Prabhu Mukhyaprana M*,Sudha Vidyasagar **,Shashikiran U***,

*Asst Professor of Medicine,
**Professor, Department Of Medicine,
***Asst. Professor of Medicine,
Dr T.M.A Pai Hospital, Udipi, Kasthurba Medical College, Manipal,

Address For Correspondence
Dr.Prabhu Mukhyaprana M MD
Asst Professor of Medicine,
Dr T.M.A .Pai Hospital, Udupi,
Kasthurba Medical College, Manipal,
Karnataka 576119 
email:mmukhyaprana@yahoo.com   
Phone # 0820-2526503, Mobile-98452-48744


ABSTRACT
BACKGROUND: Obesity in Type 2 Diabetes is uncommon in Indian population, compared to western population. Markers of obesity are Body mass index (BMI) & waist Hip ratio (WHR). Does BMI by current definition will correlate well with diabetes related complications in Indian Diabetic population? 

AIM: To study clinical profile and complications among four groups of cases of type 2 diabetes classified based on BMI Lean, Normal weight, Over-weight, and Obese Diabetics.

SETTING & DESIGN: Diabetes clinic Dr TMA Pay Hospital, Udupi attached to K M C Manipal. 
Descriptive case series analysis of 500 serial type 2 diabetics

MATERIALS & METHODS: 500 Type 2 diabetic patients from Diabetic clinic between July 2000 to January 2001 divided into 4 groups based on body mass index (BMI) were our study subjects. Detailed clinical examination and screening for micro/macro vascular complications was done in all patients

RESULTS: Lean diabetics were 7.4%. Majority of diabetics had normal weight (65%). 24% belonged to overweight group; only2.6% of diabetics were obese. Incidence of hypertension in lean diabetic was 16.2%, whereas it was 38.8% in normal weight diabetics 42% in over weight diabetics and 61.5% in obese diabetics. The incidence of ischemic heart disease (I.H.D) was very low in lean diabetics (2.7%); it was 13.88% in normal weight diabetics, 12% in over weight diabetics and 23% in obese diabetics. As body mass increases, the prevalence of macro vascular complications was increasing. But micro vascular complications prevalence was similar in all groups. Fasting and postprandial sugar levels were significantly higher in lean diabetics. Lean diabetics had significantly favorable lipid profile compared to other groups.

CONCLUSIONS: Obesity as defined body mass index is uncommon in Indian context. However subtle signs of obesity like waist hip ratio are abnormal even in diabetics with normal and lean BMI. Lean diabetics like all other groups are prone for micro vascular complications, whereas over-weight and obese diabetics are prone for macro vascular complications like I.H.D. and hypertension. Lean diabetics have more severe hyperglycemia and favorable lipid profile compared to other groups.

KEY WORDS: BMI, Waist Hip Ratio, Lean Diabetics, and micro/macro vascular complications.

Introduction:
Diabetes is the most common metabolic disorder all over the world. The incidence of diabetes is showing an alarming rise in developing countries, particularly in India (1). Obesity is considered as part of syndrome X in pathogenesis of type 2 diabetes. 60-80% of the diabetics in developed countries are obese (2) whereas in India we find that clinical profile of diabetics is different (6,7,8,9.10)
Most patients attending our diabetic clinic are not obese as defined by existing parameters such as body mass index (BMI). It has been interesting to note that most patients fall in normal weight group and some even lean. Since obesity does contribute in a considerable way to complications of diabetes, we thought it would be worthwhile to study if BMI had any implication on the complications of diabetes. Since the drug of choice has been different in diabetic based on their BMI, redefinition of obesity may help in making these clinical decisions.
We have a diabetic clinic with significant number of lean type 2 diabetics. As there were no major published studies done in South Karnataka diabetic population we undertook a study to look whether variations in body weight affect the clinical presentation, co-morbid illnesses, and the incidence of complications in diabetic patients. 

We studied 500 type2 diabetic patients from our Diabetic clinic, both outpatient and in-patient, between July 2000 and January 2001. Patient was diagnosed as type 2 diabetic based on American Diabetes Association 2000 (11). A detailed history was taken about the duration of diabetes, age of onset, family history and presence of complications. Height and weight were recorded in all cases. BMI was calculated based on formula weight (in kg) / height2 (in mt). 

