Obesity-related non-communicable diseases: South Asians vs White Caucasians

Abstract

South Asians are at higher risk than White Caucasians for the development of obesity and obesity-related non-communicable diseases (OR-NCDs), including insulin resistance, the metabolic syndrome, type 2 diabetes mellitus (T2DM) and coronary heart disease (CHD). Rapid nutrition and lifestyle transitions have contributed to acceleration of OR-NCDs in South Asians. Differences in determinants and associated factors for OR-NCDs between South Asians and White Caucasians include body phenotype (high body fat, high truncal, subcutaneous and intra-abdominal fat, and low muscle mass), biochemical parameters (hyperinsulinemia, hyperglycemia, dyslipidemia, hyperleptinemia, low levels of adiponectin and high levels of C-reactive protein), procoagulant state and endothelial dysfunction. Higher prevalence, earlier onset and increased complications of T2DM and CHD are often seen at lower levels of body mass index (BMI) and waist circumference (WC) in South Asians than White Caucasians. In view of these data, lower cut-offs for obesity and abdominal obesity have been advocated for Asian Indians (BMI; overweight >23 to 24.9 kg m⁻² and obesity ⩾25 kg m⁻²; and WC; men ⩾ 90 cm and women ⩾80 cm, respectively). Imbalanced nutrition, physical inactivity, perinatal adverse events and genetic differences are also important contributory factors. Other differences between South Asians and White Caucasians include lower disease awareness and health-seeking behavior, delayed diagnosis due to atypical presentation and language barriers, and religious and sociocultural factors. All these factors result in poorer prevention, less aggressive therapy, poorer response to medical and surgical interventions, and higher morbidity and mortality in the former. Finally, differences in response to pharmacological agents may exist between South Asians and White Caucasians, although these have been inadequately studied. In view of these data, prevention and management strategies should be more aggressive for South Asians for more positive health outcomes. Finally, lower cut-offs of obesity and abdominal obesity for South Asians are expected to help physicians in better and more effective prevention of OR-NCDs.

Introduction

South Asian countries are experiencing a rapid increase in obesity-related non-communicable diseases (OR-NCDs), including type 2 diabetes mellitus (T2DM), hypertension, dyslipidemia and coronary heart disease (CHD).^{1, 2, 3, 4} Insulin resistance and clustering of proatherogenic, cardiovascular risk factors (the metabolic syndrome) are frequently seen in South Asians, even at a young age.^{5, 6, 7} Increasing trend of obesity, both in adults and children, is directly responsible for rapid increase in NCDs.^{7, 8, 9, 10, 11, 12}

It has increasingly become apparent that South Asians have a higher risk of NCDs compared with White Caucasians, which may be contributed by a number of factors, discussed subsequently.^{13, 14, 15, 16} Early onset and heightened risk of OR-NCDs in South Asians as compared with White Caucasians have important implications for prevention and therapy. An understanding of the different mechanisms and factors responsible for increased susceptibility and different outcomes of OR-NCDs would give physicians and scientists a better insight in research and management of these chronic diseases in South Asians.^{8, 11} This would also be in line with the global strategy of WHO (World Health Organization) to reduce the burden of NCDs through integrated prevention/control of risk factors at individual, family, community and population levels.¹⁷

In this review, we attempt to analyze the evidence, possible mechanisms and determinants responsible for increased tendency of OR-NCDs, as well as discuss management issues, in South Asians, as compared with White Caucasians, and also other ethnic groups.

Definition of ‘White Caucasians’ and ‘South Asians’

Natives of first or subsequent generation immigrants originating from the Indian subcontinent (India, Pakistan, Bangladesh, Nepal and Sri Lanka) are referred to as ‘South Asians’. For this ethnic group, ‘Indo-Asians’ has also been used by various authors. ‘White Caucasians’ (henceforth referred as ‘Whites’ in the text) are defined as native Europeans/European immigrants, as well as indigenous White individuals in any other country. However, some studies have not referred ‘Caucasians’ as Whites (though probably implied), and the term ‘Caucasians’ has been used as such there. The term ‘Europeans’ has been retained wherever referred to in the original text.

Methodology for literature search

The literature search has been performed using the key words ‘obesity, insulin resistance, diabetes, hypertension, the metabolic syndrome, cardiovascular risk, coronary artery disease, Asian Indians, South Asians, Indo-Asians, Whites, Caucasians, White Caucasians and Europeans’ from medical search engine Pubmed (National Library of Medicine, Bethesda, MD, USA) from 1966 to December 2009. Manual search for other important references, and medical databases were also performed. Well-designed studies and those published in high impact journals were given preference for citation.

