Introduction
The world has experienced a significant paradigm shift towards the emergence of non-communicable diseases.1 This includes the spectrum of coronary artery disease, stroke, diabetes mellitus (DM) hypertension (HTN) and the metabolic syndrome (MS).2 The number of deaths secondary to non-communicable diseases has shown an exponential escalation from 26.5 million in 2016 to a whopping mark of 40.5 million in 2016.3 A common culprit underlying the pathogenesis of all the above said is reported to be obesity. Obesity was initially reported to be an affliction of the western world but according to various reports is escalating at alarming rates in the developing world. In 2016 according to a small survey, it was found that among other risk factors, 24 million people in the country suffered from obesity.4 Worse still is the fact that in developing nations it is affecting the lower socioeconomic strata, which results in an increased disease burden in already, challenging economic situations.5 In Pakistan, the prevalence of overweight and obesity has been reported as 25% and 10%, respectively.6
Obesity is measured internationally by the body mass index (BMI) however, emerging data suggest that at lower BMI the Asian population has an increased amount of total body fat as compared with the Caucasians.7 For this reason, new definitions of obesity measurements have emerged such as total body fat and percentage body fat, which are measured by various modalities such as dual energy absorptiometry, CT and MRI. Furthermore, the distribution of obesity described as superficial fat adiposity (SFA) more marked in the subcutaneous compartment of the lower limbs and visceral adiposity in the abdomen is of significant consideration.8 Visceral fat adiposity (VFA) which is a component of central obesity is associated with the presence of various risk factors such as the presence of DM and the MS and coronary artery disease (CAD).9 Current studies show measurement of VFA in the abdomen with CT guided software applications.10
Pakistan is a third world developing nation that is undergoing similar changes of rapid urbanisation and development. This is leading to simultaneous affliction of communicable and non-communicable diseases.11 Furthermore, due to lack of proper awareness the effects of industrialisation are leading to increased prevalence of risk factors such as obesity.12 Evaluation of central obesity with volumetric analyses is a relatively novel concept in Pakistan. No data are available for our population about the prevalence of obesity or its correlation with cardiovascular risk factors and MS. Therefore, this study will help evaluate the prevalence of MS and its association with central obesity measured by CT software in the umbilical region as well as other anthropometric measures of waist circumference (WC) and BMI as well as association with established cardiovascular risk factors and the presence of MS.
-
To assess the prevalence of MS in asymptomatic population.
-
To assess the association of central obesity measures (VFA, SFA and WC) with MS, as well as cardiovascular risk factors (DM, HTN).
Methods
Study settings
This prospective study was conducted at a tertiary care hospital in Karachi, Pakistan, to determine the prevalence of MS and its association with central obesity measures and cardiovascular risk factors.
Sample selection and study participants
In the inclusion criteria, all participants with age above 18 years who had unenhanced CT abdomen examination and relevant blood workup were enrolled.
Exclusion criteria excluded all patients with known clinical history of CAD, HTN and DM as well as pregnant patients.
Sample size
The sample size was calculated using the formula of Z2 P(1−P)/d2 for prevalence studies, at level of significance at 0.05 and CI of 95%. The prevalence of MS in the Pakistani population was taken at a level of 0.49 with a relative precision of 0.08.13 A sample size of 155 patients was estimated to determine the prevalence of MS and to assess its correlation with measures of central obesity and cardiovascular risk factors.
Data collection
Clinical data
The data were collected prospectively. Due to the absence of an obesity screening programme, patients were recruited through the general health screening programme who presented to the general health screening clinic and were undergoing unenhanced abdominal CT scans for generalised abdominal complaints from 2015 to 2022. Study participants came after a 10–12 hours fast on the day of doctor’s appointment for their routine blood samples, which were part of their general health screening. After the initial physical examination by a physician and registered nurse, blood pressure was obtained by an automatic calibrated device in the sitting position in the clinic according to the method described by the American Heart Association. All anthropometric data such as height, weight, BMI and WC were collected in the clinic. Smoking was determined by the patient’s history from review of medical records. All data were collected in a structured questionnaire.
Thereafter patients were sampled for serum cholesterol, serum triglycerides, serum low-density lipoproteins (LDLs) and serum high-density lipoproteins (HDLs), fasting blood glucose and HbA1c. All these tests were part of the screening health package that the subjects were undergoing for general health screening. Data for all laboratory parameters were extracted from radiology information system.