Patients were divided in to four groups based on the BMI as below:

<18.5 Lean Diabetics. 
18.5-<25 Normal Weight Diabetics
25-<30 Over-weight Diabetics
>30 Obese Diabetics

Waist measurement was taken as abdominal circumference at midpoint between the costal margin and anterior superior iliac spine. Hip measurement taken as maximum diameter at the greater trochanter (3). Waist Hip Ratio (WHR) was calculated in each case. Waist Hip Ratio was considered abnormal, if >0.95 for males and >0.8 for females(3). Patients were clinically screened for micro vascular and macro vascular complications. Patients were considered as hypertensive if blood pressure was >140/90 mm of hg. Patients were considered as having ischemic heart disease based on ischemic changes in the electrocardiogram (ECG) or by demonstrating hypo kinetic or akinetic segment in the echocardiogram. Ophthalmoscopy was done diagnose diabetic retinopathy. Neuropathy was diagnosed based on subjective symptoms or objective evidence in the form loss of ankle jerks or glove and stocking type of anesthesia. Peripheral vascular disease was diagnosed based on diminished/absent pulses in clinical examination or arterial Doppler. Tuberculosis was diagnosed based on sputum positivity and chest x ray. Fasting, postprandial glucose, fasting lipid profile and other relevant investigations were done in each case.

Patients were followed up regularly in the diabetic clinic. Any change in weight and BMI were recorded. Lean diabetics, whose BMI increased to normal limits after control of hyperglycemia or infection, were considered as normal weight diabetics. Lean diabetics were considered in the group only if their BMI was persistently below 18.5 in spite of control hyperglycemia and after treatment of associated infection causing weight loss such as tuberculosis. Statistical analysis was done using Microsoft excel software. Data is presented in percentage in categories. Statistical tests used included chi square test and ANOVA.


RESULTS:
Out of 500 patients studied, 325 (65%) belonged to normal weight diabetes group, 124 (24%) over weight group, 38 (7.1%) patients were lean diabetics and 13 (2.6%) were obese diabetics.

Mean age of onset of diabetes in lean diabetics was 60.34±13.5 years, whereas it was 58.2±9.8 years for obese diabetics. Lean diabetics were elderly compared to other groups. Values were statistically significant (F value 7.830028, p value < 0.001). There was no significant difference in duration of diabetes in different groups (Table 1) (F=0.8684, p =0.457)

Most of obese diabetics were females (70% of obese diabetics), whereas most of lean diabetics were males (65% of total lean patients). Sex differences in groups were statistically not significant (Table2). Family history of diabetes was present in 62.5% normal weight, 52% over weight, 46% of lean and 45% of obese diabetics.
There was a linear increase in number of patients having abnormal waist hip ratio3 with increase in BMI. Among lean diabetics 48% had abnormal waist hip ratio whereas 79% of normal weight, 80% of overweight and 96% of obese diabetics had abnormal waist hip ratio.

Micro vascular complications were found in similar proportion in all groups (Table3). Retinopathy was present in 27% of lean and 15% of obese diabetics. Neuropathy was the commonest complication among all groups was seen in 35% of lean diabetics, 34% of normal weight diabetics, 32% of over weight and 54% of obese diabetics. Nephropathy was more common in over weight (10.4%) and obese diabetics (15%) Values were statistically significant (p =0.0342).

