- Email: [email protected]
Obesity Is Associated With Family History of ESRD in Incident Dialysis Patients
Obesity Is Associated With Family History of ESRD in Incident Dialysis Patients Rebecca A. Speckman, BA, William M. McClellan, MD, MPH, Nataliya V. Volkova, BA, MPH, Claudine T. Jurkovitz, MD, MPH, Scott G. Satko, MD, Anton C. Schoolwerth, MD, MSHA, and Barry I. Freedman, MD ● Background: Obesity is an established risk factor for chronic kidney disease and aggregates in families. The objective of this study is to examine the relationship between obesity and family history of end-stage renal disease (ESRD). Methods: Data were collected from 25,883 incident patients with ESRD in US ESRD Network 6 (Georgia, North Carolina, and South Carolina) dialysis clinics between 1995 and 2003. Family history is defined as a first- or second-degree relative with ESRD. Body mass index (BMI) at dialysis therapy initiation was classified as underweight (BMI < 18.5 kg/m2), normal (BMI, 18.5 to <25 kg/m2), overweight (BMI, 25 to < 30 kg/m2), obese (BMI, 30 to <35 kg/m2), or morbidly obese (BMI > 35 kg/m2). Results: Twenty-three percent of patients reported a family history of ESRD. Of patients reporting a family history of ESRD, 5.5% were underweight, 32.5% had normal BMI, 28.0% were overweight, 17.3% were obese, and 16.7% were morbidly obese. After controlling for age, race, sex, primary cause of ESRD, history of diabetes, history of hypertension, and estimated glomerular filtration rate at dialysis therapy initiation, reported family history of ESRD was associated with being overweight (odds ratio [OR], 1.17; 95% confidence interval [CI], 1.08 to 1.26), obese (OR, 1.25; 95% CI, 1.14 to 1.37), and morbidly obese (OR, 1.40; 95% CI, 1.27 to 1.55). Conclusion: Obesity at dialysis therapy initiation was associated independently with reported family history of ESRD. This finding suggests that behavioral factors, adiposity-related genes, and gene-by-BMI interaction may contribute to familial risk for ESRD. This finding also suggests that management of obesity may be even more important for patients with a family history of ESRD than for the general population. Am J Kidney Dis 48:50-58. © 2006 by the National Kidney Foundation, Inc. INDEX WORDS: End-stage renal disease; family history; familial clustering; obesity.
F
AMILY HISTORY OF chronic kidney disease (CKD) is a risk factor for end-stage renal disease (ESRD) in African Americans and whites in the United States.1-3 Familial clustering of CKD also has been observed worldwide.4-7 Causes of familial aggregation of CKD are poorly understood; however, there is considerable evidence that genetic factors may predispose family members to ESRD. It also is likely that shared environmental exposures and gene-
environment interactions contribute to the observed clustering of ESRD within families. To unravel these complex relationships, it is important to identify shared personal characteristics and environmental exposures among family members at high risk for ESRD. One such shared characteristic among family members may be obesity. A possible association between high body mass index (BMI) and increased risk for nephropathy
From the Department of Epidemiology, Rollins School of Public Health, Emory University, Atlanta, GA; Department of Internal Medicine, Section on Nephrology, Wake Forest University School of Medicine, Winston-Salem, NC; Centers for Disease Control and Prevention, Atlanta, GA; and Section of Hypertension/Nephrology, Dartmouth Medical School, Hanover, NH. Received November 7, 2005; accepted in revised form March 23, 2006. Work done at Department of Epidemiology, Rollins School of Public Health, Emory University, Atlanta, GA. Support: None. Potential conflicts of interest: None. Presented in abstract form (F-PO321) at the American Society of Nephrology Renal Week, St Louis, MO, October 27-November 1, 2004. The analyses upon which this publication is based were performed under Contract No. 500-97-E024, entitled “EndStage Renal Disease Networks Organization #6,” sponsored by the Centers for Medicare & Medicaid Services. The content of this publication does not necessarily reflect the
views or policies of the Centers for Medicare & Medicaid Services, nor does mention of trade names, commercial products, or organizations imply endorsement by the US government. The authors assume full responsibility for the accuracy and completeness of the ideas presented. This article is a direct result of the Health Care Quality Improvement Program initiated by the Centers for Medicare & Medicaid Services, which has encouraged identification of quality improvement projects derived from analysis of patterns of care, and therefore required no special funding on the part of this contractor. Ideas and contributions to the authors concerning experience in engaging with issues presented are welcomed. Address reprint requests to Rebecca A. Speckman, BA, Department of Epidemiology, Rollins School of Public Health, Emory University, 1518 Clifton Road NE, Atlanta, GA 30322. E-mail: [email protected] © 2006 by the National Kidney Foundation, Inc. 0272-6386/06/4801-0006$32.00/0 doi:10.1053/j.ajkd.2006.03.086
50
American Journal of Kidney Diseases, Vol 48, No 1 (July), 2006: pp 50-58
OBESITY AND FAMILY HISTORY OF ESRD
51
is suggested by several clinical observations. Extreme obesity has been associated repeatedly with enhanced risk for progressive kidney disease.8-11 An association between obesity, BMI, and increased risk for kidney disease was reported in several racial and ethnic groups,12-14 and familial clustering of obesity has been reported widely.15-20 Finally, a recent genome scan in hypertensive African Americans with ESRD found strong evidence for linkage between ESRD and polymorphic markers on chromosome 13q in families with the highest BMI.21 The objective of this study is to determine whether an association exists between reported family history of ESRD and BMI among incident dialysis patients. METHODS
Participants The Family History of ESRD Study population included patients who initiated Medicare-supported renal replacement therapy in Network 6 dialysis facilities (serving North Carolina, South Carolina, and Georgia). Participation by treatment centers and patients was voluntary. The study and data-collection instruments were approved for use by the Wake Forest University School of Medicine Institutional Review Board (Winston-Salem, NC). Network 6 facilities used a standardized data-collection instrument that included the presence of ESRD in first- and second-degree relatives, total number of siblings and children, and their vital status (alive, deceased, unknown). Age, sex, race, weight, and height of patients; primary cause of ESRD; comorbidities; and laboratory results at the initiation of dialysis therapy were obtained from Centers for Medicare & Medicaid Services (CMS) Form 2728.2 The study population included 59,167 incident dialysis patients from ESRD Network 6 (North Carolina, South Carolina, Georgia) facilities who initiated dialysis therapy between 1995 and 2003. Patients residing in other states were excluded (n ⫽ 703), as were patients with ESRD caused by known Mendelian disorders (including polycystic kidney disease and Alport syndrome), urological conditions/ surgical nephrectomy, or missing data on primary cause of ESRD (n ⫽ 1,192). Of the remaining patients, 25,883 (46.28%) consented and provided family history information. Patients younger than 20 years or with a recorded race other than black or white were excluded (n ⫽ 1,055), and patients with unrealistic recorded height and weight measurements (height ⬍ 120 cm or ⬎ 206 cm, weight ⬍ 30 kg or ⬎ 300 kg) or with a missing serum creatinine concentration also were excluded, leaving a total of 23,822 patients eligible for BMI analysis (Fig 1).
Data Collection After obtaining consent, the treatment centers’ social workers or nursing staff used the standardized instrument to screen for history of ESRD in first- and second-degree
Fig 1.
Patients included in analysis.
relatives of incident dialysis patients. A family history was considered present if an incident patient reported having either a first-degree (parent, sibling, or child) or seconddegree (grandparent, aunt, uncle, grandchild, or half-sibling) relative with ESRD. ESRD was described to participants as “end-stage renal disease or kidney failure (ie, dialysis, kidney transplant, or death from kidney disease before dialysis was available or before starting dialysis).” De-identified data from the family history questionnaire were collated with information obtained from CMS Form 2728 for age, sex, race, primary cause of ESRD, comorbidities present at the initiation of dialysis therapy, and clinical characteristics at the initiation of dialysis therapy. BMI was calculated from the height and dry weight collected on CMS Form 2728. Primary cause of ESRD was classified as diabetes mellitus (DM), hypertension (HTN), primary glomerulonephritis, and “other.” The primary outcome was a reported family history of ESRD. BMI was classified by using the World Health Organization classifications: underweight (BMI ⬍ 18.5 kg/m2), normal (BMI, 18.5 to ⬍25 kg/m2), overweight (BMI, 25 to ⬍ 30 kg/m2), obese (BMI, 30 to ⬍ 35 kg/m2), and morbidly obese (BMI ⱖ 35 kg/m2).22 During the years of this study, CMS Form 2728 categories for history of DM were: “Diabetes, currently on insulin” and “Diabetes, primary or contributing.” Because these categories were not mutually exclusive, we created 4 categories for diabetes: “No DM,” “DM, currently on insulin (only),”
52
SPECKMAN ET AL
“DM, primary or contributing (only),” and “DM, both ‘currently on insulin’ and ‘primary or contributing.’” Glomerular filtration rate (GFR; milliliters per minute per 1.73 m2) was calculated from the modified Modification of Diet in Renal Disease formula.23
Statistical Analyses Associations in categorical variables were examined by using chi-square tests and Mantel-Haenszel odds ratios (ORs). Continuous variables were examined by using t-tests, analysis of variance (using Tukey test for pairwise comparisons), and Pearson correlation coefficient. Logistic regression was performed to assess the association between BMI classification and reported family history of ESRD. We adjusted for associations of other patient characteristics with a family history of ESRD by including them as covariates in the logistic regression model. Patient characteristics included in the multivariable model were age, sex, race, history of DM, history of HTN, cause of ESRD, history of smoking, number of first-degree relatives, and estimated GFR. Interactions between race and sex and between BMI class and the other covariates in the model were examined in the logistic regression model by evaluating the significance of cross-product terms. All analyses were performed using SAS, version 9.1 (SAS Institute, Cary, NC).