Increasing OR-NCDs in South Asians: lifestyle-driven and multi-factorial origin

Increasing OR-NCDs, including T2DM, hypertension, dyslipidemia and CHD, are seen in urban and even rural South Asian populations due to several reasons, such as increasing urbanization, rural-to-urban migration, mechanization and increased life expectancy (‘epidemiological transition’), ‘westernized lifestyle’ (high carbohydrate, high-fat, low-fiber diet with decreasing trends in physical activity; ‘nutrition and lifestyle transitions’).^{11, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28}

Increase in T2DM, hypertension and CHD in South Asians

In 2000, 44% of the burden of disease in South Asia measured in disability-adjusted life years lost was attributed to NCDs, and these figures are expected to rise, with cardiovascular disease (CVD) being a major contributor to premature mortality and morbidity.¹⁷ At present, India has highest global number of patients with diabetes, which is estimated to be 50.8 million in 2010 and 79.4 million in 2030.^{29, 30, 31, 32, 33} Increasing trend in Asian Indians living in India is more apparent in urban populations (exponential trend R²=0.744) than rural populations (R²=00.289).³⁴ Although the prevalence of diabetes was reported to be 12–14% in urban India, the rural population showed a relatively lower prevalence of 4–5% in 2005–06.³⁴ Neighboring South Asian countries also show a high prevalence of diabetes: Pakistan (8–10% in urban population and 2–10% in rural population),^{35, 36} Bangladesh (7–9% in urban population),³⁷ Sri Lanka (13–14% in urban areas)^{38, 39} and Nepal (nearly 19% in urban areas).⁴⁰ Furthermore, a meta-analysis has shown significant increase in hypertension in India in both urban and rural areas.⁴¹ Other South Asian countries also show high prevalence of hypertension: 19% in Pakistani adults (⩾15 years),⁴² and 19% in 2005 in Sri Lanka.³⁹ It is projected that by 2020 India will have more individuals with CVD than any other region in the world.⁴³ With an estimated 61 million cases in 2015, mortality attributed to CHD in India is expected to rise by 103.1% in men and 90% in women from 1985 to 2015 (approximately one-third of the total mortality).^{44, 45, 46} Nearly 22% of 1.4 million deaths in Pakistan (2005); 21% of 0.24 million deaths in Nepal (2002); 23% of 1.1 million deaths in Bangladesh (2002) and 34% of 0.15 million deaths in Sri Lanka (2002) were attributed to CVD.^{47, 48, 49, 50}

Obesity and body composition

Prevalence of obesity in South Asian countries is increasing in both urban and in rural areas.^{7, 11, 12, 27, 28} In urban India, data show a prevalence of 10 and 15.1% (2003);²⁵ 20.8 and 32.3% (2004);⁵¹ and 43.2 and 47.4% (2007) in men and women, respectively.⁵² Interestingly, a high prevalence has been reported in economically disadvantaged adults residing in urban slums (14%),²² specifically in postmenopausal women (28%).⁵³ Increase has also been reported in rural areas: from 8% reported in 1997⁵⁴ to 32.4% in men and 41.4% in women in 2007.⁵⁵ Of concern, childhood obesity is also increasing: from 16% in 2002 to about 24% in 2006 in North India,¹² and from 4.9% in 2003 to 6.6% in 2005 in South India.⁵⁶ Other South Asian countries show similar increasing trend in obesity: 20.3% in men and 36.5% in women (2005) in Sri Lanka³⁹ in contrast to earlier prevalence reported between 15 and 18% (1996).⁵⁷ It is to be noted that different investigators have used different criteria for diagnosis of obesity; most recent studies have used lower cut-offs for South Asians,^{11, 27, 28} which may also increase prevalence rates.

BMI and body fat: relationship and definitions

By measurement of body mass index (BMI, a ratio of height and weight), it is not possible to distinguish between fat and fat-free mass.⁵⁸ Several investigators have shown that BMI systematically underestimated percent body fat in South Asians.^{58, 59, 60, 61} Asian Indians in Singapore showed lower BMI (∼3 kg m⁻²) than other ethnic groups at any given percent body fat.^{24, 62} Furthermore, BMI did not give accurate estimate of body fat distribution, and in particular, significantly underestimated visceral adipose tissue in non-European populations.⁶³ In a recent Consensus Statement for Asian Indians, overweight is defined as BMI between 23 and 24.9 kg m⁻² and obesity as ⩾25 kg m⁻², values lower than internationally accepted cut-offs (overweight 25–29.9 kg m⁻² and obesity ⩾30 kg m⁻²) (Table 1).⁶⁴

Table 1 Cut-offs of obesity and abdominal obesity for Asian Indians vs International Criteria

Abdominal adiposity

High prevalence of abdominal obesity is characteristic of South Asians, even at BMI levels <25 kg m⁻² and is seen at a young age.^{14, 20, 22, 24, 65, 66, 67} Both migrant and British-born South Asians had higher waist-to-hip circumference ratio than the Italian/British populations.⁶⁸ A study on migrant Asian Indians in the United States of America showed that waist circumference (WC) of South Asians was 10 cm lower than Whites with similar percent body fat.⁶⁹ Similar to BMI, because of higher morbidities seen at lower level of WC in Asian Indians, cut-offs for diagnosis of abdominal obesity have been advocated as ⩾90 and ⩾80 cm, as opposed to international cut-offs of ⩾102 and ⩾88 cm for men and women, respectively (Table 1).^{64, 70}

Truncal and abdominal adipose tissue depots

Overall, South Asians have more fat in various abdominal fat depots (Table 2). In a comparative study of Asian Indians vs Whites of European origin in the United States of America, for similar value of BMI, migrant Asian Indians had significantly greater total abdominal fat and intra-abdominal adipose tissue.⁷¹ Some investigators have reported that truncal subcutaneous (SC) adipose tissue (measured by subscapular and suprailiac skinfolds and by abdominal magnetic resonance imaging) is thicker in South Asians than in Whites^{72, 73, 74, 75} and correlates with the metabolic syndrome.⁷⁶

