Overweight and Obesity Among Hospital In-Patients

Gumpeny, Reddy, Sridhar and Krishnamurty: Overweight and Obesity Among Hospital in-�Patients Introduction It is well recognized that the prevalence of overweight and obesity in India is increasing ( 1 , 2 ). The adverse effects are also well known, including the risk of metabolic syndrome and type 2 diabetes mellitus. However, most reports originate from population samples and from out-patient clinics ( 2 ). There is sparse information on subjects who are admitted as in-patients. Many of the reports among in-patients studied the outcome of obesity on surgical procedures ( 3 ) and recently, following bariatric surgery. There has not been documentation of patterns of overweight and obesity among subjects who were admitted as in patients of medical wards, particularly from rural areas of South India. Considering that preventing and treating obesity aims to prevent illness, we performed this exploratory study to determine the pattern of overweight and obese subjects who were admitted to PES Hospital, a referral centre situated in a rural area of Andhra Pradesh adjoining Karnataka and Tamil Nadu borders. Anthropometric data, prevailing medical illnesses, and gender differences amongoverweight and obese patients admitted to PES Hospital were assessed and variables were correlated. Material and Methods Source of Data: This was an inpatient based cross sectional study, considering 100 patients older than 18 years with BMI>25kg/m^2 admitted to the medical wards of PES Hospital Kuppam between August 2012 and August 2013. Inclusion criteria: All patients older than 18 years with BMI>25kg/m^2admitted to the medical wards of PES Hospital, Kuppam. Exclusion criteria: Pregnant women, patients with conditions causing fluid retention like heart failure, renal failure, and decompensated liver disease ; Patients with neurological conditions or otherwise, which interfere with measurement of anthropometric features, and patients with thyromegaly were excluded. The selected patients’ anthropometric measurements were taken as per established guidelines, and patients underwent workup for medical diseases associated with obesity like Hypertension, Type 2 Diabetes, Metabolic Syndrome, Dyslipidemia, Coronary Artery Disease, Obstructive Sleep Apnoea, Gastroesophageal Refl ux Disease, Osteoarthritis, Fatty liver, and Pancreatitis. Situations like steroid abuse, hypothyroidism and other drugs which may lead to obesity were documented. Measurement of waist, hip circumference: Waist circumference was measured at the midpoint of the lower margin of last palpable rib and the top of the iliac cress, with the subject standing up and at the end of normal expiration. Hip circumference was measured at the level of the posterior protuberance of the gluteals (widest portion of buttocks). Body mass index (BMI) was taken as body weight (kilogram) divided by square of height (expressed in meters). Classes of obesity were taken according to WHO guidelines ( 4 ). Results Most subjects were aged between 41 and 70 years ( Table 1 ); overweight (BMI 25-29.9) was the most common (60%) ( Table 2 ). Women tended to have a greater BMI, and a lower waist-hip ratio ( Table 3 ). Acanthosis and acrochordon were also more common in women ( Table 4 ). Compared to men, women tended to have greater systolic and diastolic blood pressure ( Table 5 ). In general the lipid parameters were comparable between the sexes ( Table 6 ). Interestingly, one woman consumed alcohol. Osteoarthritis was more common in women. Co-morbid conditions in decreasing order were Diabetes (54%), Hypertension (52%), Dyslipidemia (44%), Obstructive Sleep Apnea (30%), Non Alcoholic Fatty Liver Disease (30%), Osteoarthritis (30%), Cardiovascular Disease (26%), Gastroesophageal Reflux Disease (18%) and Pancreatitis (4%) ( Table 7 ). Table 1: Age distribution of patients studied Age in years Gender Total Male Female <30 0(0%) 2(3.3%) 2(2%) 30-40 6(15%) 4(6.7%) 10(10%) 41-50 6(15%) 15(25%) 21(21%) 51-60 15(37.5%) 20(33.3%) 35(35%) 61-70 9(22.5%) 12(20%) 21(21%) >70 4(10%) 7(11.7%) 11(11%) Total 40(100%) 60(100%) 100(100%) Mean ± SD 55.95±12.62 56.07±12.87 56.02±12.71 P = 0.964, not significant, student