RESULTS
A family history of ESRD was reported by 22.9% of patients included in this analysis (Table 1). Compared with those who reported no family history of ESRD, patients reporting a family history of ESRD had a greater average BMI (28.2 versus 26.6 kg/m2; P ⬍ 0.001), were younger (mean age, 57.7 versus 61.2 years; P ⬍ 0.001), had more first-degree relatives (8.7 versus 7.6; P ⬍ 0.001), were more likely to be black (74.5% versus 52.2%; P ⬍ 0.001), and were more likely to be female (56.1% versus 48.7%; P ⬍ 0.001). A history of HTN was associated with a family history of ESRD in crude analysis (Table 1), as was a history of DM. A family history of ESRD was not associated with a pre-ESRD history of smoking. Patients with primary causes of ESRD other than glomerulonephritis or HTN were less likely than patients with DM as the primary cause of ESRD to report a family history of ESRD (OR, 0.65; 95% confidence interval [CI], 0.59 to 0.71). Patients with a positive family history had a statistically significant, but clinically minor, difference in average predialysis GFR (8.0 versus 8.5 mL/min/1.73 m2 [0.13 versus 0.14 mL/s/1.73 m2]; P ⬍ 0.001). There was a high prevalence of obesity among incident patients; 6.7% of patients were underweight, 37.9%
had a normal BMI, 27.6% were overweight, 15.2% were obese, and 12.5% were morbidly obese. There was a strong association between BMI and family history of ESRD (Table 1; Fig 2). Patients with a reported family history of ESRD were more likely to be overweight (28.0% versus 27.5%), obese (17.3% versus 14.6%), or morbidly obese (16.7% versus 11.3%; Table 1). The relationship between BMI category and reported family history of ESRD was similar when stratified by race, and an interaction term between race and BMI class was nonsignificant in the multivariable model, indicating that the association between BMI class and family history of ESRD did not differ between whites and blacks. BMI was associated with all covariates. Mean BMI was 27.5 ⫾ 7.6 (SD) kg/m2 in blacks and 26.3 ⫾ 6.8 kg/m2 in whites (P ⬍ 0.001). In women, mean BMI was 27.6 ⫾ 7.9 kg/m2 compared with 26.3 ⫾ 6.6 kg/m2 in men (P ⬍ 0.001). Mean BMI was 28.2 ⫾ 8.2 kg/m2 for those younger than 55 years, 28.2 ⫾ 7.4 kg/m2 for those aged 55 to younger than 65 years, 26.4 ⫾ 6.4 kg/m2 for those aged 65 to younger than 75 years, and 24.6 ⫾ 5.6 kg/m2 for those 75 years and older (P ⬍ 0.001; all pairwise comparisons significant except between the 2 age groups). Mean BMI by primary cause was 28.5 ⫾ 7.3 kg/m2 for DM, 27.5 ⫾ 8.1 kg/m2 for glomerulonephritis, 25.9 ⫾ 7.1 kg/m2 for HTN, and 24.8 ⫾ 6.4 kg/m2 for other causes (P ⬍ 0.001; all pairwise comparisons significant). Mean BMI by history of DM category was 27.5 ⫾ 7.5 kg/m2 for patients with DM administered oral medications, 27.3 ⫾ 7.2 kg/m2 for patients with DM administered no medications, 26.3 ⫾ 6.6 kg/m2 for patients with DM administered insulin, and 25.5 ⫾ 6.9 kg/m2 for patients with no DM (P ⬍ 0.001; all pairwise comparisons significant except between patients with DM administered oral medications and patients with DM administered no medications). BMI correlated weakly with GFR (Pearson r ⫽ ⫺0.04383; P ⬍ 0.001). The association between family history and BMI classification persisted after adjustment for age, race, sex, pre-ESRD history of HTN, preESRD history of DM, primary cause of ESRD, history of smoking, number of first-degree relatives, and GFR (Table 2). Overweight patients had a 17% greater odds of reporting a family history of ESRD compared with underweight
OBESITY AND FAMILY HISTORY OF ESRD
53
Table 1. Characteristics of Patients With and Without a Reported Family History of ESRD Reported Family History of ESRD
Characteristic
BMI (kg/m2) Age (y) No. of first-degree relatives Age categories (y) ⬍55 55 to ⬍65 65 to ⬍75 ⱖ75 Sex Male Female Race Black White BMI category (kg/m2) Underweight (BMI ⬍ 18.5) Normal (BMI, 18.5-24.9) Overweight (BMI, 25-29.9) Obese (BMI, 30-34.9) Morbidly obese (BMI ⱖ 35) History of HTN No Yes History of DM No DM DM, currently on insulin DM, primary or contributing DM, both “currently on insulin” and “primary or contributing” History of smoking No Yes ESRD primary cause DM HTN Glomerulonephritis Other GFR (mL/min/1.73 m2)
Yes (n ⫽ 5,453)
No (n ⫽ 18,369)
OR (95% CI)*
P†
28.2 ⫾ 8.0 57.7 ⫾ 14.7 8.7 ⫾ 5.3
26.6 ⫾ 7.1 61.2 ⫾ 15.0 7.6 ⫾ 5.1
— — —
⬍0.001 ⬍0.001 ⬍0.001
2,224 (40.8) 1,343 (24.6) 1,171 (21.5) 715 (13.1)
5,800 (31.6) 4,038 (22.0) 4,922 (26.8) 3,609 (19.8)
Reference 0.87 (0.80-0.94) 0.62 (0.57-0.67) 0.52 (0.47-0.57)
⬍0.001
2,394 (43.9) 3,059 (56.1)
9,431 (51.3) 8,938 (48.7)
Reference 1.35 (1.27-1.43)
⬍0.001
4,065 (74.5) 1,388 (25.5)
9,580 (52.2) 8,789 (47.8)
2.69 (2.51-2.87) Reference
302 (5.5) 1,771 (32.5) 1,526 (28.0) 945 (17.3) 909 (16.7)
1,297 (7.1) 7,266 (39.6) 5,058 (27.5) 2,679 (14.6) 2,069 (11.3)
0.96 (0.83-1.09) Reference 1.24 (1.15-1.33) 1.45 (1.32-1.58) 1.80 (1.64-1.98)
752 (13.8) 4,701 (86.2)
3,083 (16.8) 15,286 (83.2)
Reference 1.26 (1.16-1.37)
⬍0.001
2,403 (44.1) 386 (7.1) 1,438 (26.4) 1,226 (22.5)
8,771 (47.8) 1,145 (6.2) 4,713 (25.7) 3,740 (20.4)
Reference 1.23 (1.09-1.39) 1.11 (1.04-1.20) 1.20 (1.11-1.29)
⬍0.001
5,004 (91.8) 449 (8.2)
16,740 (91.1) 1,629 (8.9)
Reference 0.92 (0.83-1.03)
2,757 (50.6) 1,519 (27.9) 519 (9.5) 658 (12.1) 8.0 ⫾ 4.7
8,522 (46.4) 4,945 (26.9) 1,755 (9.6) 3,147 (17.13) 8.5 ⫾ 4.6
Reference 0.95 (0.88-1.02) 0.91 (0.82-1.02) 0.65 (0.59-0.71) —
⬍0.001 ⬍0.001
0.145 ⬍0.001
⬍0.001
NOTE. N ⫽ 23,822. Values expressed as mean ⫾ SD or number (percent). To convert GFR in mL/min to mL/s, multiply by 0.01667. *Mantel-Haenszel ORs with 95% CIs. †Chi-square for categorical variables (race, sex, ESRD cause); independent sample t-tests for continuous variables (BMI, age, serum albumin level, GFR).