Table 2 Anthropometric differences between South Asians/Asian Indians and Whites/Europeans

Fat deposition in unusual sites

‘Ectopic fat’ (intramyocellular, hepatic and more recently myocardial lipids) represents the deposition of triglycerides within cells of non-adipose tissues, such as liver, muscles and so on. These sites usually contain only a small amount of fat, but recent data have shown that such ‘ectopic’ adipose tissue deposition could potentially disrupt glucose–insulin metabolism. Increasing evidence shows causal relationships between hepatic steatosis (without significant alcohol intake; non-alcoholic fatty liver disease, NAFLD) and hepatic insulin resistance, the metabolic syndrome and even CVD.^{77, 78, 79} In a recent case–control study, we have reported that urban Asian Indians with NAFLD had significantly higher fasting hyperinsulinemia, the metabolic syndrome and glucose intolerance than those without NAFLD.⁸⁰ Furthermore, we also showed that normoglycemic, normolipidemic and normotensive subjects with NAFLD had significant derangements of hepatic gluconeogenesis pathway as assessed by ³¹phosphorous magnetic resonance spectroscopy (^31PMRS).⁸¹ In an important comparative study in healthy, lean (mean BMI; 21–23 kg m⁻²) Asian Indians and Whites in the United States of America, hepatic triglycerides content, assessed by proton MRS, was two-fold higher in former (1.94 vs 0.75%, respectively; P<0.001). These authors further showed that increased hepatic triglyceride concentrations in Asian Indians was associated with higher insulin resistance (Table 3), plasma interleukin-6 (1.60 vs 0.78 pg ml⁻¹, respectively; P<0.001), plasma leptin (4.7 vs 2.4 ng ml⁻¹, respectively; P<0.001) and lower adiponectin (6.2 vs 7.3 μg ml⁻¹, respectively; P=0.02) concentrations than Whites.⁸² These preliminary data suggest that Asian Indians appear to be predisposed to develop hepatic steatosis, associated with hepatic insulin resistance and increased levels of proinflammatory cytokines.⁸² Interestingly, the prevalence of NAFLD in urban India parallels prevalence of the metabolic syndrome in India.⁸³

Table 3 Hepatic triglycerides, glycemia and insulin sensitivity in Asian Indians vs Caucasians

Intramyocellular triglyceride deposition, mostly assessed in the soleus muscle, is a potential marker for insulin resistance as shown by the studies carried out in Whites.^{84, 85} Comparative data show that intramyocellular triglyceride content was higher in Asian Indian men compared with Whites (1.03 vs 0.74%; P=0.001),⁸² but correlation of intramyocellular triglyceride with percent body fat, waist-to-hip circumference ratio and insulin sensitivity was shown in Europeans, but not in South Asians.^{86, 87, 88, 89}

Anecdotal data suggest that some previously untested phenotypic markers may predict insulin resistance, the metabolic syndrome and even T2DM in Asian Indians. According to our recent study, the presence of excess dorso-cervical fat (mild ‘buffalo hump’ termed as ‘dorso-cervical lipo-hypertrophy’) and excess fat deposition under the chin (‘double chin’) signify heightened risk for the metabolic syndrome in urban Asian Indians, and may be used as novel phenotypic markers.⁹⁰ Interestingly, ‘buffalo hump’ is a prevalent condition (28.5%) seen in obese NAFLD patients in the United States of America.⁹¹ Excess skin and fatty tissues beneath the jaw lead to a ‘double chin’, often seen in obese individuals and is reported to be a significant predictor of diabetes in patients of familial partial lipodystrophy.⁹²

Subcutaneous adipocyte size

Some studies show that increased SC abdominal adipocyte size independently predicts insulin resistance and T2DM. Weyer et al.⁹³ have showed that mean adipocyte size was 11 and 19% higher in subjects with impaired glucose tolerance and T2DM, respectively, as compared with normoglycemic individuals (P<0.001), independent of insulin resistance.⁹³ Furthermore, in normal subjects, 33% of whom developed diabetes, enlarged mean SC abdominal adipocyte size, but not high percent body fat, was an independent predictor of diabetes.⁹³ Interestingly, histological studies of SC adipose tissue have shown South Asians to have larger adipocytes as compared with Caucasians (3491 vs 1648 μm², respectively; P=0.0001), at comparable levels of total body fat, and comparable intra-abdominal fat and SC abdominal fat (Figure 1). In this study, increased adipocyte size in Asian Indians was inversely correlated to glucose disposal rate (r=−0.57; P=0.0008) and plasma adiponectin concentrations (r=−0.71; P<0.0001).⁶⁹ These interesting data need further investigation.

Figure 1

Comparative pictures of enlarged adipocytes from South Asian (left) and White (right) volunteers. Both images are obtained with SPOT digital camera using × 10 magnification. Source: Chandalia et al.⁶⁹

Skeletal muscle mass

Whereas body fat is higher, skeletal muscle mass is lower in South Asians than in Whites, particularly so in women.^{72, 94, 95} Asian Indian males have less muscle mass in the lower extremity than Swedish males.⁷⁴ Recent study by Unni et al.⁹⁶ showed that relative increase in BMI and reduced muscle mass were associated with reduced insulin sensitivity in lean Asian Indian men.⁹⁶ It is possible that low muscle mass in relation to excess adiposity in Asian Indians may have an important role in pathogenesis of insulin resistance and T2DM; however, more studies are needed.