Table 4 shows macro vascular complications in percentages various groups. Here it was observed that as the body mass increases, the incidence of hypertension and I.H.D. also increases. Incidence of hypertension in lean diabetics was 16.2%, whereas it was 39.09% in normal weight 41.6% in over weight and 61.5% in obese diabetics. Values were statistically very significant (p=0.0002) Incidence of I.H.D was very low in lean diabetics (2.7%); it was 13.84% in normal weight 12% in over weight 23.07% in obese diabetics. However values were statistically not significant (p=0.163) Evidence of peripheral vascular disease was present in 8% of lean and 9% of normal weight diabetics.
Incidence of tuberculosis was very high in lean diabetics (26.6%). (Table 4) whereas tuberculosis was seen in 4% on of normal weight diabetics. Values were statistically very significant (p <0.0001) none of our overweight and obese diabetics had tuberculosis.
It was observed that Fasting and postprandial glucose levels were higher in lean diabetics compared to other groups (Table 5). Values were statistically not significant for fasting but very significant for post prandial glucose (FBS p =0.129 whereas for PPBS p value 0.008) 
Lipid profile analysis (Table 6) showed favorable lipid profile in lean diabetics with mean total cholesterol 174.27±68.66 mg/dl, Low density cholesterol (LDL) 97.37±50.66 mg/dl triglycerides 124.28±57.88 mg/dl and High density (HDL) cholesterol 44±15.63 mg/dl. (Table7). As BMI increased from normal to over weight and obese ranges, L.D.L, Triglycerides and total cholesterol showed rising trends. Values were statistically very significant (p <0.0001,0.0001 & 0.00004 respectively)

Discussion: 

Most of the diabetics in developed countries are obese. However in India we have a significant number of diabetics who are either normal weight or even under weight(4,10). A similar pattern has also been observed in South Korea (5).
Incidence of low body weight diabetes in various Indian studies ranges from 1.6% as in Ramachandran et al (12) study to as high as 28% as in Tripathi et al study (6). Mohan et al reported an incidence of 3.5%(7). In our study we had an incidence of 7.4%, which was comparable to an earlier study (8). Most of our diabetics had normal body weight (65%). Obesity in our series was uncommon at 4% of all diabetics. 
Family history of diabetes was present in 45% of lean and 62.6% in normal body weight diabetics. These results were similar to studies by Banerji et al and Kannan et al studies (13, 14). However, absence of a strong family history in both parents and later onset distinguishes these lean diabetics from maturity onset diabetes in the young (MODY).
The diagnosis of obesity is based on BMI and WHR. Waist Hip Ratio is considered as a more sensitive indicator of obesity since abdominal fat deposition is part of syndrome X.A significant number of our lean diabetics (48%) had abnormal WHR. In this aspect they were obese even though they were underweight by BMI. Since a significant number of lean (48%) and normal weight diabetics (79%) had abnormal Waist Hip Ratio, it is probable that obesity based on BMI > 30 may not be relevant in the Indian context. Snehalatha et al (15) study suggested the healthy BMI cut off for urban Indian should be <23 kg/m2, and cut off for WHR for 0.89 for men and 0.81 for women. They have also suggested it is appropriate to use waist circumference (WC) as an index for upper-body adiposity. Possible explanations could be Asian Indians tend to have more visceral adipose tissue, higher insulin resistance despite having lean BMI. (16, 17) The WHR/ waist circumference may thus be a more sensitive indicator of obesity in Indians

In our study proportion of patients having micro-vascular complications was similar in all groups of patients. Micro-vascular complications are related directly to duration of diabetes and glycemic control. In this aspect lean and normal body weight diabetics are the same as other groups. In our study of retinopathy were seen in27% lean diabetics; nephropathy was seen in 6.6% that is similar to an earlier study8. Mohan et al (7) reported an increased prevalence of retinopathy (both background and proliferative), nephropathy and neuropathy in lean diabetics. Neuropathy was the commonest micro vascular complication in all groups of diabetics in our study. Peripheral neuropathy (49%) was the commonest presenting complication among lean diabetics in a study by Das et al (8).

Lean diabetics had higher incidence of tuberculosis (26.6%) compared to other groups. Lean diabetic patients continued to have low body weight even after treatment for tuberculosis. Sahay (18) et al and Kannan (14) et al got similar observations in their studies.

Lean diabetics are less prone to develop macro-vascular complications like hypertension and IHD. Incidence of hypertension was 8.9% and IHD was 10.2% in Nigam et al study (9). In our study only 2.7% of our lean diabetics had IHD and 16.2% had hypertension. The probable low incidence of IHD and hypertension in these patients may be because of less obesity in these patients. We also found linear increase in the incidence of hypertension and IHD with an increase in the BMI. Hypertension was seen in 61% and IHD 23% in our obese diabetics. However Spanish study by De Pablo et al (19) showed no differences in the clinical profile or complications in diabetics with varying BMI and waist hip ratio.