patients (OR, 1.17; 95% CI, 1.08 to 1.26). Obese patients had 25% greater odds (OR, 1.25; 95% CI, 1.12 to 1.37), and morbidly obese patients had 40% greater odds (OR, 1.40; 95% CI, 1.27 to 1.55). Younger age, black race, female sex, and number of first-degree relatives also were associated with a reported family history of ESRD in the adjusted analysis. We did not find significant interactions between BMI class and the other
covariates in the model, indicating that there was no synergistic relationship between BMI class and other covariates. DISCUSSION
Our study is the first report of an association between family history of ESRD and obesity among incident dialysis patients. The association was strong, increased in strength with more se-
54
SPECKMAN ET AL
Fig 2. Reported family history of ESRD and BMI (kg/m2).
vere degrees of obesity, and persisted after adjustment for other patient characteristics, including history of DM, history of HTN, and primary cause of ESRD. In addition, we noted that younger age, increasing number of first-degree relatives, female sex, black race, and history of HTN were associated with a family history of ESRD in the adjusted analysis. These observations warrant comment. First, it is not unexpected that patients with more firstdegree family members would be more likely to report a family history of ESRD. In addition, our finding of an association between younger age and family history of ESRD agrees with earlier findings24 and may be explained by a tendency for earlier onset in the presence of a genetic predisposition to ESRD. The association between family history of ESRD in blacks is consistent with findings from previous studies and with the well-known disparities between blacks and whites for incident ESRD. This association suggests that either as yet unidentified environmental factors that aggregate within black families and/or genetic factors may account for these differences in familial aggregation of ESRD, possibilities that are a focus of active investigation. The association between being female and having a family history of ESRD is interesting, and it is tempting to attribute it to sex-specific recall bias. Another possibility is that at least some of the familial aggregation of ESRD may
be transmitted matrilineally by either genomic imprinting or the mitochrondrial genome.25 However, information with which to address these possibilities are absent from our database, and evaluation of other family history databases will be necessary to determine the true significance of this association. Although it might be expected that the observed association between BMI and family history of ESRD is explained by the relationship of type 2 DM and HTN to obesity, it is important to note that the association remained after controlling for both history of DM and HTN. Furthermore, the association also was present in patients with glomerulonephritis as primary cause of ESRD. We found no interaction between BMI and other variables, indicating that BMI does not have a synergistic relationship with other patient characteristics, such as history of HTN or history of DM. These findings suggest that BMI may influence the familial aggregation of ESRD independently of other patient characteristics, including those associated with both ESRD and obesity. The association between increased BMI and positive family history of ESRD might be explained by 2 hypotheses that are not mutually exclusive. First, it is possible that the familial clustering of obesity, which is a risk factor for ESRD, contributes to increased risk for progressive kidney disease among family members. This possibility is supported by the familial clustering
OBESITY AND FAMILY HISTORY OF ESRD
55
Table 2. Adjusted Association Between BMI of Incident Patients With ESRD and Family History of ESRD Factor
BMI category (kg/m2) Underweight (BMI ⬍ 18.5) Normal (BMI, 18.5-24.9) Overweight (BMI, 25-29.9) Obese (BMI, 30-34.9) Morbidly obese (BMI ⱖ 35) Age group (y) ⬍55 55 to ⬍64 65 to ⬍74 ⱖ75 No. of first-degree relatives Sex Female Male Race Black White History of HTN Yes No History of DM No DM DM, currently on insulin DM, primary or contributing DM, both “currently on insulin” and “primary or contributing” History of smoking No Yes ESRD primary cause DM HTN Glomerulonephritis Other GFR (mL/min/1.73 m2)
OR (95% CI)
Wald P
0.93 (0.81-1.07) Reference 1.17 (1.08-1.26) 1.25 (1.14-1.37) 1.40 (1.27-1.55)
0.338 ⬍0.001 ⬍0.001 ⬍0.001
Reference 0.89 (0.82-0.97) 0.71 (0.65-0.77) 0.66 (0.60-0.73) 1.04 (1.03-1.05)*
0.007 ⬍0.001 ⬍0.001 ⬍0.001
1.25 (1.17-1.33) Reference
⬍0.001
2.38 (2.21-2.55) Reference
⬍0.001
1.12 (1.02-1.23) Reference
⬍0.001
Reference 1.10 (0.94-1.28) 1.01 (0.91-1.12) 1.09 (0.96-1.23)
0.230 0.876 0.184
Reference 1.01 (0.90-1.14)
0.851
Reference 0.94 (0.87-1.01) 0.97 (0.86-1.08) 0.79 (0.71-0.87) 1.00 (0.99-1.00)
0.100 0.564 ⬍0.001 0.217
NOTE. All covariates found to be significant were included in the model and are listed in this table. To convert GFR in mL/min to mL/s, multiply by 0.01667. *OR for “number of first-degree relatives” is the increase in odds for an increase in 1 first-degree relative.
of obesity and the observation that obesity is itself a risk factor for CKD and ESRD. Previous studies observed familial clustering of obesity15-20 and central obesity (abdominal visceral fat).26,27 Whereas several genetic loci have been linked to obesity,28 such shared environmental factors as energy intake, dietary composition, and energy expenditure also are likely to be partially responsible for the familial clustering of obesity.20,29 Obesity or extreme obesity also is associated with increased risk for progressive kidney disease or ESRD,8-13 and obesity is associated with increased risk for albuminuria.30 The association
between increased BMI and risk for CKD seems to hold even for the nonobese.14 Increasing BMI correlated with increased filtration fraction and hyperfiltration, which could predispose to progressive renal injury, in nonobese volunteers who did not have DM or HTN.31 The deleterious effects of obesity on the kidney are mediated in part by DM, hyperglycemia, HTN, and/or insulin resistance.32 However, obesity also may contribute to kidney injury through other mechanisms.33,34 Accumulation of visceral fat can increase intrarenal pressure and may lead to extracellular matrix production in the renal interstitium. Increased sympathetic nervous sys-
56
tem activity, possibly mediated by factors that include leptin, insulin, long-chain fatty acids, and angiotensin II, also contributes to renal dysfunction.35 Inflammatory mediators produced by such adipocytes as tumor necrosis factor ␣, interleukin 6, and C-reactive protein may contribute to glomerular and interstitial fibrosis.36 A second hypothesis that might partially explain the observed BMI–family history of ESRD association is that obesity increases the risk for ESRD in individuals with a genetic predisposition to progressive kidney injury through a geneby-environment interaction. This possibility is supported by 2 recent studies. A recent genome scan performed in 264 African-American pedigrees enriched for the presence of non-DM ESRD observed modest evidence for linkage between ESRD and chromosome 13q33.3 in initial analyses, with a maximum log of the odds score for linkage of 1.02.21 After stratification by BMI, log of the odds score increased to 5.20 in the 46% of families with the greatest BMI (P ⫽ 0.0002). Similarly, another study recently reported that an association between a functional variant of the podocin gene (NPHS2 R229Q) and microalbuminuria in the general population was the strongest in patients with increased BMI.37 It is important to note the clinical and public health relevance of these observations. Familytargeted prevention is an important strategy in the prevention of ESRD,38-40 and animal and human studies suggested that weight reduction or alteration in nutritional content early or late in the course of obesity-mediated renal dysfunction can improve renal function.41-43 Our findings suggest that strategies to control or reduce weight might be especially important in prevention efforts targeted to family members of patients with ESRD. Family-based ESRD prevention efforts, for example, the “Family Reunion Initiative!” recently launched by the National Kidney Disease Education Program,44 might integrate knowledge on dietary modification in patients with obesity-related renal disease with existing knowledge in the field of family-based obesity interventions. There are strengths and limitations to our study. It is the largest family history of ESRD data set ever compiled, and BMI information was obtained independently for all patients at the initiation of renal replacement therapy. Although there was rela-
SPECKMAN ET AL
tively low overall participation with respect to voluntarily reporting a family history of ESRD, characteristics of nonparticipants were similar to those of participants, and there was no significant difference between participants and nonparticipants in the distribution of BMI classification (P ⫽ 0.219, data not shown). A potential limitation of our study is error in reported dry weights provided on CMS Form 2728. It is possible, but unlikely, that the familial aggregation of increased BMI reflected greater degrees of edema in individuals reporting a positive family history. We attempted to account for this in part by controlling for the reported cause of ESRD and baseline residual GFR. However, it remains possible that better estimates of dry body weight might diminish the associations that we noted. Future studies of individuals with ESRD and those with less severe degrees of kidney disease who have data for dry weight and fat distribution will help resolve this issue. Another limitation of our study is that we did not have information on obesity or BMI in family members of patients with ESRD. We suggest that future studies on this topic address several points related to the hypotheses generated by our study. Information on BMI and other clinical characteristics of relatives is needed in addition to the presence or absence of ESRD. More precise information on body fat distribution also might help elucidate the cause of the observed BMI–family history of ESRD association. Similarly, information on socioeconomic status and dietary habits of patients and their relatives might allow for exploration of potential environmental causes of familial clustering. Studies of genetic factors for ESRD risk should consider effect modification by BMI or body fat distribution. Although low birth weight is not associated with familial aggregation of ESRD,45 future studies of maternal smoking might explore the possibility that obesity increases risks that are imparted by environmental exposures occurring in utero. In conclusion, we find that overweight and obesity are associated with a reported family history of ESRD among incident dialysis patients independent of age, race, sex, history of DM or HTN, or primary cause of ESRD. This association raises the possibility that obesitypromoting environmental factors shared in fami-
OBESITY AND FAMILY HISTORY OF ESRD
lies, obesity-related genes, and interaction between obesity and other genes that predispose to ESRD may contribute to the familial aggregation of risk for ESRD. This finding also suggests that within afflicted families, control of body weight may be an important target to reduce the risk for ESRD. ACKNOWLEDGMENT The authors thank Network 6 dialysis facility staff members for collecting data and the dialysis patients who participated in the study.