Earlier onset and increased complications of OR-NCDs in South Asians

The following section summarizes the differences (qualitatively and sometimes quantitatively) in clinical and metabolic profiles of OR-NCDs in South Asians, comparing them to other ethnic groups, mostly Whites/Europeans. Phenotypic differences in obesity and body composition between South Asians and Whites are in part responsible for greater metabolic perturbations in the former.⁹⁷ These and other ethnic differences, as discussed below, have important implications for pathophysiology, management and prevention of OR-NCDs.^{66, 71, 72, 98, 99}

Insulin resistance and the metabolic syndrome

Several studies show that insulin resistance is prevalent and occurs at an early age in South Asians (Table 4).^{7, 12, 71, 94, 97, 100, 101, 102, 103} Asian Indians had comparative lower insulin sensitivity index and lower rate of glucose disposal than European-origin Whites in the United states of America, even after adjustment for both total body fat and truncal skinfold thickness (P=0.04).⁹⁹ South Asian children in the United Kingdom also showed higher degree of hyperinsulinemia, correlating with adiposity, than in White children of similar age.⁹⁷ Similarly, the metabolic syndrome has been seen to be more prevalent in South Asians as compared with Whites (Table 5).^{94, 104, 105, 106, 107, 108} Furthermore, South Asians having the metabolic syndrome showed higher diastolic blood pressure, plasma triglycerides, fasting insulin and lower high-density lipoprotein-cholesterol (HDL-C) levels compared with UK Whites.¹⁰⁹

Table 4 Differences in insulin resistance parameters between South Asians/Asian Indians vs Whites/Europeans

Table 5 Prevalence of the metabolic syndrome in South Asians vs Whites/Europeans

T2DM and its complications

T2DM manifests nearly a decade earlier,^{4, 110, 111, 112, 113, 114} and occurs at a lower BMI and WC in South Asians than in Whites.^{115, 116} Importantly, South Asians are more predisposed to develop microvascular and macrovascular complications than European Whites.¹¹⁷ Several investigators have reported increased predisposition to microalbuminuria and diabetic nephropathy in South Asians (Figure 2).^{118, 119, 120} Notably, in Sweden, nearly 40-fold increased risk for end-stage renal disease consequent to T2DM was reported in Indo-Asians compared with native Dutch Caucasians, in spite of similar duration of diabetes.¹²¹ Nephropathy is reported to occur with shorter duration of diabetes (<10 years) in South Asians than in Whites.^{122, 123, 124, 125, 126, 127} Furthermore, as compared with White Europeans, South Asian T2DM patients in the United Kingdom were younger at diagnosis and had significantly higher prevalence of diabetic retinopathy, maculopathy, site-threatening retinopathy (45 vs 37%; P=0.059, respectively),^{128, 129} and specific vascular lesions (for example, arterio-venous nicking; 8.5 vs 7.5%, respectively; P<0.005).¹³⁰

Figure 2

Kaplan–Meier survival plots of Indo-Asian, African-Caribbean and White patients with type 2 diabetes showing doubling of serum creatinine (as end point) on the Y axis and follow-up (months) on X axis. All Indo-Asian patients experienced doubling of their creatinine compared with only 45 and 50% of African-Caribbean and Whites, respectively. (log-rank test χ²=7.36; P=0.025). Source: Earle et al.²⁷⁷

South Asians with T2DM showed an increased predisposition to CHD as compared with Europeans. Samanta et al.¹¹² showed diabetes to be associated with a higher relative risk of acute myocardial infarction (MI) in UK South Asians than in European Whites (adjusted relative risks being 3.3 and 1.3, respectively).¹³¹ T2DM in South Asians causes higher morbidity and mortality than other ethnic groups.^{3, 132, 133, 134, 135, 136, 137} There are, however, contradictory reports regarding other macrovascular complications, but it appears that the risk of foot ulcers and diabetes-related amputation may be lesser in South Asians than in European Whites/Caucasians.^{138, 139, 140, 141, 142} A recent report also indicates that Asian Indians with diabetes in the United Kingdom have substantially less large and small fiber neuropathy than European Whites, despite comparable traditional risk factors, possibly due to better skin microvascularization, thus explaining reduced foot ulcer risk.¹⁴³

Dyslipidemia

Similar to T2DM, dyslipidemia occurs at lower levels of BMI and body fat in South Asians than in Whites.^{11, 24, 66, 144} Higher triglycerides and lower HDL levels (Table 6) associated with insulin resistance have been consistently shown in Asian Indians/South Asians as compared with Europeans/Whites.^{14, 71, 99, 109, 114, 137, 145, 146, 147} Interestingly, Asian Indians not only have low levels of protective HDL-C, but also have a preponderance of small-dense dysfunctional HDL-C particles, which are less protective and proinflammatory even at normal HDL levels.^{148, 149} Furthermore, Kulkarni et al.¹⁵⁰ demonstrated higher prevalence of small, dense low-density lipoprotein in Asian Indians compared with Whites in the United States of America (44 vs 21%, respectively; P<0.05) Others have demonstrated similar findings in adult Asian Indians in the United Kingdom,¹⁵¹ and in adolescent (15–16 years) South Asians in the United States of America.¹⁵² Overall, a possible explanation of excess CVD risk in South Asians may be due to high prevalence of dysfunctional HDL and small dense low-density lipoprotein as compared with Whites.¹⁵³