Lean diabetics have more severe hyperglycemia and poor metabolic control according to various earlier studies (7,8,9,20). In our study also statistically significant higher fasting and postprandial sugar levels in lean diabetics was found compared to other diabetics. This has been explained based low beta cell reserve in these patients. Lean diabetics are insulinopenic and highly insulin sensitive (8, 10). Most of these patients may require insulin with oral hypoglycemic agents (OHA) for control of hyperglycemia. According to study by Kannan et al 1433% of lean diabetics doing well on OHA s had to be shifted to insulin. According to Italian study by Pointoroly et al (12) secondary failure to oral hypoglycemic agents in non-obese patients with non-insulin-dependent diabetes is related to reduced insulin release. These data indicate that secondary failure is more frequent in lean patients with NIDDM, and is related to reduced insulin release (8, 12). In our study also we found that a significant number of lean diabetics required insulin for control of sugars. In Das et al study 8 basal, post glucose and post glucagon values and insulin release value (IRV) were significantly lower in lean diabetics. There was no statistically significant difference in corresponding C peptide levels. A possible explanation for this is excess extraction of insulin by liver in lean diabetics compared to others. In obese patients, as in non-obese patients, the lower beta-cell function seems likely to be the major pathogenic factor in the appearance of secondary failure, while being overweight plays only a minor role, thus showing that Type 2 diabetes is the same disease in obese and non obese patients (21).

Analysis of the lipid profile in various groups showed interesting results. All the parameters of lipid profile were lower in lean diabetics compared to all other groups. These differences were statistically very significant. Studies by Banerji et al 13and Das et al8 had showed slight increase in triglycerides and HDL in lean diabetes. Japanese study by Ikeda et al (22) showed no major differences in lipid profile in lean diabetes, irrespective of glycemic status.

There are still controversies regarding role of body mass index predicting mortality. In a study by Ross et al (23) there was no relationship between body weight and mortality, as leanness did not confer any mortality benefit in diabetics. It has been suggested that normal body weight is the most desirable in diabetic patients.

Our study has limitations, as it was hospital based in tertiary care setting incidence of complications might be higher compared to general population or primary care setting. We did not do fasting insulin, C peptide levels and GAD antibodies in our lean diabetics due to financial constraints. Actually some of these patients may be late onset type 1 diabetics or latent autoimmune diabetes (LADA).


Conclusions:
Most of diabetics in our population have normal body weight (65%). However, if obesity were to be defined as increase in WHR, a significant number of normal and even lean diabetics could be termed obese. Lean diabetics form significant number (7.4%) of type 2 diabetics. Lean diabetics are prone for micro vascular complications like all other diabetics. Neuropathy is the commonest micro vascular complication among all diabetics. Tuberculosis is more common in lean diabetics. Incidence of macro vascular complications like hypertension and ischemic heart disease increases as body weight (BMI) increases. Lean diabetics have more severe hyperglycemia among all diabetic groups pointing to difficulty in diabetic control in these patients. 

Future directions: As our study was hospital based and has its limitations, further epidemiological studies are needed in Indian Diabetic population to look specifically for complications of diabetes and relationship to obesity. Secondly we may need to redefine obesity in Indian context.





References:

