International Journal of Obesity (2005) 29, 11151120 & 2005 Nature Publishing Group All rights reserved 0307-0565/05

$30.00 www.nature.com/ijo

PAPER
Effect of weight loss using formula diet on renal function in obese patients with diabetic nephropathy
A Saiki1, D Nagayama2, M Ohhira1, K Endoh1, M Ohtsuka1, N Koide1, T Oyama1, Y Miyashita1* and K Shirai2
1 2

Center of Diabetes, Endocrine and Metabolism, Sakura Hospital, School of Medicine, Toho University, Chiba, Japan; and Internal Medicine, Sakura Hospital, School of Medicine, Toho University, Chiba, Japan

OBJECTIVE: To evaluate the effect and safety of treatment with low-calorie formula diet on renal function and proteinuria in obese patients with diabetic nephropathy. DESIGN: Prospective study on safety and efficacy of a 4-week low-calorie (1119 kcal/kg/day) normal-protein (0.91.2 g/kg/ day) diet partly supplemented with formula diet. SUBJECTS: In all, 22 obese patients with diabetic nephropathy (BMI: 30.475.3 kg/m2, HbA1c: 7.171.4%, serum creatinine: 172.4757.5 mmol/l, urinary protein: 3.372.6 g/day). RESULTS: The mean body weight decreased by 6.273.0 kg. The mean systolic blood pressure, creatinine, blood urea nitrogen, urinary protein, and 8-hydroxydeoxyguanosine decreased significantly by 7.5712.7 mmHg, 41.6723.9 mmol/l, 1.5071.61 mmol/l, 1.871.7 g/day, and 3.173.6 ng/mg creatinine, respectively. No patient had increased serum creatinine and urinary protein. Mean creatinine clearance (40.6717.9 to 46.1714.6 ml/s/1.73 m2) and serum albumin showed no significant changes. Dserum creatinine and Durinary protein correlated with Dbody weight (r 0.62 and 0.49, respectively) and Dvisceral fat area (r 0.58 and 0.58, respectively), but did not correlate with Dsystolic blood pressure, Dfasting blood glucose and Dsubcutaneous fat area. CONCLUSION: These results suggested that weight reduction using formula diet might improve renal function and proteinuria safely for a short term in obese patients with diabetic nephropathy. International Journal of Obesity (2005) 29, 11151120. doi:10.1038/sj.ijo.0803009; published online 31 May 2005 Keywords: diabetic nephropathy; formula diet; oxidative stress; visceral fat

Introduction
Diabetic nephropathy is considered to be one of the most common causes of renal failure in Japan. Once overt proteinuria has set in, there are no therapeutic interventions that can halt the progressive loss of renal function. The decline to end-stage renal disease is often rapid with a mean rate of decline of glomerular filtration rate (GFR) of approximately 1 ml/min/month.1 Although there is no therapy for progressive loss of renal function, it can be delayed by antihypertensive therapy2 and may be influenced by the quality of metabolic control.3 Recent report has indicated a strong association of oxidative stress with diabetic nephropathy.4 In dietary treatment of diabetic nephropathy,

*Correspondence: Dr Y Miyashita, Center of Diabetes, Endocrine and Metabolism, Sakura Hospital, School of Medicine, Toho University, 564-1, Shimoshizu, Sakura-City, Chiba 285-8741, Japan. E-mail: mumon@sf6.so-net.ne.jp Received 18 September 2004; revised 28 March 2005; accepted 4 April 2005; published online 31 May 2005

several studies reported that high-calorie, low-protein diet was beneficial for renal function and proteinuria.57 On the other hand, obesity is recognized as a risk factor of renal dysfunction. The mechanism by which obesity deteriorates the renal function is not fully understood. Weight loss has been shown to induce a dramatic decrease in proteinuria in patients with obesity-related proteinuria,8 but no study has investigated the effect of weight loss in obese patients with diabetic nephropathy. Formula diet, which is normal-protein diet, is used for the treatment of obesity. Currently, normal-protein diet is not recommended for diabetic nephropathy. However, Nakamura et al9 reported that the GFR responded differently to various proteins, and that some proteins do not have adverse effects on GFR. Their report indicated that formula diet, which mainly consists of soybean and dairy product, might be safe for diabetic nephropathy. We therefore designed a prospective study to evaluate the effect and safety of low-calorie formula diet on renal function and proteinuria in obese patients with diabetic nephropathy. Furthermore, we studied whether

Formula diet for diabetic nephropathy A Saiki et al

1116
weight loss might affect the oxidative stress in diabetic nephropathy. meals of 400 kcal each (total; 970 kcal), or two meals of formula diet and one ordinary meal (total; 740 kcal). Salt intake was 2.79 g/day for the 740 kcal diet and 4.90 g/day for the 970 kcal/day diet.