Table 6 Differences in the HDL levels between South Asians/Asian Indians vs Whites/Europeans

Lipoprotein(a) (Lp(a)), a low-density lipoprotein-like particle (with apolipoprotein B-100 linked to apolipoprotein (a)), is an inherited atherogenic lipoprotein and is thrombogenic.¹⁵⁴ Importantly, Lp(a) concentration of South Asians are higher than Whites.^{155, 156, 157, 158, 159, 160} Some studies show that elevated plasma levels of Lp(a) confer almost three-fold risk of CVD in Asian Indians.¹⁶¹ Physical inactivity, atherogenic diet and abdominal obesity may additionally magnify the effect of high Lp(a) levels in Asian Indians.¹⁶² Significance of these findings in South Asians with respect to the development of CHD is not clear, and need further research.

Hypertension

Unlike discussion on other variables shown above, comparative data are not conclusive regarding the prevalence and severity of hypertension between South Asians and Whites. Several investigators from the United Kingdom showed that South Asians tend to have higher prevalence of hypertension than UK-based European Whites.^{65, 163, 164} However, a meta-analysis comparing hypertension in South Asians and Whites showed mixed results. Out of 12 studies, 7 studies showed lower mean systolic blood pressures, whereas other 7 studies showed higher diastolic pressures in South Asian men compared with Whites.¹⁶⁵ Karlsen et al.¹⁶⁶ demonstrated age-standardized ratio of mean diastolic blood pressure to be higher in Asian Indian men, but lower in Pakistani and Bangladeshi men than in the general UK population.¹⁶⁶ However, this heterogeneity within different South Asian ethnic groups has been reflected in other variables also (such as smoking, physical activity and CHD) and has been ascribed to socioeconomic factors and patient's knowledge and outlook about diseases.¹⁶⁷

Higher cardiovascular risk and clustering often at lower levels of adiposity in South Asians than in Whites

Raji et al.⁷¹ demonstrated that even at normal BMI, Asian Indians showed clustering of cardiovascular risk factors (insulin resistance, dyslipidemia, procoagulant state).⁷¹ Overall, with similar level of weight and obesity even at ‘normal’ BMI and WC levels, South Asians have greater risk of metabolic perturbations (‘metabolically obese’ individuals), T2DM and CVD than Europeans/Whites.^{9, 14, 68, 168, 169, 170}

Earlier onset and higher rates of acute MI and CHD, as well as mortality due to CHD and stroke (Figures 3a, b)¹⁷¹ is higher in South Asians, and is often associated with abdominal obesity, insulin resistance and diabetes than Europeans/Whites.^{13, 107, 132, 133, 157, 172, 173, 174, 175, 176, 177} Furthermore, South Asians are more likely than Whites to have evidence of CHD in the absence of symptoms or clinical findings.^{157, 178} Finally, South Asians have more severe angiographic disease (significant left main, distal, along with more diffuse triple vessel disease), and more anterior and larger MI as compared with Europeans/Whites.^{135, 136, 173, 179}

Figure 3

(a) Mortality from coronary heart disease (CHD) per 100 000 by country of birth, adults aged 30–69 years, England and Wales (1999–2003). Source: CHD Statistics 2008: Mortality.¹⁷¹ (b) Mortality from stroke per 100 000 by country of birth, adults aged 30–69 years, England and Wales (1999–2003). Source: CHD Statistics 2008: Mortality.¹⁷¹

Subclinical inflammation, endothelial dysfunction and procoagulant factors

Subclinical inflammation, denoted by increased levels of high-sensitivity C-reactive protein (hs-CRP), has been reported to correlate with obesity, the metabolic syndrome, T2DM and increased cardiovascular risk.^{180, 181} Asian Indians have higher hs-CRP concentrations as compared with European/UK Whites, which is strongly associated with obesity and insulin resistance.^{182, 183, 184} These observations have also been reported in children.¹⁸⁵ Chambers et al.¹⁸² showed that hs-CRP concentrations significantly predicted the risk of MI, and out of 40% excess risk of CHD reported in Asian Indians compared with European Whites, about two-fifth was thought to be attributable to elevated hs-CRP levels.¹⁸² Some investigators have suggested that increased abdominal adiposity may be responsible for subclinical inflammation in South Asians.¹⁸⁶

