic persons than in those with type 1 diabetes, yet the opposite relation by diabetes status found in these data requires further study. Although this would support the hypothesis that adiponectin increases in response to microvascular insults in persons with type 1 diabetes, the inverse relation in nondiabetic persons is less clear. Mechanistic explanations regarding how adipose tissue is signaled by vascular disease and thereby increases adiponectin production are needed.

As renal function has been proposed as an important determinant of adiponectin levels, its accurate assessment is of great importance, and estimating glomerular filtration rate and its limitations have been recently reviewed (35). Therefore, we utilized both the Cockcroft-Gault and Modification of Diet in Renal Disease equations to estimate glomerular filtration rate, as well as serum creatinine and urinary albumin excretion to estimate renal function. The estimated glomerular filtration rate by both equations had a similar inverse relation to adiponectin levels (Web figures 1–4). The relation of adiponectin to GFRCG could be biased (as body weight is included in the GFRCG equation); in our data, the albumin excretion rate explained 9.3 percent of the difference in adiponectin levels between diabetic and nondiabetic persons.

Adiponectin has also been shown to be highly correlated with triglycerides and HDL cholesterol (9), and mechanistically an association with lipoprotein lipase (20) and an inverse relation with hepatic lipase activity (21) have been described. In our cross-sectional data, HDL cholesterol was the strongest correlate of adiponectin and explained 15.1 percent of the difference in adiponectin levels between persons with type 1 diabetes and nondiabetic persons. This may reflect a common role of these two lipases on HDL cholesterol metabolism and adiponectin or a common etiologic factor influencing both HDL cholesterol and adiponectin. Adiponectin is known to have a beneficial effect on intravascular inflammatory insults (22); however, adiponectin's relation to markers of inflammation (C-reactive protein) and thrombotic markers (PAI-1, fibrinogen) varied in these data. PAI-1 levels were higher in nondiabetic than in diabetic persons and explained 9.0 percent of the difference in adiponectin levels. PAI-1 has been shown to be inversely related to serum adiponectin in overweight and obese women (38). Racial and ethnic differences in adiponectin have been reported (39). Race entered our final model and explained an additional 6.4 percent of the difference in adiponectin, since our diabetic cohort has fewer persons who are not non-Hispanic White than our nondiabetic cohort.

Adiponectin is inversely correlated with insulin resistance (3, 9), although determination of insulin resistance in type 1 diabetes is challenging. For a large epidemiologic study, the daily insulin dose and the EGDR (32) remain the only established markers of insulin resistance. The daily insulin dose (in units/kilogram) in diabetic persons was inversely associated with adiponectin levels, but 1/EGDR was not. Medications have also been reported to affect adiponectin levels (40). Angiotensin-converting enzyme/angiotensin receptor blocker use was correlated with adiponectin in our univariate (but not multivariate) analysis, and no persons were taking peroxisome proliferator-activated receptor gamma agonists.

The effect of hemoglobin A1c on adiponectin levels differed univariately by diabetes status. In multivariate analysis, hemoglobin A1c was not significant in nondiabetic persons when controlling for obesity-related measures, but hemoglobin A1c was significant in persons with type 1 diabetes adjusting for the same variables. As there is little overlap in hemoglobin A1c between persons with type 1 diabetes and nondiabetic persons, interpretation of this interaction requires caution. Nevertheless, further adjustments for hemoglobin A1c to our final model explained an additional 26 percent of the difference in adiponectin between persons with type 1 diabetes and nondiabetic persons. While elevation of total adiponectin in type 1 diabetes might be due to differential effects on adiponectin multimers, posttranslational modifications of adiponectin (such as glycosylation) may be altered in type 1 diabetes and could potentially affect adiponectin's activity (such as its antiinflammatory, cardioprotective qualities). Exposure of differentiated adipocytes to high glucose (15 mmol) led to an altered posttranslational modification of secreted adiponectin (28), and interference with normal adiponectin hydroxylation/glycosylation in adiponectin-null mice decreases its ability to stimulate phosphorylation of adenosine monophosphate-activated protein kinase in liver tissue (29). Increased adiponectin in persons with type 1 diabetes could be a compensatory mechanism to reduce hepatic glucose output in the hyperglycemic milieu of type 1 diabetes. Due to collinearity, further study is needed to clarify whether the effect of hemoglobin A1c is that of hyperglycemia or whether hemoglobin A1c is simply acting as a marker of type 1 diabetes and other pathophysiologic mechanisms.