Methods
Patients In all, 22 obese patients (16 males and six females) with diabetic nephropathy were studied and observed under hospitalization. We selected patients who satisfied the following criteria: obesity (body mass index (BMI) 425 kg/ m2), presence of diabetic retinopathy, proteinuria (urinary albumin 4300 mg/day) and serum creatinine level less than 265.2 mmol/l. Excluded were patients with unstable diabetic retinopathy and presence of pleural effusion and severe leg edema. Baseline characteristics of these study patients are shown in Table 1. Before the study, the patients were prescribed a daily caloric intake of 2530 kcal/kg and 0.8 g/ kg protein at least for 3 months. Drugs such as statins, probucol, angiotensin-converting enzyme inhibitors, angiotensin-receptor blockers, b-blockers and diuretics that had been taken for more than 2 months prior to the study were continued during this study, on condition that the doses remained unchanged. Drugs for diabetes mellitus such as insulin and sulfonyl ureas were reduced or discontinued at the time of hypoglycemia. Clinical and laboratory measurements Body weight measurement and blood collection for laboratory evaluations were performed after an overnight fast. Blood pressures were measured in the afternoon after 5 min of rest in a sitting position. Twenty-four-hour urine sample was collected for urinary protein and creatinine clearance measurements. Creatinine clearance was calculated by measuring intrinsic creatinine in serum and urine. Area of visceral fat was evaluated using CT. The CT scan was performed at the umbilical level with the subject resting in the supine position. The visceral fat area was traced manually along the inside of the abdominal wall, and the number of pixels showing CT values between 50 and 150 Hounsfield units was calculated for this region. The subcutaneous fat area was calculated as follows: total fat area visceral fat area. Urinary 8-hydroxydeoxyguanosine (8OHdG) is considered a biological marker of in vivo oxidative DNA damage.10 We utilized the 8-OHdG as an oxidative stress marker in this study. Body weight, systolic blood pressure, diastolic blood pressure, blood urea nitrogen (BUN), serum creatinine, serum albumin, fasting blood glucose, HbA1c, total cholesterol (TC), triglycerides (TG) and high-density lipoprotein cholesterol (HDL-C), urinary protein and creatinine clearance were measured once a week (week 04). Visceral fat, subcutaneous fat and 8-OHdG were measured before and after the study.

Diets The patients were prescribed a total daily caloric intake of 740 or 970 kcal (1119 kcal/kg) A formula diet consisting of liquid protein (ObeCure; US Cure Inc.,. Tokyo, Japan) providing 170 kcal/pack was used as part of the diet. Every day for 4 weeks, the patients either consumed one meal of formula diet meal containing 170 kcal and two ordinary

Table 1 Baseline characteristics and changes in clinical and biochemical variables 0 Week Age (y) Height (m) Weight (kg) Body mass index (kg/m2) Systolic blood pressure (mmHg) Diastolic blood pressure (mmHg) Serum albumin (g/l) Blood urea nitrogen (mmol/l) Serum creatinine (mmol/l) Creatinine clearance (ml/s/1.73 m2) Urinary protein (g/24 h) Fasting blood glucose (mmol/l) HbA1c (%) Total cholesterol (mmol/l) Triglyceride (mmol/l) HDL-C (mmol/l) Visceral fat area (cm2) Subcutaneous fat area (cm2) 8-OHdG (ng/mg Cr) 53.678.4 167.177.4 85.2717.0 30.475.3 126.5714.2 79.4710.7 39.174.6 11.575.8 172.4757.5 0.6870.29 3.2772.63 6.8971.57 7.1171.42 5.2471.48 1.7670.91 1.0670.19 233.1766.5 244.57101.2 14.073.9 4 Weeks F F 79.0717.2 28.275.5 119.0713.0 72.179.3 39.774.7 9.975.8 130.8746.9 0.7770.24 1.5071.28 6.3371.55 6.6871.21 4.6270.96 1.3870.56 1.0370.29 191.0767.0 215.9774.4 10.974.7 P-value F F o0.0001 o0.0001 o0.05 o0.05 NS o0.005 o0.0001 NS o0.0001 NS o0.05 o0.05 o0.05 NS o0.0001 o0.05 o0.05