Endothelial dysfunction is believed to be associated with insulin resistance, and may be one of the first event in its pathogenesis.^{187, 188} Endothelial-dependent dilatation and circulating numbers of endothelial progenitor cells (EPCs) and EPC colony-forming units (indicating dysfunctional endothelium) were significantly lower in South Asian than in White men (Table 7).¹⁸⁹ The strongest predictor of flow-mediated dilatation (used for endothelial function assessment) was EPC number. Further, ethnicity was shown to be the strongest predictor of EPC count. In addition to quantitative difference, evidence of lower EPC functional capacity (proliferative capability and migratory response) was also seen in South Asians as compared with Whites.¹⁸⁹ Further, Chambers et al.¹⁹⁰ showed that Asian Indians with vascular endothelial dysfunction had more visceral and abdominal obesity, insulin resistance and dyslipidemia as compared with European Whites (3.2 vs 5.9%, P=0.03). More importantly, relation between Asian Indian ethnicity and flow-mediated dilatation was independent of insulin resistance and associated metabolic markers, raising the possibility that endothelial function among Asian Indians could be genetically mediated.¹⁹⁰ These authors further showed that reduced levels of endothelial nitric oxide in Asian Indians may contribute to vascular injury by facilitating platelet–vascular wall interactions, adhesion of circulating monocytes to the endothelial surface and vascular smooth muscle proliferation.¹⁹⁰

Table 7 Differences in conduit, resistance vessel function and endothelial pathological features in South Asians vs Whites

Hemostatic and fibrinolytic factors, including plasminogen-activator inhibitor-1 (PAI-1), tissue plasminogen activator and fibrinogen constitute another dimension of insulin resistance and the metabolic syndrome.^{191, 192} Several studies show that South Asians have higher procoagulant tendency; increased C-3, PAI-1, fibrinogen and decreased tissue plasminogen activator levels as compared with European and North American Whites.^{71, 109, 157, 160, 193, 194} Importantly, Kain et al.¹⁹⁵ showed that significantly higher mean fibrinogen levels were seen in South Asians than UK Whites, even after adjustments for BMI, hypertension, and serum triglycerides. Further, Raji et al.⁷¹ showed that PAI-1 levels inversely correlated with glucose disposal rate in Asian Indians but not in Whites.⁷¹ Interestingly, South Asians without the metabolic syndrome and UK Whites with the metabolic syndrome had similar plasma levels of C3 and tissue plasminogen activator, as well as comparable degree of insulin resistance.¹⁰⁹ Finally, the same authors showed that South Asians with the metabolic syndrome had significantly higher plasma levels of C3, PAI-1 and tissue plasminogen activator, in addition to significantly higher degree of insulin resistance, than UK Whites with the metabolic syndrome.¹⁰⁹

Non-esterified fatty acids, leptin and adiponectin

Excess lipolysis, with release of large amounts of non-esterified fatty acids (NEFAs), is believed to increase triglyceride levels in skeletal muscle,¹⁹⁶ enhance gluconeogenesis,¹⁹⁷ and stimulate triglyceride production,¹⁹⁸ thus causing increased triglycerides/lipid overload in non-adipose tissues. This metabolic process, known as ‘lipotoxicity’ causes pancreatic β-cell dysfunction and apoptosis.^{199, 200, 201} Both fasting and baseline plasma concentrations of NEFAs were reported to be significantly higher in Asian Indians as compared with Whites, along with impaired insulin-mediated plasma NEFA suppression during oral glucose tolerance test.²⁰² Higher fasting plasma NEFAs, despite relative hyperinsulinemia, indicated higher release of NEFAs from adipose tissue of Asian Indians, although reasons for defective insulin action in adipose tissue were not clear.²⁰²

Asian Indians have significantly higher fasting serum leptin (5.15 vs 2.3 ng ml⁻¹, respectively; P<0.001), along with lower insulin sensitivity index and lower insulin clearance than Whites.²⁰³ Higher leptin concentrations in South Asians as compared with Whites, even after statistical adjustments for body fat, suggest excessive leptin production from adipocytes, independent of overall obesity or abdominal and/or truncal obesity.²⁰² Furthermore, lower serum adiponectin levels, correlating with low HDL-C and central obesity, was seen in Indo-Asians compared with Whites.²⁰⁴ Importantly, thicker truncal skinfolds (indicating high truncal SC fat) in Asian Indians were strongly correlated with low plasma adiponectin levels.²⁰² All these data indicate an intrinsic and yet unknown abnormality in adipose tissue metabolism of Asian Indians, the precise mechanisms of which remain unclear.

Environmental, perinatal and genetic factors affecting OR-NCDs

The following section briefly summarizes important differences and observations pertaining to OR-NCDs on environmental, perinatal and genetic role in South Asians. For detailed discussion on these topics, readers should refer to other reviews.^{10, 100, 205, 206, 207, 208, 209, 210, 211, 212, 213}

Higher intakes of carbohydrate, saturated fatty acids, trans fatty acids and ω-6 polyunsaturated fatty acids, along with lower intakes of monounsaturated fatty acids and fiber have been reported in South Asians.¹⁰ Burden et al.²¹⁴ showed higher carbohydrate content (% energy) in South Asian than European diets to be associated with higher fasting glucose, triacylglycerols, higher HDL and post-glucose load insulin levels.²¹⁵ Another contributory factor to insulin resistance in Asian Indians may be imbalanced dietary ω-6 and ω-3 polyunsaturated fatty acids. Importantly, South Asians consume higher proportion of total fatty acids as ω-6 polyunsaturated fatty acids and a lower proportion of long chain ω-3 polyunsaturated fatty acids as compared with Europeans.^{216, 217} Furthermore, South Asians, including children, consume less fresh fruits, vegetables and fiber than Whites.^{97, 218, 219, 220} Many of these imbalanced dietary factors have important implications for OR-NCDs in South Asians. Recent consensus dietary guidelines under preparation have aimed to address some of these issues (Misra A and Indian Dietary Guidelines Consensus Group, unpublished observations, 2010).