REFERENCES 1. Ferguson R, Grim CE, Opgenorth TJ: A familial risk of chronic renal failure among blacks on dialysis? J Clin Epidemiol 41:1189-1196, 1998 2. Freedman BI, Soucie JM, McClellan WM: Family history of end-stage renal disease among incident dialysis patients. J Am Soc Nephrol 8:1942-1945, 1997 3. Freedman BI, Volkova NV, Satko SG, et al: Populationbased screening for family history of end-stage renal disease among incident dialysis patients. Am J Nephrol 25:529-535, 2005 4. Gumprecht J, Zychma MJ, Moczulski DK, Gosek K, Grzeszczak W: Family history of end-stage renal disease among hemodialyzed patients in Poland. J Nephrol 16:511515, 2003 5. Nelson RG, Knowler WC, McCance DR, et al: Determinants of end-stage renal disease in Pima Indians with type 2 (non-insulin-dependent) diabetes mellitus and proteinuria. Diabetologia 36:1087-1093, 1993 6. Canani LH, Gerchman F, Gross JL: Familial clustering of diabetic nephropathy in Brazilian type 2 diabetic patients. Diabetes 48:909-913, 1999 7. Chandie Shaw PK, van Es LA, Paul LC, Rosendaal FR, Souverijn JH, Vandenbroucke JP: Renal disease in relatives of Indo-Asian type 2 diabetic patients with endstage diabetic nephropathy. Diabetologia 46:618-624, 2003 8. Weisinger JR, Kempson RL, Eldridge FL, Swenson RS: The nephrotic syndrome: A complication of massive obesity. Ann Intern Med 81:440-447, 1974 9. Kasiske BL, Crosson JT: Renal disease in patients with massive obesity. Arch Intern Med 146:1105-1109, 1986 10. Jennette JC, Charles L, Grubb W: Glomerulomegaly and focal segmental glomerulosclerosis associated with obesity and sleep-apnea syndrome. Am J Kidney Dis 10:470472, 1987 11. Kasiske BL, Napier J: Glomerular sclerosis in patients with massive obesity. Am J Nephrol 5:45-50, 1985 12. Stengel B, Tarver-Carr ME, Powe NR, Eberhardt MS, Brancati FL: Lifestyle factors, obesity and the risk of chronic kidney disease. Epidemiology 14:479-487, 2003 13. Fox CS, Larson MG, Leip EP, Culleton B, Wilson PW, Levy D: Predictors of new-onset kidney disease in a community-based population. JAMA 291:844-850, 2004 14. Vupputuri S, Sandler DP: Lifestyle risk factors and chronic kidney disease. Ann Epidemiol 13:712-720, 2003
57
15. Burns TL, Moll PP, Lauer RM: The relation between ponderosity and coronary risk factors in children and their relatives. The Muscatine Ponderosity Family Study. Am J Epidemiol 129:973-987, 1989 16. Ayatollahi SM: Obesity in school children and their parents in southern Iran. Int J Obes Relat Metab Disord 16:845-850, 1992 17. Guillaume M, Lapidus L, Beckers F, Lambert A, Bjorntorp P: Familial trends of obesity through three generations: The Belgian-Luxembourg child study. Int J Obes Relat Metab Disord 19:S5-S9, 1995 (suppl 3) 18. Magnusson PK, Rasmussen F: Familial resemblance of body mass index and familial risk of high and low body mass index. A study of young men in Sweden. Int J Obes Relat Metab Disord 26:1225-1231, 2002 19. Fuentes RM, Notkola IL, Shemeikka S, Tuomilehto J, Nissinen A: Familial aggregation of body mass index: A population-based family study in eastern Finland. Horm Metab Res 34:406-410, 2002 20. Mirmiran P, Mirbolooki M, Azizi F: Familial clustering of obesity and the role of nutrition: Tehran Lipid and Glucose Study. Int J Obes Relat Metab Disord 26:16171622, 2002 21. Freedman BI, Langefeld CD, Rich SS, et al: A genome scan for end-stage renal disease in African American families enriched for non-diabetic nephropathy. J Am Soc Nephrol 15:2719-2727, 2004 22. World Health Organization: Obesity: Preventing and Managing the Global Epidemic. Report from a WHO Consultation. Geneva, Switzerland, World Health Organization, WHO Technical Report Series 894, 2000 23. Levey AS, Greene T, Kusek JW, Beck GL, for the MDRD Study Group: A simplified equation to predict glomerular filtration rate from serum creatinine. J Am Soc Nephrol 11:155A, 2000 (abstr) 24. Spray BJ, Atassi NG, Tuttle AB, Freedman BI: Familial risk, age at onset, and cause of end-stage renal disease in white Americans. J Am Soc Nephrol 5:1806-1810, 1995 25. Watson B Jr, Khan MA, Desmond RA, Bergman S: Mitochondrial DNA mutations in black Americans with hypertension-associated end-stage renal disease. Am J Kidney Dis 38:529-536, 2001 26. Hong Y, Rice T, Gagnon J, et al: Familial clustering of insulin and abdominal visceral fat: The HERITAGE Family Study. J Clin Endocrinol Metab 83:4239-4245, 1998 27. Davey G, Ramachandran A, Snehalatha C, Hitman GA, McKeigue PM: Familial aggregation of central obesity in Southern Indians. Int J Obes Relat Metab Disord 24:15231527, 2000 28. Snyder EE, Walts B, Perusse L, et al: The human obesity gene map: The 2003 update. Obes Res 12:369-439, 2004 29. Davison KK, Birch LL: Child and parent characteristics as predictors of change in girls’ body mass index. Int J Obes Relat Metab Disord 25:1834-1842, 2001 30. Pinto-Sietsma SJ, Navis G, Janssen WM, de Zeeuw D, Gans RO, de Jong PE, for the PREVEND Study Group: A central body fat distribution is related to renal function impairment, even in lean subjects. Am J Kidney Dis 41:733741, 2003
58
31. Bosma RJ, van der Heide JJ, Oosterop EJ, de Jong PE, Navis G: Body mass index is associated with altered renal hemodynamics in non-obese healthy subjects. Kidney Int 65:259-265, 2004 32. Schelling JR, Sedor JR: The metabolic syndrome as a risk factor for chronic kidney disease: More than a fat chance? J Am Soc Nephrol 15:2773-2774, 2004 33. Hall JE, Henegar JR, Dwyer TM, et al: Is obesity a major cause of chronic kidney disease? Adv Ren Replace Ther 11:41-54, 2004 34. Abrass CK: Overview: Obesity: What does it have to do with kidney disease? J Am Soc Nephrol 15:2768-2772, 2004 35. Bagby SP: Obesity-initiated metabolic syndrome and the kidney: A recipe for chronic kidney disease? J Am Soc Nephrol 15:2775-2791, 2004 36. Wisse BE: The inflammatory syndrome: The role of adipose tissue cytokines in metabolic disorders linked to obesity. J Am Soc Nephrol 15:2792-2800, 2004 37. Pereira AC, Pereira AB, Mota GF, et al: NPHS2 R229Q functional variant is associated with microalbuminuria in the general population. Kidney Int 65:1026-1030, 2004 38. McClellan WM, Ramirez SP, Jurkovitz C: Screening for chronic kidney disease: Unresolved issues. J Am Soc Nephrol 14:S81-S87, 2003 (suppl 2)
SPECKMAN ET AL
39. Satko SG, Freedman BI: The importance of family history on the development of renal disease. Curr Opin Nephrol Hypertens 13:337-341, 2004 40. Boulware LE, Jaar BG, Tarver-Carr ME, Brancati FL, Powe NR: Screening for proteinuria in US adults: A costeffectiveness analysis. JAMA 290:3101-3114, 2003 41. Chagnac A, Weinstein T, Herman M, Hirsh J, Gafter U, Ori Y: The effects of weight loss on renal function in patients with severe obesity. J Am Soc Nephrol 14:14801486, 2003 42. Maddox DA, Alavi FK, Santella RN, Zawada ET Jr: Prevention of obesity-linked renal disease: Age-dependent effects of dietary food restriction. Kidney Int 62:208-219, 2002 43. Velasquez MT, Bhathena SJ, Ranich T, et al: Dietary flaxseed meal reduces proteinuria and ameliorates nephropathy in an animal model of type II diabetes mellitus. Kidney Int 64:2100-2107, 2003 44. National Kidney Disease Education Program: Family Reunion Initiative. Available at: www.nkdep.nih.gov/ familyreunion. Accessed November 1, 2005 45. Fan ZJ, Lackland DT, Kenderes B, Krisher J, Freedman BI: Impact of birth weight on familial aggregation of end-stage renal disease. Am J Nephrol 23:117-120, 2003
F
AMILY HISTORY OF chronic kidney disease (CKD) is a risk factor for end-stage renal disease (ESRD) in African Americans and whites in the United States.1-3 Familial clustering of CKD also has been observed worldwide.4-7 Causes of familial aggregation of CKD are poorly understood; however, there is considerable evidence that genetic factors may predispose family members to ESRD. It also is likely that shared environmental exposures and gene-