Informed consent We explained to the patients that the study would be discontinued under the following circumstances: aggravation of serum creatinine and proteinuria, progression of diabetic retinopathy, and problem relevant to weight loss using formula diet in the study. With full informed consent, the investigation was permitted by all patients.

Statistical analysis The results are expressed as mean7s.d. A t-test was used for group comparisons. P-values of o0.05 were considered significant.

Results
During the 4 weeks, no patient showed progression of diabetic retinopathy in this study. Mean body weight decreased significantly (Po0.0001) from 85.2717.0 to 79.0717.2 kg (Table 1). Mean systolic blood pressure, diastolic blood pressure, HbA1c, TC and TG showed

International Journal of Obesity

Formula diet for diabetic nephropathy A Saiki et al

1117
significant decreasesMean fasting blood glucose HDL-C serum albumin showed no significant changesMean serum creatinine decreased significantly (Po0.0001) from 172.4757.5 to 130.8746.9 mmol/l and mean BUN decreased significantly (Po0.005) from 11.575.8 to 9.975.8 mmol/l. Urinary protein decreased significant (Po0.0001) from 3.2772.63 to 1.5071.28 g/24 h. All patients showed decreases of serum creatinine and urinary protein. Mean creatinine clearance showed no significant increase (from 0.6870.29 to 0.7770.24 ml/s/1.73 m2). Visceral fat decreased significantly (Po0.0001) from 233.1766.5 to 191.0767.0 cm2, and subcutaneous fat also decreased significantly (Po0.05) from 244.57101.2 to 215.9774.4 cm2. 8-OHdG showed a significant decrease (Po0.05) during 4 weeks from 14.073.9 to 10.974.7 ng/mg creatinine. Since both serum creatinine and urinary protein decreased significantly in this study, the correlation between Dserum creatinine or Durinary protein with other factors was studied. Decrease in serum creatinine correlated significantly with weight loss (r 0.62, Po0.005) (Figure 1a) and decrease in visceral fat (r 0.58, Po0.05) (Figure 2a). The Dserum creatinine showed no significant correlation with Dsystolic blood pressure, diastolic blood pressure, fasting blood glucose, HbA1c, TC, TG, HDL-C (data not shown) and subcutaneous fat (Figure 2b). Decrease in urinary protein also correlated significantly with weight loss (r 0.49, Po0.05)

Figure 1 (a) Correlation between change of the body weight and change of serum creatinine. Decrease in serum creatinine and weight loss show a significant correlation (r 0.62, Po0.005). (b) Correlation between change of the body weight and change of urinary protein. Decrease in proteinuria correlates with weight loss (r 0.49, Po0.05).

Figure 2 (a) Correlation between change of visceral fat area and change of serum creatinine. Decrease in serum creatinine and decrease in visceral fat show a significant correlation (r 0.58, Po0.05). (b) Correlation between change of subcutaneous fat area and change of the serum creatinine. Dserum creatinine does not correlate with Dsubcutaneous fat. (c) Correlation between change of visceral fat area and change of the urinary protein. Decrease in proteinuria correlates significantly with decrease in visceral fat (r 0.58, Po0.05). (d) Correlation between change of subcutaneous fat area and change of urinary protein. DUrinary protein showed no significant correlation with Dsubcutaneous fat.

