**3. Results**

The prevalence of underweight, stunting and low BMI (below-2 *Z*-score) [29] in these children was 48.2%, 42.0% and 16.1% respectively. The prevalence of hyperhomocysteinemia was 18.4%. Among the 1000 children, 827 (83%) had low HDL cholesterol (<0.9 mmol/L), 108 (11%) had high triglyceride ( mmol/L), 190 (19%) had high blood pressure ( 0th percentile of reference), 153 (15%) had high waist circumference ( WK SHUFHQWLOH \$ WRWDO RI  FKLOGUHQ PHW WKH criteria for MetS.

7DEOHLQGLFDWHVWKDWFKLOGUHQZLWKHOHYDWHG+F\!ȝPRO/KDGKLJKHU%0,*p* = 0.002), waist circumference (*p* = 0.043), systolic (*p* = 0.071) and diastolic (*p* = 0.052) blood pressure, and triglycerides (*p* = 0.016) and total lipid concentrations (*p* = 0.048). HDL concentrations were lower in subjects with higher vs. lower Hcy levels (*p* = 0.031). In logistic regression analyses adjusted for age and sex (Table 3), an elevated Hcy level was associated with an increased risk of low HDL cholesterol [OR = 1.77 (95% CI = 1.08–2.88); *p* = 0.022], high BP [OR = 1.65 (95% CI = 1.10–2.46); *p* = 0.015] and MetS [OR = 1.75 (95% CI = 1.06–2.90); *p* = 0.029]. No association was observed between elevated Hcy and high TG, waist circumference or PG.

We detected an increased risk of high BMI (WKSHUFHQWLOH[OR = 1.98 (95% CI = 1.33–2.96); *p* = 0.001] with hyperhomocysteinemia. Moreover, we observed that the relative odds of having combined low HDL cholesterol and high BP was 1.76 (95% CI = 1.14–2.70; *p* = 0.010). Similarly the odds ratio for combined low HDL cholesterol and high BMI was 2.31 (95% CI = 1.51–3.53; *p* < 0.001). We didn't see additional risk for combined low HDL, high BP and high BMI with hyperhomocysteinemia (*p* = 0.351) (Table 3).


**Table 2.** Anthropometric measures, blood pressure, and plasma biochemical biomarkers of study children by plasma homocysteine levels 1 .

1 Values are means±SD; <sup>2</sup> *p* compares mean values in Homocysteine ( ȝPRO/ *vs*. Homocysteine ! ȝPRO/ JURXSV XVLQJ LQGHSHQGHQW VDPSOH *t*-test; <sup>3</sup> Data were missing for BMI (*n* = 1), waist circumference (*n* = 3), systolic blood pressure (*n* = 8), diastolic blood pressure (*n* = 8), LDL (*n* = 367) and FPI (*n* = 324); BMI, body mass index; TG, triglyceride; LDL, low density lipoprotein; HDL, high density lipoprotein; PG, plasma glucose; FPI, fasting plasma insulin; HbA1c, glycosylate hemoglobin.

**Table 3.** Risk of MetS and its components in 6–8 years old children related to hyperhomocysteinemia (*n* = 1000) <sup>1</sup> .


1 Values are OR (95% CI) based on separate logistic regression analyses for each set of outcomes, adjusted for age and gender; <sup>2</sup> MetS defined as presence of any three of the following constituents: elevated waist circumference, high PG, high systolic or diastolic BP, high TG and low HDL.

### **4. Discussion**

We observed an increased risk of MetS and its components, specifically dyslipidemia, BMI, and hypertension, with elevated Hcy in young school-aged children in rural Nepal. The distributions of these components of MetS were similar to those reported previously for the larger cohort of 3524 children from whom this sample was derived [9]. That study identified intrauterine exposure to folate as a factor that ameliorated the risk of MetS in childhood; the current study expands on those findings by identifying a more proximal potential risk factor for MetS.

Previous studies have revealed that individuals who meet the criteria of MetS are at increased risk for developing CHD and diabetes mellitus [30], both of which are increasing in South Asia. It has been reported that 20%–25% of South Asians have MetS, with the frequency expected to rise in the future [31]. Nepal is not an exception, with CHD considered the leading cause of death in 2010 followed by stroke, hypertension and diabetes mellitus among other major causes of death [32]. However, data for these reports have been generated from urban areas with no study, so far, carried out to establish the burden of CVD in rural Nepal. Shaik, *et al*. collected hospital admission data in one of the tertiary hospitals which served the terai of Nepal and observed that around two patients out of ten were admitted due to stroke [33]. The extent of CVD in the region surrounding Nepal is not different, as it has been reported that 32% of all deaths in rural India are due to CVD [34].

Not limited to adults, South Asian children are also reported to have increased susceptibility to MetS and CVD, as South Asian children living in the UK showed higher mean heart rate, elevated mean triglyceride and fibrinogen levels, and higher mean fasting and post-glucose load insulin concentrations compared to white children [35]. Moreover, evidence also suggests that Indian children who are born small for gestational age have higher blood pressure, cholesterol levels (total cholesterol and LDL-C), as well as increased adiposity at the age of eight years [36]. Our results are in line with the above-mentioned studies, as we also observed high mean triglycerides and low mean HDL in these children, who were nonetheless thin (prevalence of low BMI 16%), underweight and stunted (>40% prevalence for both) [18]. Our data also suggest a role for elevated homocysteine with hypertension, dyslipidemia, and adiposity.