Lower level of physical activity in Asian Indian, Pakistani and Bangladeshi adults and children (South Asian populations) than Europeans/Whites are important for the development of OR-NCDs in the former (Table 8).^{206, 221} Lesser physical activity in Indians, Pakistanis or Bangladeshis than Europeans inversely correlated with BMI, WC, systolic blood pressure, plasma glucose and insulin levels.²⁰⁶ In a recent consensus statement, physical activity guidelines for Asian Indians emphasize 60 min of physical activity daily, and is inclusive of aerobic and resistance exercises, all days of the week.⁶⁴

Table 8 Differences in physical activity patterns in South Asians/Asian Indians vs Whites/Europeans

Adverse intrauterine milieu is believed to have profound and permanent effect on structure and functional development of fetus, and consequent development of chronic diseases in later life (‘intrauterine fetal programming’).²¹² Low birth weight of Indian babies (mean <2.7 kg; which is lower by 1 kg than White babies)²²² was associated with more adiposity and poorer muscle mass compared with White babies.²¹² In addition, higher truncal SC fat, insulin and leptin levels in Asian Indian neonates as compared with White neonates have been reported.²²³ Higher susceptibility of Asian Indians to insulin resistance, T2DM, hypertension and CHD has been attributed to ‘intrauterine programming’ in adverse nutrient milieu, and consequent to low birth weight. Interestingly, most adverse risk profile were seen in children who showed ‘catch-up obesity’ (were small at birth but had a higher weight and fat mass at 8 years).²²⁴ Overall, implication of these findings to pathogenesis of OR-NCDs is not clear, given the continued heightened risk to develop these diseases in heterogeneous perinatal conditions and continued risk in Asian Indian children born to affluent families in urban environment.

Several studies show genetic predisposition for development of obesity, diabetes, dyslipidemia and CVD in South Asians. Discussion on this subject is beyond the scope of this article, and could be accessed in original articles and reviews.^{209, 210, 225, 226, 227, 228, 229, 230, 231, 232, 233} Overall, predominant contribution of gene(s) or gene(s)–environmental interaction(s) in the pathogenesis of OR-NCDs in South Asians remains to be researched.

Problems and barriers to diagnosis and management

South Asian patients have more atypical presentations (for example, discomfort over interscapular region, more diffuse and variable anatomical location of discomfort) than classical chest pain during acute coronary syndromes than Europeans/North American Whites.^{234, 235, 236, 237, 238, 239, 240, 241, 242} Atypical presentation could be a contributory factor to delayed diagnosis, less aggressive therapy (underutilized thrombolytic therapy) and higher mortality in South Asians.^{237, 243} Furthermore, British South Asian patients were shown to be less likely to be referred for stress testing, angiography and coronary artery bypass grafting, and were less likely to receive intensive care comparable with European Whites.²⁴⁴ Asian Indians were also less likely to complete the exercise stress test, probably due to poorer exercise tolerance than Europeans.²⁴⁵ Lawrence and Littler²⁴⁶ showed longer mean time from arrival in casualty to admission in coronary care unit with South Asians than European Whites, mostly attributable to communication difficulties.²⁴⁶ Bradley et al.²⁴⁷ also showed that door-to-drug (fibrinolytic) time (in terms of arrival to hospital and reperfusion therapy) was significantly longer for Asian Indian patients with MI than Whites in the United States of America.²⁴⁷

Difficulties in achieving treatment compliance in South Asian patients with diabetes in the United Kingdom are believed to be due to inappropriate or negligible health education, language barriers and lower awareness about disease risk and its complications than Europeans/UK Whites.^{248, 249, 250, 251, 252, 253, 254} Many South Asian patients could not name any complication of diabetes, were not aware of purpose of attendances at the clinic (screening for early complications) and did not know what a chiropodist did.²⁵⁵

Sociocultural and religious factors influence health beliefs, diet and management in South Asians.^{256, 257, 258} Cultural factors contribute to higher default from clinic appointments by South Asians than Whites (for example, South Asians having festival/holiday on that day). Furthermore, compliance with therapy (insulin or oral treatment) is less in South Asians during holidays, and control is unsatisfactory due to religious fasting.^{248, 249, 259} In contrast, an advantageous sociocultural factor seen in South Asians is the extended family structure, often helping patients to cope better with insulin therapy and morbidities.²⁴⁹ South Asian patients with diabetes had a more negative attitude, and believed they were made to wait longer in clinics than the UK Whites.²⁵¹ Interestingly, cause of diabetes by migrant Bangladeshis in the United Kingdom was believed to lie in taking particular foods, balance of food entering the body and excess emission of body fluids.²⁶⁰ Furthermore, physical exercise was viewed as having a potential to exacerbate illness and physical weakness, instead ritual Muslim prayers (namaz) were cited as worthy and healthy form of exercise.²⁶⁰ Finally, being asymptomatic was interpreted as having well-controlled diabetes, and need for regular surveillance, when asymptomatic was rarely acknowledged.²⁶⁰ A reluctance to start insulin therapy is a frequently encountered problem in Asian Indian Hindus, as it is perceived as a failure of self-care and progression from ‘mild’ to more ‘serious’ diabetes.²⁶¹ These distorted perceptions may have implications in lesser health-seeking behavior and consequent poor prevention and control of the disease in South Asians than in Whites.