environment interactions contribute to the observed clustering of ESRD within families. To unravel these complex relationships, it is important to identify shared personal characteristics and environmental exposures among family members at high risk for ESRD. One such shared characteristic among family members may be obesity. A possible association between high body mass index (BMI) and increased risk for nephropathy
From the Department of Epidemiology, Rollins School of Public Health, Emory University, Atlanta, GA; Department of Internal Medicine, Section on Nephrology, Wake Forest University School of Medicine, Winston-Salem, NC; Centers for Disease Control and Prevention, Atlanta, GA; and Section of Hypertension/Nephrology, Dartmouth Medical School, Hanover, NH. Received November 7, 2005; accepted in revised form March 23, 2006. Work done at Department of Epidemiology, Rollins School of Public Health, Emory University, Atlanta, GA. Support: None. Potential conflicts of interest: None. Presented in abstract form (F-PO321) at the American Society of Nephrology Renal Week, St Louis, MO, October 27-November 1, 2004. The analyses upon which this publication is based were performed under Contract No. 500-97-E024, entitled “EndStage Renal Disease Networks Organization #6,” sponsored by the Centers for Medicare & Medicaid Services. The content of this publication does not necessarily reflect the
views or policies of the Centers for Medicare & Medicaid Services, nor does mention of trade names, commercial products, or organizations imply endorsement by the US government. The authors assume full responsibility for the accuracy and completeness of the ideas presented. This article is a direct result of the Health Care Quality Improvement Program initiated by the Centers for Medicare & Medicaid Services, which has encouraged identification of quality improvement projects derived from analysis of patterns of care, and therefore required no special funding on the part of this contractor. Ideas and contributions to the authors concerning experience in engaging with issues presented are welcomed. Address reprint requests to Rebecca A. Speckman, BA, Department of Epidemiology, Rollins School of Public Health, Emory University, 1518 Clifton Road NE, Atlanta, GA 30322. E-mail: [email protected] © 2006 by the National Kidney Foundation, Inc. 0272-6386/06/4801-0006$32.00/0 doi:10.1053/j.ajkd.2006.03.086
50
American Journal of Kidney Diseases, Vol 48, No 1 (July), 2006: pp 50-58
OBESITY AND FAMILY HISTORY OF ESRD
51
is suggested by several clinical observations. Extreme obesity has been associated repeatedly with enhanced risk for progressive kidney disease.8-11 An association between obesity, BMI, and increased risk for kidney disease was reported in several racial and ethnic groups,12-14 and familial clustering of obesity has been reported widely.15-20 Finally, a recent genome scan in hypertensive African Americans with ESRD found strong evidence for linkage between ESRD and polymorphic markers on chromosome 13q in families with the highest BMI.21 The objective of this study is to determine whether an association exists between reported family history of ESRD and BMI among incident dialysis patients. METHODS
Participants The Family History of ESRD Study population included patients who initiated Medicare-supported renal replacement therapy in Network 6 dialysis facilities (serving North Carolina, South Carolina, and Georgia). Participation by treatment centers and patients was voluntary. The study and data-collection instruments were approved for use by the Wake Forest University School of Medicine Institutional Review Board (Winston-Salem, NC). Network 6 facilities used a standardized data-collection instrument that included the presence of ESRD in first- and second-degree relatives, total number of siblings and children, and their vital status (alive, deceased, unknown). Age, sex, race, weight, and height of patients; primary cause of ESRD; comorbidities; and laboratory results at the initiation of dialysis therapy were obtained from Centers for Medicare & Medicaid Services (CMS) Form 2728.2 The study population included 59,167 incident dialysis patients from ESRD Network 6 (North Carolina, South Carolina, Georgia) facilities who initiated dialysis therapy between 1995 and 2003. Patients residing in other states were excluded (n ⫽ 703), as were patients with ESRD caused by known Mendelian disorders (including polycystic kidney disease and Alport syndrome), urological conditions/ surgical nephrectomy, or missing data on primary cause of ESRD (n ⫽ 1,192). Of the remaining patients, 25,883 (46.28%) consented and provided family history information. Patients younger than 20 years or with a recorded race other than black or white were excluded (n ⫽ 1,055), and patients with unrealistic recorded height and weight measurements (height ⬍ 120 cm or ⬎ 206 cm, weight ⬍ 30 kg or ⬎ 300 kg) or with a missing serum creatinine concentration also were excluded, leaving a total of 23,822 patients eligible for BMI analysis (Fig 1).
Data Collection After obtaining consent, the treatment centers’ social workers or nursing staff used the standardized instrument to screen for history of ESRD in first- and second-degree
Fig 1.
Patients included in analysis.
relatives of incident dialysis patients. A family history was considered present if an incident patient reported having either a first-degree (parent, sibling, or child) or seconddegree (grandparent, aunt, uncle, grandchild, or half-sibling) relative with ESRD. ESRD was described to participants as “end-stage renal disease or kidney failure (ie, dialysis, kidney transplant, or death from kidney disease before dialysis was available or before starting dialysis).” De-identified data from the family history questionnaire were collated with information obtained from CMS Form 2728 for age, sex, race, primary cause of ESRD, comorbidities present at the initiation of dialysis therapy, and clinical characteristics at the initiation of dialysis therapy. BMI was calculated from the height and dry weight collected on CMS Form 2728. Primary cause of ESRD was classified as diabetes mellitus (DM), hypertension (HTN), primary glomerulonephritis, and “other.” The primary outcome was a reported family history of ESRD. BMI was classified by using the World Health Organization classifications: underweight (BMI ⬍ 18.5 kg/m2), normal (BMI, 18.5 to ⬍25 kg/m2), overweight (BMI, 25 to ⬍ 30 kg/m2), obese (BMI, 30 to ⬍ 35 kg/m2), and morbidly obese (BMI ⱖ 35 kg/m2).22 During the years of this study, CMS Form 2728 categories for history of DM were: “Diabetes, currently on insulin” and “Diabetes, primary or contributing.” Because these categories were not mutually exclusive, we created 4 categories for diabetes: “No DM,” “DM, currently on insulin (only),”
52
SPECKMAN ET AL
“DM, primary or contributing (only),” and “DM, both ‘currently on insulin’ and ‘primary or contributing.’” Glomerular filtration rate (GFR; milliliters per minute per 1.73 m2) was calculated from the modified Modification of Diet in Renal Disease formula.23
Statistical Analyses Associations in categorical variables were examined by using chi-square tests and Mantel-Haenszel odds ratios (ORs). Continuous variables were examined by using t-tests, analysis of variance (using Tukey test for pairwise comparisons), and Pearson correlation coefficient. Logistic regression was performed to assess the association between BMI classification and reported family history of ESRD. We adjusted for associations of other patient characteristics with a family history of ESRD by including them as covariates in the logistic regression model. Patient characteristics included in the multivariable model were age, sex, race, history of DM, history of HTN, cause of ESRD, history of smoking, number of first-degree relatives, and estimated GFR. Interactions between race and sex and between BMI class and the other covariates in the model were examined in the logistic regression model by evaluating the significance of cross-product terms. All analyses were performed using SAS, version 9.1 (SAS Institute, Cary, NC).