HTN was defined as blood pressure >140/90 mm Hg or >130/80 mm Hg.14 DM was defined as any one of the following: fasting blood sugar at or above 110 mg/dL (7.0 mmol/L)/random blood sugar at or above 200 mg/dL (11.1 mmol/L).15 Obesity was classified as being overweight with a BMI 23–24.9 kg/m2 or obese with BMI ≥25 kg/m2 as defined by WHO.16 Smoking history was classified significant as >1 pack/day or ex-smoker defined as an individual who stopped smoking ≥2 years ago.17 Dyslipidaemia was defined as cholesterol >200 mg/dL and LDL >100 mg/dL.
MS was defined as any three of the following criteria.
-
The presence of abdominal obesity, with WC, in males >40 inches; females >35 inches.
-
The presence of triglyceride levels >150 mg/dL.
-
High-density lipoproteins (HDL) <40 mg/dL in males; <50 mg/dL in females.
-
The presence of HTN systolic ≥130 and diastolic ≥85 mm Hg.
-
Fasting glucose levels of ≥110 mg/dL.18
Volumetric CT adiposity measurements
After the initial physician evaluation and blood sample collection, the study participants were referred to radiology for relevant examination such as a chest X-ray and unenhanced abdominal CT as per their clinical indication. The scan parameters for abdominal non-contrast imaging included sections of the chest from the level of T3 to the symphysis pubis.
Volumetric analyses of visceral adiposity and subcutaneous adiposity were performed at the lumbar 4–5 levels in supine position, and single 5 mm CT slices were taken during suspended respiration after normal expiration. The fat areas in each subject were determined from an image at the level of the umbilicus using commercially supplied software on a 320-slice Toshiba scanner. Subcutaneous fat was taken as the extraperitoneal fat between the skin and muscles, with attenuation ranging from −150 to −50 HU. The intraperitoneal part with the same density as the subcutaneous fat layer was defined as visceral fat. The VFA and SFA were determined by automatic planimetry.19
Data entry was done on Excel and then re-entered in SPSS data analysis on SPSS V.19 thus double-checking all data entries.
Patient and public involvement
Patients were not involved in the research question, design, methodology and outcomes of the study.
Statistical analysis
Collected data were analysed using SPSS V.19.0. Normality assumption was assessed for continuous variables using the Shapiro-Wilk test. Mean±SD was calculated for normally distributed variables, whereas median (IQR) was calculated for non-normally distributed continuous variables. For categorical variables, frequencies and percentages were determined. χ2 and Fisher’s exact tests were used to determine the differences in the distribution of independent variables among individuals with HTN, diabetes and MS separately. T-test and Mann-Whitney U tests were used to test the association of gender, diabetes, HTN, and MS with VFA, and SFA. To assess the significance, the p value cut-off was kept at 0.05.
Results
Baseline characteristics
The study enrolled 165 patients. The study population’s average age was 43.2±13.7. 124 male subjects (75.2%) were listed according to the inclusion criteria. 112 subjects (67.9%) had a BMI of >23 kg/m2, and 75.2% had a family history of heart disease. Only 29.7% of the participants had HTN and MS, whereas 23.2% had diabetes. The mean VFA was 126.7±58.5, and the mean SFA was 225.5±102.6. According to the laboratory results, the mean cholesterol level was 172.5 g/dL. 85 subjects (51.5%) had above-ideal LDL levels, whereas 80 subjects (50.9%) had optimal HDL levels (table 1).
Subject characteristics with proportions (%) and SD
Study outcome
Participants were stratified into groups based on their comorbidities, the clinical outcomes, which included diabetes, HTN and MS. There were significant variations in the mean age of participants with diabetes and HTN (p value <0.0001). These comorbidities were prevalent in older people (p≤0.0001). Similarly, subjects with diabetes and HTN had higher median WCs (p=0.01 and p=0.001, respectively) as compared with subjects with normal blood pressure and glycaemic control. Mean HDL was higher in subjects with a normal glycaemic level (p=0.007) than subjects with diabetes whereas no statistically significant difference in mean HDL was found in subjects having HTN than those without it (p=0.7) (table 2).
Association of age, gender and other covariates with hypertension and diabetes mellitus
Males had a higher proportion of MS (59.2%) than females (40.8%) (p=0.002). Similarly, mean BMI (81.6%) was higher in subjects with MS than those without it (18.4%) (p=0.01). There were no statistically significant variations in smoking history, cholesterol or LDL levels between individuals with and without MS (p=0.4, p=0.7 and p=0.4, respectively) (table 3).