t test Table 2: Obesity class Obesity class Gender Total Male Female Obesity 1 8(20%) 23(38.3%) 31(31%) Obesity 2 1(2.5%) 6(10%) 7(7%) Obesity 3 0(0%) 2(3.3%) 2(2%) Overweight 31(77.5%) 29(48.3%) 60(60%) Total 40(100%) 60(100%) 100(100%) Table 3: Anthropometric variables according to gender Anthropometric variables Gender Total P value Male Female Age in years 55.95±12.62 56.07±12.87 56.02±12.71 0.964 Height (cm) 163.53±7.20 152.48±6.86 156.90±8.83 <0.001** Weight (kg) 75.70±9.32 70.83±10.34 72.78±10.18 0.018* BMI (kg/m 2 ) 28.16±2.62 30.25±3.83 29.41±3.54 0.003** Neck (cm) 37.38±3.27 35.37±2.46 36.17±2.97 0.001** Waist circumference (cm) 95.88±8.57 91.48±10.39 93.24±9.90 0.029* Hip circumference (cm) 104.25±11.02 103.08±10.12 103.55±10.45 0.587 Waist-Height ratio 0.58±0.05 0.60±0.07 0.59±0.06 0.276 Waist-Hip ratio 0.92±0.04 0.88±0.05 0.90±0.05 <0.001** Table 4: Acanthosis/Acrochordon according to gender Gender Total (n=100) Male (n=40) Female (n=60) Acanthosis 12(30%) 28(46.7%) 40(40%) Acrochordon 11(27.5%) 24(40%) 35(35%) Table 5: Comparison of BP according to gender Gender Total P value Male Female SBP (mm Hg) 130.65±21.92 143.60±28.15 138.42±26.50 0.016* DBP (mm Hg) 82.05±12.39 88.07±15.48 85.66±14.56 0.042* Table 6: Comparison of lipid parameters according to gender Gender Total P value Male Female Total cholesterol 193.50±77.70 183.90±67.57 187.74±71.57 0.514 TGL 155.93±54.88 162.20±80.03 159.69±70.80 0.666 HDL 38.73±12.81 36.62±13.54 37.46±13.23 0.438 LDL 87.85±35.62 82.57±32.88 84.68±33.92 0.448 Table 7: Comorbid conditions Comorbid conditions Gender Total (n=100) P value Male (n=40) Female (n=60) Hypertension 13(32.5%) 39(65%) 52(52%) 0.001** Type 2 Diabetes Mellitus 19(47.5%) 35(58.3%) 54(54%) 0.287 Dyslipidemia 21(52.5%) 23(38.3%) 44(44%) 0.162 Cardiovascular disease 11(27.5%) 15(25%) 26(26%) 0.780 Obstructive Sleep Apnea Syndrome 11(27.5%) 19(31.7%) 30(30%) 0.656 Gastro esophageal Reflux Disease 7(17.5%) 11(18.3%) 18(18%) 0.915 Osteoarthritis 7(17.5%) 22(36.7%) 29(29%) 0.119 Fatty liver 13(32.5%) 17(28.3%) 30(30%) 0.119 Pancreatitis 3(7.5%) 1(1.7%) 4(4%) Table 8: A comparison of different study populations Study Average BMI Smita P. Patil et al (2012) 11 , Maharasthra 23.48±2.71 in males and 26.41±4.86 in females Present Study (2013), Kuppam, Andhra Pradesh 30.25±3.83 in males and 28.16±2.62 in females Table 9: A comparison of different study populations Study Waist-hip ratio Danilo Ramos Haun et al, (2010) 12 , Brazil 0.91±0.07 in males, and 0.83±0.08 in females. Smita Patil et al (2012)(11), Maharasthra 0.96±0.05 in males and 0.94±0.12 in females Present study (2013), Kuppam, Andhra Pradesh 0.92±0.04 in males and 0.88±0.05 in females. Discussion Prevalence of BMI >25 in non hospitalized subjects from India range from 8.5% to 3.24% (1) . Among rural women, the prevalence ranged from 10% to 17.3% ( 1 ). A recent study reported from Punjab in 2013 showed that BMI and waist hip ratios were higher in urban women and lower in rural women ( 5 ). A study on young rural adolescents from south India showed that obesity and overweight were low in low socio economic group, due, mostly to lifestyle factors ( 6 ). In this hospital-based study, the pattern of overweight and obesity was assessed in subjects from southern India. This may be considered a reflection of morbidities seen in people who were not of normal weight. As expected, middle aged and elderly formed the most common age group. Overweight, rather than obesity was found in 60% of the whole sample, but it was more in men (77.5%) than in women (48.3%). In general, more women than men had obesity of all grades (viz 1-3); interestingly none of the men had grade 3 obesi ty. This implies that to be admitted in hospital, women had worse anthropometric parameters, and may have had more severe disorders to be brought to medical attention, in line with gender differences in access to medical care ( 7 ). An earlier study reported that American adults in the age group 40-59 were more likely to be obese ( 8 ). Besides, most admitted men in the current study were because of alcohol abuse and of chronic smoking. This could also be related to rural setting of the hospital, and most men did manual