Although this is the largest type 1 diabetes cohort reported on for adiponectin levels, some limitations should be mentioned. These data are cross-sectional, and therefore cause and effect cannot be definitely determined. We do not currently have data on other important potential determinants of adiponectin levels, such as vascular cell adhesion molecule-1 and tumor necrosis factor-alpha (13), or genetic data on various haplotypes that have been shown to be related to adiponectin levels. Furthermore, controversy exists regarding which isoform of adiponectin is the biologically active form of the protein (41, 42) and, at the time this study was initiated, no commercially available assays to measure isoforms had been validated in persons with type 1 diabetes. It is possible that high-molecular-weight isoforms may be more metabolically relevant and different in type 1 diabetes.

In conclusion, we report that adiponectin is elevated in persons with type 1 diabetes compared with nondiabetic persons and that 41 percent of this difference is explained by the variables in our final model, with the addition of hemoglobin A1c to the final model explaining a cumulative 67 percent of the difference. Further studies including longitudinal data are needed to better understand the role of adiponectin in persons with type 1 diabetes and its relation to patient outcomes.


   ACKNOWLEDGMENTS
 
Support for this study was provided by National Heart, Lung, and Blood Institute grant R01 HL61753 from the National Institutes of Health and by the Diabetes and Endocrinology Research Center Clinical Investigation Core grant P30 DK57516. The study was performed at the Adult General Clinical Research Center at the University of Colorado at Denver and the Health Sciences Center supported by grant M01 RR00051 from the National Institutes of Health, at the Barbara Davis Center for Childhood Diabetes, and at the Colorado Heart Imaging Center in Denver, Colorado.

Conflict of interest: none declared.