1. Jonathan.E.Show, David McCarty, Paul Z. Zimmet, Maximillan De Courtan. 
Type 2 diabetes world wide according to new classification and criterion;
Diabetes Care,Vol 23, 2000,Suppl 2.
2. WHO study group on diabetes mellitus. World Health Organization technical report series 1985; 727.
3. Lean M.E.J, Morrison Waist circumference as a measure for indicating need for weight management .BMJ, Vol 311 1995 
4. Das. S Introduction, Low body weight Type 2 Diabetes Mellitus. Technical series of Indian College of Physicians
5. Min HK. Department of Internal Medicine, Seoul National University Hospital, Korea. Non-insulin-dependent diabetes mellitus (NIDDM) in Korea. Diabetes Med 1996 ;( 9 Suppl 6):S13-5
6. Tripathy B B and Kar B C Observations and clinical patterns of diabetes mellitus in India : Diabetes 1965 ; 14: 404-12
7. Mohan V, Vijayaprabha R, Rema M, Premalatha G, Poongothai S, Deepa et al Madras Diabetes Research Foundation, India. Clinical profile of lean NIDDM in South India. Diabetes Res Clin Pract 1998 Aug;41(2):149-50
8. Das S, Samal K.C, Baliarsinha A.K., Tripathy B.B., Lean (Underweight) NIDDM -peculiarities and differences in metabolic and hormonal status. JAPI 1995; Vol 43:339-42.
9. Nigam A .Clinical and epidemiological characteristics of NIDDM patients with low body weight. Diabetes 1997; 46(1):97 A.
10. Siddharth Das Low body weight Type 2 Diabetes mellitus.Int J. Diab.Dev .Countries 2003 Vol 23.
11. Americal Diabetes Association 2000:Diabetes Care 23 Supp 1) 2000
12 Ramachandran A, Snehalatha C, Latha E, Vijay V, Viswanathan M: Rising prevalence of NIDDM in urban population in India. Diabetologia 40:232-237, 1997[Medline]
13. Banerjee S,,Paul UK.Clinical ,biochemical and hormonal profile. Experiences from
Calcutta In:Low body weight type 2 Diabetes Mellitus, (Ed),Siddarth Das ,Association of Physicians of India, Mumbai,1999,70-75
14 Kannan K.Lean type II diabetes mellitus - a distinct entityIn: Kapur A(ed).Proceedings of the Novo -Nordisk diabetes update.Health care communication,Bombay 1993:147-51.
15 Chamukuttan Snehalatha, Vijay Viswanathan, Ambady Ramachandran, Cutoff Values for Normal Anthropometric Variables in Asian Indian Adults. Diabetes Care 26:1380-1384, 2003
16 Banerji MA, Faridi N, Atluri R, Chaiken RL, Lebovitz HE: Body composition, visceral fat, leptin and insulin resistance in Asian Indian men. J Clin Endocrinol Metab 84:137-144, 1999
17 Chandalia M, Abate N, Garg A, Stray-Gundersen J, Grundy SM: Relationship between generalized and upper body obesity to insulin resistance in Asian Indian men. J Clin Endocrinol Metab 84:2329-2335, 1999
18. Sahay BK.Profile of lean-NIDDM as seen in Hyderabad.IN:Kapur A (ed). Proceedings of the Novo -Nordisk diabetes update.Health care communication,Bombay 1993:147-51
19. De Pablos PL, Martinez J, Martinez MP, Doreste JA. Endocrinology Dept, Hospital, Las Palmas, Spain. Prevalence of micro- and macroalbuminuria in a Canarian population of type 2 diabetic patients. Relationship with blood pressure, lipid profile, obesity and metabolic control. Diabetes Metab 1998, 4:337-43
20 Yajnik C S, Shelgikar K M, Sardesai B S et al. Association of obesity with clinical, biochemical, metabolic and endocrine measurements in newly diagnosed NIDDM patients: Novo Nordisk Diabetes Update Proceedings 1993;Health care communications ,Bombay 1993; 139-46
21 Prando R, Cheli V, Melga P, Giusti R, Ciuchi E, Odetti P:.Department of Internal Medicine, University of Genoa, Italy. Is type 2 diabetes a different disease in obese and nonobese patients? Diabetes Care 1998 (10):1680-85
22 Ikeda T, Ochi H, Ohtani I, Fujiyama K, Hoshino T, Tanaka Y et al : Department of Internal Medicine, Tottori University School of Medicine, Yonago, Japan. Serum lipid and apolipoprotein levels in non-hypertensive lean NIDDM patients. J Intern Med 1991 ; 230(2):131
23 Ross C, Langer RD, Barrett-Connor E. Given diabetes, is fat better than thin?
Diabetes Care 1997 (4):650-2 

© Copyright of articles belongs to the respective authors unless otherwise specified.Verbatim copying, redistribution and storage of this article permitted provided no restrictions are imposed on the access and a hyperlink to the original article in Calicut Medical Journal maintained. All opinion stated are exclusively that of the author(s).
Calicut Medical Journal upholds the policy of Open Access to Scientific literature.