International Journal of Obesity

Formula diet for diabetic nephropathy A Saiki et al

1118
taken by obese subjects with diabetic nephropathy may improve renal function and proteinuria in the short term. In this study, both serum creatinine and the urinary protein correlated significantly with weight loss. These results indicate that weight loss may be the most important factor for improving renal function and proteinuria in diabetic nephropathy with obesity. However, several confounding factors should be examined. First, glycemic control and blood pressure have been reported to impact renal function and proteinuria. The United Kingdom Prospective Diabetes Study (UKPDS)12 reported that tight blood pressure control in type 2 diabetes suppressed proteinuria and progression to end-stage renal failure. Moreover, the Diabetes Control and Complications Trial (DCCT)13 on type 1 diabetes and the Kumamoto study14 on type 2 diabetes have reported that intensive glycemic control suppresses the progression of nephropathy. In the present study in which medications remained unchanged and only the diet was modified, the mean blood pressure decreased significantly while blood glucose was apparently lowered but not significantly. However, changes in serum creatinine and urinary protein did not correlate with changes in blood pressures and blood glucose, whereas they correlated with weight loss. Therefore, we speculate that the blood pressure effect may have partially contributed to improving renal function and proteinuria, but weight loss has the major effect. Another factor is salt intake. In the present study, dietary salt intake was low (2.70 or 4.90 g/day). Cianciaruso et al15 observed improvements in renal function and proteinuria with low sodium intake compared to high sodium intake in patients with progressive renal disease. Allen et al16 reported suppression of hyperfiltration, renal enlargement and albuminuria by salt restriction in diabetic rats. In the present study, we did not measure urinary electrolytes and the effect of salt restriction on renal function and urinary protein is unknown. However, the significant correlation between weight loss and serum creatinine and urinary protein excretion indicates that weight loss, at least, contributes greatly to the improvement. Several studies showed that obese patients could develop proteinuria and progressive loss of renal function,17,18 and weight loss improved them.8,19 The proposed mechanisms by which obesity increased renal risk include increasing GFR,20 focal segmental glomerulosclerosis,21 activation of the renninangiotensin system,8 and adipocyte-derived cytokines.22 However, the impact mechanism underlying the increased renal risk of obesity is not fully understood. In this study, both serum creatinine and urinary protein correlated significantly with visceral fat area. These results suggested that the adipocyte-derived cytokines might affect nephropathy directly or indirectly. Wolf et al22 reported that leptin stimulated renal glomerular endothelial cell proliferation. A recent study showed that human adipose angiotensinogen gene expression was associated with abdominal fat distribution in obesity,23 and that might have an effect on nephropathy. Moreover, the progression of diabetic nephro-

Figure 3 Change of serum creatinine per weight loss (a) and change of
urinary protein per weight loss (b) stratified by four grades of weight loss. No intergroup differences in change of serum creatinine (a) and change of urinary protein (b) were observed. Bars represent the mean7s.d.

(Figure 1b) and decrease in visceral fat (r 0.58, Po0.05) (Figure 2c). The Durinary protein showed no significant correlation with Dsystolic blood pressure, diastolic blood pressure, fasting blood glucose, HbA1c, TC, TG, HDL-C (data not shown) and subcutaneous fat (Figure 2d). During the 4 weeks in this study, mean 8-OHdG (considered an oxidative stress marker) decreased significantly (Table 1). However, the mean 8-OHdG showed no significant correlation with weight loss (r 0.15, P 0.66), decrease in serum creatinine (r 0.02, P 0.94), or decrease in urinary protein (r 0.23, P 0.50). The changes of serum creatinine per weight loss and the changes of urinary protein per weight loss were stratified by 4 grades of weight loss. No between-group differences were detected for the change of serum creatinine (Figure 3a) and for the change of urinary protein (Figure 3b).

Discussion
Usually, high-calorie low-protein diet is considered to protect renal function and prevent aggravation of proteinuria in diabetic nephropathy.57 Correction of the abnormal glomerular hemodynamics has been reported to be the mechanism by which low protein diet improves renal damage in diabetic nephropathy.11 However, the effect of weight loss on renal function in obese patients with diabetic nephropathy is not fully understood. In this study, we observed obese patients with diabetic nephropathy treated with a low-calorie normal-protein diet for 4 weeks under hospitalization. As a result, the mean serum creatinine decreased significantly by 41.6728.9 mmol/l, and the mean urinary protein also decreased significantly by 1.871.7 g/ day. Although the serum creatinine might be affected by reduction of skeletal muscle, the mean creatinine clearance increased showed an apparent increase in this study (not significant). These results suggest that a low-calorie diet
International Journal of Obesity