   References
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 References
 

  1. Arita Y, Kihara S, Ouchi N, et al. Paradoxical decrease of an adipose-specific protein, adiponectin, in obesity. Biochem Biophys Res Commun (1999) 257:79–83.[CrossRef][ISI][Medline]
  2. Hotta K, Funahashi T, Arita Y, et al. Plasma concentrations of a novel, adipose-specific protein, adiponectin, in type 2 diabetic patients. Arterioscler Thromb Vasc Biol (2000) 20:1595–9.[Abstract/Free Full Text]
  3. Weyer C, Funahashi T, Tanaka S, et al. Hypoadiponectinemia in obesity and type 2 diabetes: close association with insulin resistance and hyperinsulinemia. J Clin Endocrinol Metab (2001) 86:1930–5.[Abstract/Free Full Text]
  4. Matsubara M, Maruoka S, Katayose S. Decreased plasma adiponectin concentrations in women with dyslipidemia. J Clin Endocrinol Metab (2002) 87:2764–9.[Abstract/Free Full Text]
  5. Nishizawa H, Shimomura I, Kishida K, et al. Androgens decrease plasma adiponectin, an insulin-sensitizing adipocyte-derived protein. Diabetes (2002) 51:2734–41.[Abstract/Free Full Text]
  6. Diez JJ, Iglesias P. The role of the novel adipocyte-derived hormone adiponectin in human disease. Eur J Endocrinol (2003) 148:293–300.[Abstract]
  7. Chandran M, Phillips SA, Ciaraldi T, et al. Adiponectin: more than just another fat cell hormone? Diabetes Care (2003) 26:2442–50.[Free Full Text]
  8. Ouchi N, Kihara S, Funahashi T, et al. Reciprocal association of C-reactive protein with adiponectin in blood stream and adipose tissue. Circulation (2003) 107:671–4.[Abstract/Free Full Text]
  9. Cnop M, Havel PJ, Utzschneider KM, et al. Relationship of adiponectin to body fat distribution, insulin sensitivity and plasma lipoproteins: evidence for independent roles of age and sex. Diabetologia (2003) 46:459–69.[ISI][Medline]
  10. Imagawa A, Funahashi T, Nakamura T, et al. Elevated serum concentration of adipose-derived factor, adiponectin, in patients with type 1 diabetes. Diabetes Care (2002) 25:1665–6.[Free Full Text]
  11. Perseghin G, Lattuada G, Danna M, et al. Insulin resistance, intramyocellular lipid content and plasma adiponectin in patients with type 1 diabetes. Am J Physiol Endocrinol Metab (2003) 285:1174–81.
  12. Stenvinkel P, Marchlewska A, Pecoits-Filho R, et al. Adiponectin in renal disease: relationship to phenotype and genetic variation in the gene encoding adiponectin. Kidney Int (2004) 65:274–81.[CrossRef][ISI][Medline]
  13. Schalkwijk CG, Chaturvedi N, Schram MT, et al. Adiponectin is inversely associated with renal function in type 1 diabetic patients. J Clin Endocrinol Metab (2006) 91:129–35.[Abstract/Free Full Text]
  14. Frystyk J, Tarnow L, Hansen TK, et al. Increased serum adiponectin levels in type 1 diabetic patients with microvascular complications. Diabetologia (2005) 48:1911–18.[CrossRef][ISI][Medline]
  15. Hadjadj S, Aubert R, Fumeron F, et al. Increased plasma adiponectin concentrations are associated with microangiopathy in type 1 diabetic subjects. Diabetologia (2005) 48:1088–92.[CrossRef][ISI][Medline]
  16. Saraheimo M, Forsblom C, Fagerudd J, et al. Serum adiponectin is increased in type 1 diabetic patients with nephropathy. Diabetes Care (2005) 28:1410–14.[Abstract/Free Full Text]
  17. Okamoto Y, Arita Y, Nishida M, et al. An adipocyte-derived plasma protein, adiponectin, adheres to injured vascular walls. Horm Metab Res (2000) 32:47–50.[ISI][Medline]
  18. Ouchi N, Kihara S, Arita Y, et al. Novel modulator for endothelial adhesion molecules: adipocyte-derived plasma protein adiponectin. Circulation (1999) 100:2473–6.[Abstract/Free Full Text]
  19. Kubota N, Terauchi Y, Yamauchi T, et al. Disruption of adiponectin causes insulin resistance and neointimal formation. J Biol Chem (2002) 277:25863–6.[Abstract/Free Full Text]
  20. von Eynatten M, Schneider JG, Humpert PM, et al. Decreased plasma lipoprotein lipase in hypoadiponectinemia: an association independent of systemic inflammation and insulin resistance. Diabetes Care (2004) 27:2925–9.[Abstract/Free Full Text]
  21. Schneider JG, von Eynatten M, Schiekofer S, et al. Low plasma adiponectin levels are associated with increased hepatic lipase activity in vivo. Diabetes Care (2005) 28:2181–6.[Abstract/Free Full Text]
  22. Matsuda M, Shimomura I, Sata M, et al. Role of adiponectin in preventing vascular stenosis. The missing link of adipo-vascular axis. J Biol Chem (2002) 277:37487–91.[Abstract/Free Full Text]
  23. Maahs DM, Ogden LG, Kinney GL, et al. Low plasma adiponectin levels predict progression of coronary artery calcification. Circulation (2005) 111:747–53.[Abstract/Free Full Text]
  24. Pischon T, Girman CJ, Hotamisligil GS, et al. Plasma adiponectin levels and risk of myocardial infarction in men. JAMA (2004) 291:1730–7.[Abstract/Free Full Text]
  25. Costacou T, Zgibor JC, Evans RW, et al. The prospective association between adiponectin and coronary artery disease among individuals with type 1 diabetes. The Pittsburgh Epidemiology of Diabetes Complications Study. Diabetologia (2005) 48:41–8.[CrossRef][ISI][Medline]
  26. Zoccali C, Mallamaci F, Panuccio V, et al. Adiponectin is markedly increased in patients with nephrotic syndrome and is related to metabolic risk factors. Kidney Int Suppl (2003) (84):S98–102.
  27. Havel PJ. Update on adipocyte hormones: regulation of energy balance and carbohydrate/lipid metabolism. Diabetes (2004) 53(suppl 1):S143–51.[Abstract/Free Full Text]
  28. Richards AA, Stephens T, Charlton HK, et al. Adiponectin multimerization is dependent on conserved lysines in the collagenous domain: evidence for regulation of multimerization by alterations in posttranslational modifications. Mol Endocrinol (2006) 20:1673–87.[Abstract/Free Full Text]
  29. Wang Y, Lam KS, Chan L, et al. Post-translational modifications of the four conserved lysine residues within the collagenous domain of adiponectin are required for the formation of its high molecular weight oligomeric complex. J Biol Chem (2006) 281:16391–400.[Abstract/Free Full Text]
  30. Maahs DM, Kinney GL, Wadwa P, et al. Hypertension prevalence, awareness, treatment, and control in an adult type 1 diabetes population and a comparable general population. Diabetes Care (2005) 28:301–6.[Abstract/Free Full Text]
  31. Dabelea D, Kinney G, Snell-Bergeon JK, et al. Effect of type 1 diabetes on the gender difference in coronary artery calcification: a role for insulin resistance?: The Coronary Artery Calcification in Type 1 Diabetes (CACTI) Study. Diabetes (2003) 52:2833–9.[Abstract/Free Full Text]
  32. Williams KV, Erbey JR, Becker D, et al. Can clinical factors estimate insulin resistance in type 1 diabetes? Diabetes (2000) 49:626–32.[Abstract]
  33. Cockcroft DW, Gault MH. Prediction of creatinine clearance from serum creatinine. Nephron (1976) 16:31–41.[ISI][Medline]
  34. Levey AS, Bosch JP, Lewis JB, et al. A more accurate method to estimate glomerular filtration rate from serum creatinine: a new prediction equation. Modification of Diet in Renal Disease Study Group. Ann Intern Med (1999) 130:461–70.[Abstract/Free Full Text]
  35. Stevens LA, Coresh J, Greene T, et al. Assessing kidney function—measured and estimated glomerular filtration rate. N Engl J Med (2006) 354:2473–83.[Free Full Text]
  36. Thule PM, Campbell AG, Kleinhenz DJ, et al. Hepatic insulin gene therapy prevents deterioration of vascular function and improves adipocytokine profile in STZ-diabetic rats. Am J Physiol Endocrinol Metab (2006) 290:114–22.[CrossRef]
  37. Chudek J, Adamczak M, Karkoszka H, et al. Plasma adiponectin concentration before and after successful kidney transplantation. Transplant Proc (2003) 35:2186–9.[CrossRef][ISI][Medline]
  38. Mertens I, Ballaux D, Funahashi T, et al. Inverse relationship between plasminogen activator inhibitor-I activity and adiponectin in overweight and obese women. Interrelationship with visceral adipose tissue, insulin resistance, HDL-chol and inflammation. Thromb Haemost (2005) 94:1190–5.[ISI][Medline]
  39. Hulver MW, Saleh O, MacDonald KG, et al. Ethnic differences in adiponectin levels. Metabolism (2004) 53:1–3.[CrossRef][ISI][Medline]
  40. Furuhashi M, Ura N, Higashiura K, et al. Blockade of the renin-angiotensin system increases adiponectin concentrations in patients with essential hypertension. Hypertension (2003) 42:76–81.[Abstract/Free Full Text]
  41. Pajvani UB, Du X, Combs TP, et al. Structure-function studies of the adipocyte-secreted hormone Acrp30/adiponectin. Implications for metabolic regulation and bioactivity. J Biol Chem (2003) 278:9073–85.[Abstract/Free Full Text]
  42. Bluher M, Brennan AM, Kelesidis T, et al. Total and high-molecular weight adiponectin in relation to metabolic variables at baseline and in response to an exercise treatment program: comparative evaluation of three assays. Diabetes Care (2007) 30:280–5.[Abstract/Free Full Text]