RESULTS
A family history of ESRD was reported by 22.9% of patients included in this analysis (Table 1). Compared with those who reported no family history of ESRD, patients reporting a family history of ESRD had a greater average BMI (28.2 versus 26.6 kg/m2; P ⬍ 0.001), were younger (mean age, 57.7 versus 61.2 years; P ⬍ 0.001), had more first-degree relatives (8.7 versus 7.6; P ⬍ 0.001), were more likely to be black (74.5% versus 52.2%; P ⬍ 0.001), and were more likely to be female (56.1% versus 48.7%; P ⬍ 0.001). A history of HTN was associated with a family history of ESRD in crude analysis (Table 1), as was a history of DM. A family history of ESRD was not associated with a pre-ESRD history of smoking. Patients with primary causes of ESRD other than glomerulonephritis or HTN were less likely than patients with DM as the primary cause of ESRD to report a family history of ESRD (OR, 0.65; 95% confidence interval [CI], 0.59 to 0.71). Patients with a positive family history had a statistically significant, but clinically minor, difference in average predialysis GFR (8.0 versus 8.5 mL/min/1.73 m2 [0.13 versus 0.14 mL/s/1.73 m2]; P ⬍ 0.001). There was a high prevalence of obesity among incident patients; 6.7% of patients were underweight, 37.9%
had a normal BMI, 27.6% were overweight, 15.2% were obese, and 12.5% were morbidly obese. There was a strong association between BMI and family history of ESRD (Table 1; Fig 2). Patients with a reported family history of ESRD were more likely to be overweight (28.0% versus 27.5%), obese (17.3% versus 14.6%), or morbidly obese (16.7% versus 11.3%; Table 1). The relationship between BMI category and reported family history of ESRD was similar when stratified by race, and an interaction term between race and BMI class was nonsignificant in the multivariable model, indicating that the association between BMI class and family history of ESRD did not differ between whites and blacks. BMI was associated with all covariates. Mean BMI was 27.5 ⫾ 7.6 (SD) kg/m2 in blacks and 26.3 ⫾ 6.8 kg/m2 in whites (P ⬍ 0.001). In women, mean BMI was 27.6 ⫾ 7.9 kg/m2 compared with 26.3 ⫾ 6.6 kg/m2 in men (P ⬍ 0.001). Mean BMI was 28.2 ⫾ 8.2 kg/m2 for those younger than 55 years, 28.2 ⫾ 7.4 kg/m2 for those aged 55 to younger than 65 years, 26.4 ⫾ 6.4 kg/m2 for those aged 65 to younger than 75 years, and 24.6 ⫾ 5.6 kg/m2 for those 75 years and older (P ⬍ 0.001; all pairwise comparisons significant except between the 2 age groups). Mean BMI by primary cause was 28.5 ⫾ 7.3 kg/m2 for DM, 27.5 ⫾ 8.1 kg/m2 for glomerulonephritis, 25.9 ⫾ 7.1 kg/m2 for HTN, and 24.8 ⫾ 6.4 kg/m2 for other causes (P ⬍ 0.001; all pairwise comparisons significant). Mean BMI by history of DM category was 27.5 ⫾ 7.5 kg/m2 for patients with DM administered oral medications, 27.3 ⫾ 7.2 kg/m2 for patients with DM administered no medications, 26.3 ⫾ 6.6 kg/m2 for patients with DM administered insulin, and 25.5 ⫾ 6.9 kg/m2 for patients with no DM (P ⬍ 0.001; all pairwise comparisons significant except between patients with DM administered oral medications and patients with DM administered no medications). BMI correlated weakly with GFR (Pearson r ⫽ ⫺0.04383; P ⬍ 0.001). The association between family history and BMI classification persisted after adjustment for age, race, sex, pre-ESRD history of HTN, preESRD history of DM, primary cause of ESRD, history of smoking, number of first-degree relatives, and GFR (Table 2). Overweight patients had a 17% greater odds of reporting a family history of ESRD compared with underweight
OBESITY AND FAMILY HISTORY OF ESRD
53
Table 1. Characteristics of Patients With and Without a Reported Family History of ESRD Reported Family History of ESRD
Characteristic
BMI (kg/m2) Age (y) No. of first-degree relatives Age categories (y) ⬍55 55 to ⬍65 65 to ⬍75 ⱖ75 Sex Male Female Race Black White BMI category (kg/m2) Underweight (BMI ⬍ 18.5) Normal (BMI, 18.5-24.9) Overweight (BMI, 25-29.9) Obese (BMI, 30-34.9) Morbidly obese (BMI ⱖ 35) History of HTN No Yes History of DM No DM DM, currently on insulin DM, primary or contributing DM, both “currently on insulin” and “primary or contributing” History of smoking No Yes ESRD primary cause DM HTN Glomerulonephritis Other GFR (mL/min/1.73 m2)
Yes (n ⫽ 5,453)
No (n ⫽ 18,369)
OR (95% CI)*
P†
28.2 ⫾ 8.0 57.7 ⫾ 14.7 8.7 ⫾ 5.3
26.6 ⫾ 7.1 61.2 ⫾ 15.0 7.6 ⫾ 5.1
— — —
⬍0.001 ⬍0.001 ⬍0.001
2,224 (40.8) 1,343 (24.6) 1,171 (21.5) 715 (13.1)
5,800 (31.6) 4,038 (22.0) 4,922 (26.8) 3,609 (19.8)
Reference 0.87 (0.80-0.94) 0.62 (0.57-0.67) 0.52 (0.47-0.57)
⬍0.001
2,394 (43.9) 3,059 (56.1)
9,431 (51.3) 8,938 (48.7)
Reference 1.35 (1.27-1.43)
⬍0.001
4,065 (74.5) 1,388 (25.5)
9,580 (52.2) 8,789 (47.8)
2.69 (2.51-2.87) Reference
302 (5.5) 1,771 (32.5) 1,526 (28.0) 945 (17.3) 909 (16.7)
1,297 (7.1) 7,266 (39.6) 5,058 (27.5) 2,679 (14.6) 2,069 (11.3)
0.96 (0.83-1.09) Reference 1.24 (1.15-1.33) 1.45 (1.32-1.58) 1.80 (1.64-1.98)
752 (13.8) 4,701 (86.2)
3,083 (16.8) 15,286 (83.2)
Reference 1.26 (1.16-1.37)
⬍0.001
2,403 (44.1) 386 (7.1) 1,438 (26.4) 1,226 (22.5)
8,771 (47.8) 1,145 (6.2) 4,713 (25.7) 3,740 (20.4)
Reference 1.23 (1.09-1.39) 1.11 (1.04-1.20) 1.20 (1.11-1.29)
⬍0.001
5,004 (91.8) 449 (8.2)
16,740 (91.1) 1,629 (8.9)
Reference 0.92 (0.83-1.03)
2,757 (50.6) 1,519 (27.9) 519 (9.5) 658 (12.1) 8.0 ⫾ 4.7
8,522 (46.4) 4,945 (26.9) 1,755 (9.6) 3,147 (17.13) 8.5 ⫾ 4.6
Reference 0.95 (0.88-1.02) 0.91 (0.82-1.02) 0.65 (0.59-0.71) —
⬍0.001 ⬍0.001
0.145 ⬍0.001
⬍0.001
NOTE. N ⫽ 23,822. Values expressed as mean ⫾ SD or number (percent). To convert GFR in mL/min to mL/s, multiply by 0.01667. *Mantel-Haenszel ORs with 95% CIs. †Chi-square for categorical variables (race, sex, ESRD cause); independent sample t-tests for continuous variables (BMI, age, serum albumin level, GFR).
patients (OR, 1.17; 95% CI, 1.08 to 1.26). Obese patients had 25% greater odds (OR, 1.25; 95% CI, 1.12 to 1.37), and morbidly obese patients had 40% greater odds (OR, 1.40; 95% CI, 1.27 to 1.55). Younger age, black race, female sex, and number of first-degree relatives also were associated with a reported family history of ESRD in the adjusted analysis. We did not find significant interactions between BMI class and the other
covariates in the model, indicating that there was no synergistic relationship between BMI class and other covariates. DISCUSSION
Our study is the first report of an association between family history of ESRD and obesity among incident dialysis patients. The association was strong, increased in strength with more se-
54
SPECKMAN ET AL
Fig 2. Reported family history of ESRD and BMI (kg/m2).