Formula diet for diabetic nephropathy A Saiki et al

1119
pathy has been suggested to involve oxidative stress.4 On the other hand, obesity has been shown to increase oxidative stress.24,25 In this study, 8-OHdG, a recognized oxidative stress marker, decreased significantly after 4 weeks. Although the decreases in body weight, serum creatinine and urinary protein did not correlate with the decrease in 8-OHdG, our results suggested that the decrease in oxidative stress might be one of the mechanisms by which weight loss improved renal function and proteinuria in obese subjects with diabetic nephropathy. The safety of weight loss in obese subjects with diabetic nephropathy was studied. Although rapid glycemic control might accelerate the progression of diabetic retinopathy,26 no patient showed progression of the retinopathy in this study. Serum albumin did not change significantly despite using the formula diet. It is important to note that all patients showed decreases of serum creatinine and urinary protein. To find out the optimal weight loss to improve renal function, the changes of serum creatinine and urinary protein per weight loss were stratified by grades of weight loss. The rates of improving renal function and proteinuria were not different among the grades of weight loss. These findings imply that for a short observation period of 4 weeks, serum creatinine and urinary protein are improved as long as there is weight loss, irrespective of the amount of loss. In other words, short-term use of formula diet to achieve weight loss appears to be a safe method to improve renal function and proteinuria in obese patient with diabetic nephropathy because the effect is achieved as long as the patient loses weight with no risk of excessive weight loss. In this study, we utilized a formula diet providing 6264 g of protein (0.91.2 g/kg) a day as a normal protein diet for prevention of protein catabolism. Usually, low-protein diet is recommended for diabetic nephropathy. Nakamura et al9 reported that alanine, glycine and arginine loads induced glomerular hyperfiltration in vivo. The amino-acid composition of ObeCure, a normal-protein diet, might not adversely affect diabetic nephropathy. In the future, we should compare a low-protein diet for obese subjects with diabetic nephropathy. In this study, weight loss might have improved nephropathy related to obesity, but whether weight loss improves diabetic nephropathy is not fully understood. We should conduct long-term study on the effect of weight loss on diabetic nephropathy in obese subjects. derived cytokines and oxidative stress might be the mechanisms by which weight loss improved renal function and proteinuria in diabetic nephropathy with obesity.

References
1 De Fronzo RA. Diabetic nephropathy: etiologic and therapeutic considerations. Diabetes Rev 1995; 3: 510564. 2 Mogensen CE. How to protect the kidney in diabetic patients with special reference to IDDM. Diabetes 1997; 46: S104S110. 3 The Diabetes Control and Complication Trial Research Group. The effect of intensive treatment of diabetes on the development and progression of long-term complications in insulin-dependent diabetes mellitus. N Engl J Med 1990; 329: 977986. 4 Kedziora-Kornatowska KZ, Luciak M, Blaszczyk J, Pawlak W. Lipid peroxidation and activities of antioxidant enzymes in erythrocytes of patients with non-insulin dependent diabetes with or without nephropathy. Nephrol Dial Transplant 1998; 13: 28292832. 5 Barsotti G, Ciardella F, Morelli E, Cupisti A, Mantovanelli A, Giovannetti S. Nutritional treatment of renal failure in type 1 diabetic nephropathy. Clin Nephrol 1988; 29: 280287. 6 Walker JD, Bending JJ, Dodds RA, Mattock MB, Murrells TJ, Keen H, Viberti GC. Restriction of dietary protein and progression of renal failure in diabetic nephropathy. Lancet 1989; ii: 14111415. 7 Zeller K, Wittaker E, Sullivan L, Raskin P, Jacobson HR. Effect of restricting dietary protein on the progression of renal failure in patients with insulin-dependent diabetes mellitus. N Engl J Med 1991; 324: 7884. 8 Praga M, Hernandez E, Andres A, Leon M, Ruilope LM, Rodicio JL. Effect of body-weight loss and captopril treatment on proteinuria associated with obesity. Nephron 1995; 70: 3541. 9 Nakamura H, Ebe N, Ito S, Shibata A. Renal effects of different types of protein in healthy volunteer subjects and diabetic patients. Diabetes Care 1993; 16: 10711075. 10 Mark K, Carlos J, Bruce N. Urinary 8-hydroxy-20 -deoxyguanosine as a biological marker of in vivo oxidative DNA damage. Proc Natl Acad Sci USA 1989; 86: 96979701. 11 Parving HH, Kastrup H, Smidt UM, Andersen AR, Feldt-Rasmussen B, Christiansen JS. Impaired autoregulation of glomerular filtration rate in type 1 (insulin-dependent) diabetic patients with nephropathy. Diabetologia 1984; 27: 547556. 12 United Kingdom Prospective Diabetes Study Group. Tight blood pressure control and risk of macrovascular and microvascular complications in type 2 diabetes: UKPDS 38. BMJ 1998; 317: 703713. 13 The Diabetes Control and Complications Trial Research Group. The effect of intensive treatment of diabetes on the development and progression of long-term complications in insulin-dependent diabetes mellitus. N Engl J Med 1993; 329: 977986. 14 Ohkubo Y, Kishikawa H, Araki E, Miyata T, Isami S, Motoyoshi S, Kojima Y, Furuyoshi N, Shichiri M. Intensive insulin therapy prevents the progression of diabetic microvascular complications in Japanese patients with non-insulin-dependent diabetes mellitus: a randomized prospective 6-year study. Diabetes Res Clin Pract 1995; 28: 103117. 15 Cianciaruso B, Bellizzi V, Minutolo R, Tavera A, Capuano A, Conte G, De Nicola L. Salt intake and renal outcome in patients with progressive renal disease. Miner Electrolyte Metab 1998; 24: 296301. 16 Allen TJ, Waldron MJ, Casley D, Jerums G, Cooper ME. Salt restriction reduces hyperfiltration, renal enlargement, and albuminuria in experimental diabetes. Diabetes 1997; 46: 1924. 17 Jennette JC, Charles L, Grubb W. Glomerulomegaly and focal segmental glomerulosclerosis associated with obesity and sleepapnea syndrome. Am J Kidney Dis 1987; 10: 470472.