Although Hcy concentrations have been investigated in children in Western populations [20,37], no previous study has examined this risk factor in young South Asian children. The 95th percentile of Hcy ȝPRO/ PHDQ FRQFHQWUDWLRQ +F\ RI ȝPRO/ DQG SUHYDOHQFH RI K\SHUKRPRF\VWHLQHPLD (18.4%) observed in our study area reflect elevated levels of Hcy in rural Nepal compared to Western populations [20,37]. A high burden of hyperhomocysteinemia in children could be a risk factor for CVD in later life, as elevated Hcy (95th percentile) in Dutch children was associated with a subsequent 4-fold increased risk of ischemic cerebrovascular disease [38]. High levels of Hcy in children of rural Nepal could be due to genetic polymorphisms, environmental exposures (e.g., high blood lead), physical activity patterns, and diet, particularly one that may lead to low vitamin B12 concentrations [39].

Atherosclerosis is known to begin in childhood, as autopsy reports in children and young adults with unexpected death have revealed positive associations between atherosclerotic lesions and risk factors such as LDL-C, triglycerides, systolic and diastolic blood pressure, body mass index and cigarette smoking, making it imperative to maintain healthy lipid profiles, blood pressure [40] and plasma Hcy to minimize the burden of diseases like CVD and MetS. High levels of Hcy and high cholesterol are both associated with CVD risk. However, very few studies have reported the combined effect of hyperhomocysteinemia and dyslipidemia on the risk of CVD [41]. Since we observed a high prevalence of low HDL (83%) and considerable prevalence of hyperhomocysteinemia (18.5%) in these children, a potential interaction between Hcy and HDL cholesterol seems clinically relevant. Although exact mechanisms relating Hcy and cholesterol are not known, animal studies have revealed that hypomethylation due to hyperhomocysteinemia could be accountable for lipid accumulation in tissues [42]. Moreover, Hcy could also modulate activity of some inhibiting enzymes which play a role in HDL-particle assembly [42].

The association we observed between Hcy and blood pressure also merits discussion. We observed that an increase in Hcy was associated with rise in systolic and diastolic blood pressure, a finding consistent with National Health and Nutrition Examination Survey (1994–1998) and other studies of adult populations in developed countries [43,44]. Possible mechanisms to explain a link of Hcy with increased blood pressure are Hcy-induced arteriolar constriction [45] vascular endothelial damage [46] and decreased vasodilator responsiveness [47].

To the best of our knowledge no previous study has addressed the relationship of Hcy with MetS or components of MetS in children. Those studies which have addressed the relationship of Hcy with MetS in adults present contradictory results. Relationships we observed were similar to data published by Kang *et al*. showing a positive association between Hcy and triglycerides, BMI and systolic and diastolic blood pressure and a negative association with HDL in Korean adults [48]. Likewise, Hcy was positively associated with waist circumference, BMI, blood pressure, LDL-C, and triglycerides, but inversely associated with HDL in a Chinese sample of 1680 adults [49]. On the other hand, there are published reports where authors did not find associations [50] or observed associations with few components of MetS [51]. A lack of association of hyperhomocysteinemia with insulin resistance is consistent with other reports [52,53].

This study had some limitations. Due to the cross-sectional design we cannot establish a temporal or causal relationship between Hcy and MetS. The elevated Hcy and MetS could be results of common pathways such as poor diet, environment, inadequate physical activity [54,55], and genetic polymorphisms. Since no stringent range has been set to define hyperhomocysteinemia in children, our definition of hyperhomocysteinemia may underestimate the prevalence of hyperhomocysteinemia as WKHFXWRII!ȝPRO/LVJHQHUDOO\XVHGIRUDGXOWV0RUHRYHUZHsomewhat overestimated the actual mean of total cholesterol, HDL, TG and plasma glucose by assigning values for these analytes equivalent to the lowest detectable concentration of these analytes, although prevalence of abnormal findings would not have been affected We believe that this study provides unique information about the high prevalence of hyperhomocysteinemia and its role as a risk factor for MetS in a population of otherwise undernourished children residing in rural Nepal. Our findings reveal important issues to be considered further on the relationship of Hcy and MetS in a population which is known to have a high burden of CVD in adulthood.

#### **5. Conclusions**

We conclude that hyperhomocysteinemia exerts risk towards development of MetS and that high prevalence of hyperhomocysteinemia and MetS in this low income population suggest that Nepalese children are at a greater risk of developing CVD and diabetes in future. It is therefore necessary to understand the causes of hyperhomocysteinemia, such as dietary habits in this population, in order to attenuate the future development of diabetes and cardiovascular disease symptoms, both of which constitute a growing concern among populations of South Asia.

### **Acknowledgments**

This study was supported by Grants #GH614 and OPPGH5241 from the Bill & Melinda Gates Foundation, Seattle, WA, USA. The original Nepal Nutrition Intervention Project- Sarlahi (NNIPS)-2 trial was supported through the Vitamin A for Health Cooperative Agreement (HRN-A-00-97-00015-00) between Johns Hopkins University and the Office of Health, Infectious Diseases and Nutrition, United States Agency for International Development (USAID), Washington DC, with additional support from the Sight and Life Research Institute, Baltimore, MD, USA and Basel, Switzerland.