vere degrees of obesity, and persisted after adjustment for other patient characteristics, including history of DM, history of HTN, and primary cause of ESRD. In addition, we noted that younger age, increasing number of first-degree relatives, female sex, black race, and history of HTN were associated with a family history of ESRD in the adjusted analysis. These observations warrant comment. First, it is not unexpected that patients with more firstdegree family members would be more likely to report a family history of ESRD. In addition, our finding of an association between younger age and family history of ESRD agrees with earlier findings24 and may be explained by a tendency for earlier onset in the presence of a genetic predisposition to ESRD. The association between family history of ESRD in blacks is consistent with findings from previous studies and with the well-known disparities between blacks and whites for incident ESRD. This association suggests that either as yet unidentified environmental factors that aggregate within black families and/or genetic factors may account for these differences in familial aggregation of ESRD, possibilities that are a focus of active investigation. The association between being female and having a family history of ESRD is interesting, and it is tempting to attribute it to sex-specific recall bias. Another possibility is that at least some of the familial aggregation of ESRD may
be transmitted matrilineally by either genomic imprinting or the mitochrondrial genome.25 However, information with which to address these possibilities are absent from our database, and evaluation of other family history databases will be necessary to determine the true significance of this association. Although it might be expected that the observed association between BMI and family history of ESRD is explained by the relationship of type 2 DM and HTN to obesity, it is important to note that the association remained after controlling for both history of DM and HTN. Furthermore, the association also was present in patients with glomerulonephritis as primary cause of ESRD. We found no interaction between BMI and other variables, indicating that BMI does not have a synergistic relationship with other patient characteristics, such as history of HTN or history of DM. These findings suggest that BMI may influence the familial aggregation of ESRD independently of other patient characteristics, including those associated with both ESRD and obesity. The association between increased BMI and positive family history of ESRD might be explained by 2 hypotheses that are not mutually exclusive. First, it is possible that the familial clustering of obesity, which is a risk factor for ESRD, contributes to increased risk for progressive kidney disease among family members. This possibility is supported by the familial clustering
OBESITY AND FAMILY HISTORY OF ESRD
55
Table 2. Adjusted Association Between BMI of Incident Patients With ESRD and Family History of ESRD Factor
BMI category (kg/m2) Underweight (BMI ⬍ 18.5) Normal (BMI, 18.5-24.9) Overweight (BMI, 25-29.9) Obese (BMI, 30-34.9) Morbidly obese (BMI ⱖ 35) Age group (y) ⬍55 55 to ⬍64 65 to ⬍74 ⱖ75 No. of first-degree relatives Sex Female Male Race Black White History of HTN Yes No History of DM No DM DM, currently on insulin DM, primary or contributing DM, both “currently on insulin” and “primary or contributing” History of smoking No Yes ESRD primary cause DM HTN Glomerulonephritis Other GFR (mL/min/1.73 m2)
OR (95% CI)
Wald P
0.93 (0.81-1.07) Reference 1.17 (1.08-1.26) 1.25 (1.14-1.37) 1.40 (1.27-1.55)
0.338 ⬍0.001 ⬍0.001 ⬍0.001
Reference 0.89 (0.82-0.97) 0.71 (0.65-0.77) 0.66 (0.60-0.73) 1.04 (1.03-1.05)*
0.007 ⬍0.001 ⬍0.001 ⬍0.001
1.25 (1.17-1.33) Reference
⬍0.001
2.38 (2.21-2.55) Reference
⬍0.001
1.12 (1.02-1.23) Reference
⬍0.001
Reference 1.10 (0.94-1.28) 1.01 (0.91-1.12) 1.09 (0.96-1.23)
0.230 0.876 0.184
Reference 1.01 (0.90-1.14)
0.851
Reference 0.94 (0.87-1.01) 0.97 (0.86-1.08) 0.79 (0.71-0.87) 1.00 (0.99-1.00)
0.100 0.564 ⬍0.001 0.217
NOTE. All covariates found to be significant were included in the model and are listed in this table. To convert GFR in mL/min to mL/s, multiply by 0.01667. *OR for “number of first-degree relatives” is the increase in odds for an increase in 1 first-degree relative.
of obesity and the observation that obesity is itself a risk factor for CKD and ESRD. Previous studies observed familial clustering of obesity15-20 and central obesity (abdominal visceral fat).26,27 Whereas several genetic loci have been linked to obesity,28 such shared environmental factors as energy intake, dietary composition, and energy expenditure also are likely to be partially responsible for the familial clustering of obesity.20,29 Obesity or extreme obesity also is associated with increased risk for progressive kidney disease or ESRD,8-13 and obesity is associated with increased risk for albuminuria.30 The association
between increased BMI and risk for CKD seems to hold even for the nonobese.14 Increasing BMI correlated with increased filtration fraction and hyperfiltration, which could predispose to progressive renal injury, in nonobese volunteers who did not have DM or HTN.31 The deleterious effects of obesity on the kidney are mediated in part by DM, hyperglycemia, HTN, and/or insulin resistance.32 However, obesity also may contribute to kidney injury through other mechanisms.33,34 Accumulation of visceral fat can increase intrarenal pressure and may lead to extracellular matrix production in the renal interstitium. Increased sympathetic nervous sys-
56
tem activity, possibly mediated by factors that include leptin, insulin, long-chain fatty acids, and angiotensin II, also contributes to renal dysfunction.35 Inflammatory mediators produced by such adipocytes as tumor necrosis factor ␣, interleukin 6, and C-reactive protein may contribute to glomerular and interstitial fibrosis.36 A second hypothesis that might partially explain the observed BMI–family history of ESRD association is that obesity increases the risk for ESRD in individuals with a genetic predisposition to progressive kidney injury through a geneby-environment interaction. This possibility is supported by 2 recent studies. A recent genome scan performed in 264 African-American pedigrees enriched for the presence of non-DM ESRD observed modest evidence for linkage between ESRD and chromosome 13q33.3 in initial analyses, with a maximum log of the odds score for linkage of 1.02.21 After stratification by BMI, log of the odds score increased to 5.20 in the 46% of families with the greatest BMI (P ⫽ 0.0002). Similarly, another study recently reported that an association between a functional variant of the podocin gene (NPHS2 R229Q) and microalbuminuria in the general population was the strongest in patients with increased BMI.37 It is important to note the clinical and public health relevance of these observations. Familytargeted prevention is an important strategy in the prevention of ESRD,38-40 and animal and human studies suggested that weight reduction or alteration in nutritional content early or late in the course of obesity-mediated renal dysfunction can improve renal function.41-43 Our findings suggest that strategies to control or reduce weight might be especially important in prevention efforts targeted to family members of patients with ESRD. Family-based ESRD prevention efforts, for example, the “Family Reunion Initiative!” recently launched by the National Kidney Disease Education Program,44 might integrate knowledge on dietary modification in patients with obesity-related renal disease with existing knowledge in the field of family-based obesity interventions. There are strengths and limitations to our study. It is the largest family history of ESRD data set ever compiled, and BMI information was obtained independently for all patients at the initiation of renal replacement therapy. Although there was rela-
SPECKMAN ET AL
tively low overall participation with respect to voluntarily reporting a family history of ESRD, characteristics of nonparticipants were similar to those of participants, and there was no significant difference between participants and nonparticipants in the distribution of BMI classification (P ⫽ 0.219, data not shown). A potential limitation of our study is error in reported dry weights provided on CMS Form 2728. It is possible, but unlikely, that the familial aggregation of increased BMI reflected greater degrees of edema in individuals reporting a positive family history. We attempted to account for this in part by controlling for the reported cause of ESRD and baseline residual GFR. However, it remains possible that better estimates of dry body weight might diminish the associations that we noted. Future studies of individuals with ESRD and those with less severe degrees of kidney disease who have data for dry weight and fat distribution will help resolve this issue. Another limitation of our study is that we did not have information on obesity or BMI in family members of patients with ESRD. We suggest that future studies on this topic address several points related to the hypotheses generated by our study. Information on BMI and other clinical characteristics of relatives is needed in addition to the presence or absence of ESRD. More precise information on body fat distribution also might help elucidate the cause of the observed BMI–family history of ESRD association. Similarly, information on socioeconomic status and dietary habits of patients and their relatives might allow for exploration of potential environmental causes of familial clustering. Studies of genetic factors for ESRD risk should consider effect modification by BMI or body fat distribution. Although low birth weight is not associated with familial aggregation of ESRD,45 future studies of maternal smoking might explore the possibility that obesity increases risks that are imparted by environmental exposures occurring in utero. In conclusion, we find that overweight and obesity are associated with a reported family history of ESRD among incident dialysis patients independent of age, race, sex, history of DM or HTN, or primary cause of ESRD. This association raises the possibility that obesitypromoting environmental factors shared in fami-
OBESITY AND FAMILY HISTORY OF ESRD
lies, obesity-related genes, and interaction between obesity and other genes that predispose to ESRD may contribute to the familial aggregation of risk for ESRD. This finding also suggests that within afflicted families, control of body weight may be an important target to reduce the risk for ESRD. ACKNOWLEDGMENT The authors thank Network 6 dialysis facility staff members for collecting data and the dialysis patients who participated in the study.