Conclusion
Weight loss using formula diet improved the renal function and proteinuria in obese patients with diabetic nephropathy in a short-term study. The mean serum creatinine and mean urinary protein decreased significantly, and both serum creatinine and urinary protein correlated significantly with weight loss. These results suggest that weight loss in obese patients with diabetic nephropathy achieved by taking a formula diet might improve renal function and proteinuria safely in the short term, and the changes of adipocyte-

International Journal of Obesity

Formula diet for diabetic nephropathy A Saiki et al

1120
18 Kambham N, Markowitz G, Valeri AM, Lin J, DAgati VD. Obesityrelated glomerulopathy: an emerging epidemic. Kidney Int 2001; 59: 14981509. 19 Morales E, Angeles M, Leon M, Hernandez E, Praga M. Beneficial effects of weight loss in overweight patients with chronic proteinuric nephropathies. Am J Kidney Dis 1980; 41: 319327. 20 Stokholm KH, Brochner-Mortensen J, Hoilund-Carlsen PF. Increased glomerular filtration rate and adrenocortical function in obese women. Int J Obes 1980; 4: 5763. 21 Kasiske BL, Crosson JT. Renal disease in patients with massive obesity. Arch Intern Med 1986; 146: 11051109. 22 Wolf G, Hamann A, Han DC, Helmchen U, Thaiss F, Ziyadeh FN, Stahl RA. Leptin stimulates proliferation and TGF-b expression in renal glomerular endothelial cells: potential role in glomerulosclerosis. Kidney Int 1999; 56: 860872. 23 van Harmelen V, Elizalde M, Ariapart P, Bergstedt-Lindqvist S, Reynisdottir S, Hoffstedt J, Lundkvist I, Bringman S, Arner P. The association of human adipose angiotensinogen gene expression with abdominal fat distribution in obesity. Int J Obes Relat Metab Disord 2000; 24: 673678. 24 Erdos B, Snipes JA, Miller AW, Busija DW. Cerebrovascular dysfunction in Zucker obese rats is mediated by oxidative stress and protein kinase C. Diabetes 2004; 53: 13521359. 25 Sonta T, Inoguchi T, Tsubouchi H, Sekiguchi N, Kobayashi K, Matsumoto S, Utsumi H, Nawata H. Evidence for contribution of vascular NAD(P)H oxidase to increased oxidative stress in animal models of diabetes and obesity. Free Radic Biol Med 2004; 37: 115123. 26 Funatsu H, Yamashita H, Ohashi Y, Ishigaki T. Effect of rapid glycemic control on progression of diabetic retinopathy. Jpn J Ophthalmol 1992; 36: 356367.

International Journal of Obesity

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.