Differences in response to therapy and disease outcomes

Glycemic control is often significantly worse in South Asians than in Whites, in spite of South Asians being younger and having lower BMI.^{249, 251, 262} South Asians showed smaller improvements in cholesterol and blood pressure compared with Whites.²⁶³ Further, South Asians were less likely to receive statins in spite of similar calculated 10-year CHD risk (30% by Framingham algorithm), possibly due to lower average total serum cholesterol concentrations.²⁶⁴ Even when same proportion of patients received lipid-lowering therapy, proportion of patients with uncontrolled total cholesterol and poorly controlled blood pressure were higher in Bangladeshi South Asians than in White Europeans.¹¹⁰ On the contrary, other studies have shown no differences in therapy outcomes in South Asians.^{265, 266}

South Asians, with and without diabetes, not only are more likely to suffer a cardiac event, but are also less likely to survive it than Whites, in spite of intervention. As compared with UK Whites, South Asians with MI had lesser survival possibilities, both at 30 days (82.9 vs 93.2%, respectively, P=0.12) as well as 6 months (89.1 vs 95.0%, respectively, P=0.12), in spite of greater proportion of South Asians receiving thrombolytic drugs (81.2 vs 73.8%, respectively, P=0.12). Importantly, the presence of diabetes was considered to be a likely explanation for therapy differences.¹³⁴ South Asian ethnicity seems to be an independent predictor of operative and overall mortality after coronary artery bypass grafting.²⁶⁷ Postoperative morbidity (for example, MI, sepsis, sternal wound infection, postoperative hospital stay after coronary artery bypass grafting and use of inotrope drugs) was also higher in South Asians than Whites in Canada.²⁶⁷ Furthermore, revascularization rates at follow-up were also higher for South Asian patients (6.5 vs 3.1%; P=0.001).²⁶⁸ On the other hand, some reports do not show any difference in incidence of MI or long-term adverse survival in South Asians with diabetes than in UK Whites.^{269, 270}

Response to drugs has been suspected to be different in South Asians than Caucasians, but has not been adequately investigated. Increase in rosuvastatin plasma exposure and area under the plasma concentration–time curve from time 0 to the time of last quantifiable concentration was higher in Asian Indians (1.63-fold) as compared with Whites.²⁷¹ On the basis of these data, it has been suggested that care should be exercised while prescribing rosuvastatin in Asian Indians, and this has led to revised labeling of rosuvastatin.^{271, 272, 273} Of note, therapy should be started at lower dose (5 mg once daily) and 40 mg per day dose (if needed) should be given with caution in Asian Indians.²⁷⁴ More insulin-resistant Asian Indians as compared with Whites respond more favorably to pioglitazone with marked improvements in insulin sensitivity, cardiovascular/inflammatory risk markers and vascular responses to insulin. A significant 32% increase in glucose disposal in Asian Indians was seen with pioglitazone treatment, whereas the effect was of smaller magnitude in Whites. Importantly, >50% decrease was seen in hs-CRP and PAI-1 levels in Asian Indians with pioglitazone therapy as compared with Whites.²⁷⁵ On the contrary, one report suggests that metformin may be more effective in Whites.²⁷⁶ Importantly, more studies are needed to get clarity on apparent differential response of insulin sensitizers in South Asians than in Whites and the factors responsible for it.

Drug therapy for obesity is presently based on clinical studies based on cut-offs of White population. The current consensus statement for Asian Indians suggest lower cut-offs of BMI for initiation of pharmacotherapy of obesity and for bariatric surgery.⁶⁴

Conclusions

Increasing disease burden due to obesity and OR-NCDs is being seen in South Asian populations due to rapid industrialization, ‘westernization’ and consequent lifestyle changes. Apart from these environmental factors, genetics, certain innate body composition and differences in metabolic profile are also responsible for higher predisposition and increased prevalence of OR-NCDs in South Asians than in Whites. Higher abdominal and/or truncal obesity in South Asians vs Whites is hypothesized to be responsible for increased morbidity/mortality associated with OR-NCDs. Higher prevalence of insulin resistance, the metabolic syndrome and T2DM and its other components confer South Asians with an increased cardiovascular risk. Poorer nutritional habits and lesser physical activity along with adverse perinatal mileau (low birth weight and ‘catch-up’ obesity) seen in South Asians vs Whites are also believed to be contributory factors. Interestingly, studies in migrant South Asians have shown differences in disease outcomes and therapeutic response than Whites. However, more research is needed in this area so as to optimize the prevention and management strategies, leading to more positive health outcomes. Recent guidelines advocating lower cut-offs of obesity and abdominal obesity for South Asians as compared with international guidelines are expected to help physicians in better and more effective preventive management. However, greater awareness among physicians about the differences in OR-NCDs between South Asians and Whites is needed and would enable them to achieve better treatment outcomes.