REFERENCES 1. Ferguson R, Grim CE, Opgenorth TJ: A familial risk of chronic renal failure among blacks on dialysis? J Clin Epidemiol 41:1189-1196, 1998 2. Freedman BI, Soucie JM, McClellan WM: Family history of end-stage renal disease among incident dialysis patients. J Am Soc Nephrol 8:1942-1945, 1997 3. Freedman BI, Volkova NV, Satko SG, et al: Populationbased screening for family history of end-stage renal disease among incident dialysis patients. Am J Nephrol 25:529-535, 2005 4. Gumprecht J, Zychma MJ, Moczulski DK, Gosek K, Grzeszczak W: Family history of end-stage renal disease among hemodialyzed patients in Poland. J Nephrol 16:511515, 2003 5. Nelson RG, Knowler WC, McCance DR, et al: Determinants of end-stage renal disease in Pima Indians with type 2 (non-insulin-dependent) diabetes mellitus and proteinuria. Diabetologia 36:1087-1093, 1993 6. Canani LH, Gerchman F, Gross JL: Familial clustering of diabetic nephropathy in Brazilian type 2 diabetic patients. Diabetes 48:909-913, 1999 7. Chandie Shaw PK, van Es LA, Paul LC, Rosendaal FR, Souverijn JH, Vandenbroucke JP: Renal disease in relatives of Indo-Asian type 2 diabetic patients with endstage diabetic nephropathy. Diabetologia 46:618-624, 2003 8. Weisinger JR, Kempson RL, Eldridge FL, Swenson RS: The nephrotic syndrome: A complication of massive obesity. Ann Intern Med 81:440-447, 1974 9. Kasiske BL, Crosson JT: Renal disease in patients with massive obesity. Arch Intern Med 146:1105-1109, 1986 10. Jennette JC, Charles L, Grubb W: Glomerulomegaly and focal segmental glomerulosclerosis associated with obesity and sleep-apnea syndrome. Am J Kidney Dis 10:470472, 1987 11. Kasiske BL, Napier J: Glomerular sclerosis in patients with massive obesity. Am J Nephrol 5:45-50, 1985 12. Stengel B, Tarver-Carr ME, Powe NR, Eberhardt MS, Brancati FL: Lifestyle factors, obesity and the risk of chronic kidney disease. Epidemiology 14:479-487, 2003 13. Fox CS, Larson MG, Leip EP, Culleton B, Wilson PW, Levy D: Predictors of new-onset kidney disease in a community-based population. JAMA 291:844-850, 2004 14. Vupputuri S, Sandler DP: Lifestyle risk factors and chronic kidney disease. Ann Epidemiol 13:712-720, 2003
57
15. Burns TL, Moll PP, Lauer RM: The relation between ponderosity and coronary risk factors in children and their relatives. The Muscatine Ponderosity Family Study. Am J Epidemiol 129:973-987, 1989 16. Ayatollahi SM: Obesity in school children and their parents in southern Iran. Int J Obes Relat Metab Disord 16:845-850, 1992 17. Guillaume M, Lapidus L, Beckers F, Lambert A, Bjorntorp P: Familial trends of obesity through three generations: The Belgian-Luxembourg child study. Int J Obes Relat Metab Disord 19:S5-S9, 1995 (suppl 3) 18. Magnusson PK, Rasmussen F: Familial resemblance of body mass index and familial risk of high and low body mass index. A study of young men in Sweden. Int J Obes Relat Metab Disord 26:1225-1231, 2002 19. Fuentes RM, Notkola IL, Shemeikka S, Tuomilehto J, Nissinen A: Familial aggregation of body mass index: A population-based family study in eastern Finland. Horm Metab Res 34:406-410, 2002 20. Mirmiran P, Mirbolooki M, Azizi F: Familial clustering of obesity and the role of nutrition: Tehran Lipid and Glucose Study. Int J Obes Relat Metab Disord 26:16171622, 2002 21. Freedman BI, Langefeld CD, Rich SS, et al: A genome scan for end-stage renal disease in African American families enriched for non-diabetic nephropathy. J Am Soc Nephrol 15:2719-2727, 2004 22. World Health Organization: Obesity: Preventing and Managing the Global Epidemic. Report from a WHO Consultation. Geneva, Switzerland, World Health Organization, WHO Technical Report Series 894, 2000 23. Levey AS, Greene T, Kusek JW, Beck GL, for the MDRD Study Group: A simplified equation to predict glomerular filtration rate from serum creatinine. J Am Soc Nephrol 11:155A, 2000 (abstr) 24. Spray BJ, Atassi NG, Tuttle AB, Freedman BI: Familial risk, age at onset, and cause of end-stage renal disease in white Americans. J Am Soc Nephrol 5:1806-1810, 1995 25. Watson B Jr, Khan MA, Desmond RA, Bergman S: Mitochondrial DNA mutations in black Americans with hypertension-associated end-stage renal disease. Am J Kidney Dis 38:529-536, 2001 26. Hong Y, Rice T, Gagnon J, et al: Familial clustering of insulin and abdominal visceral fat: The HERITAGE Family Study. J Clin Endocrinol Metab 83:4239-4245, 1998 27. Davey G, Ramachandran A, Snehalatha C, Hitman GA, McKeigue PM: Familial aggregation of central obesity in Southern Indians. Int J Obes Relat Metab Disord 24:15231527, 2000 28. Snyder EE, Walts B, Perusse L, et al: The human obesity gene map: The 2003 update. Obes Res 12:369-439, 2004 29. Davison KK, Birch LL: Child and parent characteristics as predictors of change in girls’ body mass index. Int J Obes Relat Metab Disord 25:1834-1842, 2001 30. Pinto-Sietsma SJ, Navis G, Janssen WM, de Zeeuw D, Gans RO, de Jong PE, for the PREVEND Study Group: A central body fat distribution is related to renal function impairment, even in lean subjects. Am J Kidney Dis 41:733741, 2003
58
31. Bosma RJ, van der Heide JJ, Oosterop EJ, de Jong PE, Navis G: Body mass index is associated with altered renal hemodynamics in non-obese healthy subjects. Kidney Int 65:259-265, 2004 32. Schelling JR, Sedor JR: The metabolic syndrome as a risk factor for chronic kidney disease: More than a fat chance? J Am Soc Nephrol 15:2773-2774, 2004 33. Hall JE, Henegar JR, Dwyer TM, et al: Is obesity a major cause of chronic kidney disease? Adv Ren Replace Ther 11:41-54, 2004 34. Abrass CK: Overview: Obesity: What does it have to do with kidney disease? J Am Soc Nephrol 15:2768-2772, 2004 35. Bagby SP: Obesity-initiated metabolic syndrome and the kidney: A recipe for chronic kidney disease? J Am Soc Nephrol 15:2775-2791, 2004 36. Wisse BE: The inflammatory syndrome: The role of adipose tissue cytokines in metabolic disorders linked to obesity. J Am Soc Nephrol 15:2792-2800, 2004 37. Pereira AC, Pereira AB, Mota GF, et al: NPHS2 R229Q functional variant is associated with microalbuminuria in the general population. Kidney Int 65:1026-1030, 2004 38. McClellan WM, Ramirez SP, Jurkovitz C: Screening for chronic kidney disease: Unresolved issues. J Am Soc Nephrol 14:S81-S87, 2003 (suppl 2)
SPECKMAN ET AL
39. Satko SG, Freedman BI: The importance of family history on the development of renal disease. Curr Opin Nephrol Hypertens 13:337-341, 2004 40. Boulware LE, Jaar BG, Tarver-Carr ME, Brancati FL, Powe NR: Screening for proteinuria in US adults: A costeffectiveness analysis. JAMA 290:3101-3114, 2003 41. Chagnac A, Weinstein T, Herman M, Hirsh J, Gafter U, Ori Y: The effects of weight loss on renal function in patients with severe obesity. J Am Soc Nephrol 14:14801486, 2003 42. Maddox DA, Alavi FK, Santella RN, Zawada ET Jr: Prevention of obesity-linked renal disease: Age-dependent effects of dietary food restriction. Kidney Int 62:208-219, 2002 43. Velasquez MT, Bhathena SJ, Ranich T, et al: Dietary flaxseed meal reduces proteinuria and ameliorates nephropathy in an animal model of type II diabetes mellitus. Kidney Int 64:2100-2107, 2003 44. National Kidney Disease Education Program: Family Reunion Initiative. Available at: www.nkdep.nih.gov/ familyreunion. Accessed November 1, 2005 45. Fan ZJ, Lackland DT, Kenderes B, Krisher J, Freedman BI: Impact of birth weight on familial aggregation of end-stage renal disease. Am J Nephrol 23:117-120, 2003