work in contrast to fewer women who were sedentary. Rather than just residence, a study from Karnataka, located in proximity to Kuppam, showed that migration of women from rural to urban areas was a risk factor for overweight ( 9 ). Even in a township from Maharastra, among 365 shopkeepers who were sedentary, obesity was reported in 22% and overweight in 43.3% ( 10 ). In the current study, the body mass index was also higher in women, who also had greater prevalen ce of acanthosis and acrochordon, both manifestations of insulin resistance. Interestingly, there was no significant difference in the lipid parameters between the sexes. A higher BMI among women was reported by Patil et al ( 11 ) ( Table 8 ). The waist hip ratio was higher in the current study than that reported from Brazil in 2010 ( 12 ) ( Table 9 ). Acanthosis and acrochordon are external signs of insulin resistance representing proliferative conditions due to growth factors causing keratinocytes and dermal fibroblast proliferation. Obesity is associated with hyperinsulinism and activation of IGF receptors; multiple skin tags are recognized as markers for diabetes ( 13 ). Indians have been reported to have abdominal adiposity as well as ectopic fat deposition and metabolic syndrome, both under and over nutrition pose a double edged problem ( 1 ). The co-morbidities found in the present study in the descending order of prevalence were Diabetes (54%), Hypertension (52%), Dyslipidem ia (44%), Obstructive Sleep Apnea (30%), Non Alcoholic Fatty Liver Disease (30%), Osteoarthritis(30%),Cardiovascular Disease (26%), Gastroesophageal Reflux Disease (18%) and Pancreatitis (4%) ( Table 7 ). Except hypertension, which was more common in women and chronic obstructive pulmonary disease which was more in men, the other disorders were equally distributed (viz diabetes mellitus, dyslipidemia, ischemic heart disease, obstructive sleep apnea and gastroesophageal reflux). The greater female preponderance of hypertension could be due to non-compliance to treatment in women, and reasons for admission in men being organ failure, critical illness with edema, rendering them ineligible for the study. The lack of dyslipidemia is of interest. Despite admission for a variety of conditions, the lipids were not abnormal, caused in part by lack of access to animal fat belonging as they do, to lower socioeconomic status. What is of importance is that despite lower BMI and not having dys lipidemia, co-morbid conditions can still occur even in rural areas. Other behavioural risk factors such as smoking and the consumption of alcohol must be looked into to prevent illness. Osteoarthritis was more than twice as common in women. Increasing age, and greater weight could account for the discrepancy. Obesity is a modifiable risk factor for osteoarthritis ( 10 ). Osteoarthritis of both knee and hip were increased by obesity ( 14 ). There are few studies on association of inpatients subjects globally. A report from San Diego California showed that obese patients who underwent total joint arthroplasty were at greater risk of infection ( 3 ). The proportion of obese women who underwent surgery was higher compared to men, in line with a greater number of admissions in the current study for joint symptoms among women. It is therefore important to try to prevent overweight and obesity so that the adverse consequences can be minimized. Not only should focus be placed on diet and exercise, but other potentially preventable aspects such as smoking and consumption of alcohol should be attended to ( 15 ). Strengths and limitations: This is an exploratory study to assess the morbidity among subjects who were overweight and obese admitted to a rural medical college hospital in south India. To our knowledge this is the first such evaluation from India. Being preliminary, the sample size and geographical location are limited, as is the limitation inherent in a hospital-based study. Future studies can be planned including larger number of subjects extending over wider geographical regions, finally expanding to community based surveys. This display is generated from NISO JATS XML with jats-html.xsl . The XSLT engine is SAXON 9.1.0.2 from Saxonica.