Add to CiteULike CiteULike   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us    What's this?


This article has been cited by other articles:


Home page
 DiabetesHome page
V.-P. Makinen, C. Forsblom, L. M. Thorn, J. Waden, D. Gordin, O. Heikkila, K. Hietala, L. Kyllonen, J. Kyto, M. Rosengard-Barlund, et al.
Metabolic Phenotypes, Vascular Complications, and Premature Deaths in a Population of 4,197 Patients With Type 1 Diabetes
Diabetes, September 1, 2008; 57(9): 2480 - 2487.
[Abstract] [Full Text] [PDF]

Home page
 Diabetes CareHome page
M. Saraheimo, C. Forsblom, L. Thorn, J. Waden, M. Rosengard-Barlund, O. Heikkila, K. Hietala, D. Gordin, J. Frystyk, A. Flyvbjerg, et al.
Serum Adiponectin and Progression of Diabetic Nephropathy in Patients With Type 1 Diabetes
Diabetes Care, June 1, 2008; 31(6): 1165 - 1169.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow Figures
Right arrow All Versions of this Article:
166/6/731    most recent
kwm125v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrowRequest Permissions
Right arrow Disclaimer
Google Scholar
Right arrow Articles by Maahs, D. M.
Right arrow Articles by Rewers, M.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Maahs, D. M.
Right arrow Articles by Rewers, M.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?