Association of age, gender and other covariates with metabolic syndrome
VFA was found to be linked with diabetes among research participants, in whom mean VFA (152.7±55.4, p=0.001) was higher in group with diabetes than those without diabetes. Another substantial association of VFA with HTN and MS (p<0.0001) was found, in the group with HTN and MS having higher mean VFA (161.3±54.3 and 155.3±7.9), respectively than subjects who were disease free. Similarly, median SFA was found to be significantly higher in the female gender (p=0.0004) than in males, subjects with HTN (p=0.0005) than those without it, and those having MS (p<0.0001) (table 4).
Association of CT body fat measures with gender and other comorbidities
Subset analysis of VFA and BMI with outcome variables
For the sake of further analysis, tertiles were calculated for VFA so that comparison of BMI categories (normal, overweight and obese) could be evaluated with VFA for the presence of MS using χ2 in the male subjects (n=124). Although both the obesity measures, VFA and BMI revealed a significant association with MS, in subjects with BMI more than 26 kg/m2, only 20 out of 61 males (33%) had MS while in the highest tertile of VFA, 14 out of 31 (47%) subjects had MS. Male subjects, in the highest tertile, also revealed a significant association with HTN where 17 out of 30 subjects (55%) had HTN. This was not a significant outcome in the obese category of BMI where only 20 subjects out of 61 in the obese category had HTN. Diabetes did not show a significant association with either the BMI category of >26 kg/m2 or the highest tertile of VFA (table 5).
Association of third quartile of the visceral fat adiposity (VFA) and the obese category of body mass index (BMI) with metabolic syndrome, diabetes and hypertension
Discussion
In this study, mean BMI, VFA and WC were all found to be higher in individuals having MS, diabetes and HTN in comparison to the disease free cohort. SFA was also found to be higher in individuals with MS and HTN than those without it. The prevalence of MS and HTN was found to be 29.7% whereas DM was seen in 23% of the individuals.
MS is a multifactorial disorder that has been implicated as a culprit in multiple non-communicable diseases. In Pakistan, Iqbal Hydrie et al reported it as high as 46% when different definitions of the syndrome were used in 2007.13 In 2019 another local study reported a prevalence of 95.7% of MS in adults who were above a BMI of 23 kg/m2.20 In a recent meta-analysis published in 2023, the pooled prevalence of MS was found to be 28.8% with a very high prevalence in the province of Sindh due to urbanisation (63.7%).21 The latter result is very close to this study where a prevalence of 29.7% was found in asymptomatic study subjects.
One of the recurring risk factors and predictor of MS has been postulated as obesity. Various measurements of adiposity have been associated with MS. BMI alone has not emerged as a significant predictor of obesity as it does not quantify the amount of fat in different body compartments.22 23 In this study we found a significant association of BMI in subjects who had MS however a significant proportion of subjects with an increased BMI of >23 kg/m2 did not have MS raising the possibility of compartmental fat distribution such as increased SFA which is a lesser predisposing factor for MS. Similarly, in subjects with BMI >26 kg/m2 (n=60) only one-third of the subjects had MS whereas approximately 50% of patients having the highest tertile of VFA had MS thereby signifying the metabolic effect of the visceral fat component as compared with generalised body mass (table 5).
There was a significant disparity in the prevalence of MS within the gender distribution. Among 41 female study participants, half of them (n=20) had MS. Since this was a sample of asymptomatic patients, this was a significant proportion of patients with the presence of MS. Although previous studies have reported a higher prevalence in male subjects, this study may warrant a more gender classified research regarding the prevalence of MS with respect to gender.24 A possible explanation for the lower prevalence of female subjects may be because in a low- and middle-income country state females may have restricted opportunities for general healthcare and therefore would exist with subclinical disease.
Central obesity has already been implicated in multiple studies as a predictor of MS and is one of the five criteria in the diagnosis of MS in the form of WC.23 Further classification of central obesity has led to the differentiation of fat in the visceral compartment, intraperitoneal fat (VFA) and that in the subcutaneous region known as SFA. Distribution of fat in the visceral space has emerged as a harmful predictor for MS and cardiovascular disease.25 In this study, there was a significant association of mean volume of VFA with DM, HTN and MS while SFA was associated with HTN and MS only. Furthermore, in gender-based analysis, SFA was significantly higher in females as compared with male subjects. Matsha et al reported a similar result in which they found VFA as a significant predictor for MS in both males and females while increased quartiles of SFA revealed positive prediction with MS in male subjects only.26 In this study, a significant association of highest tertile of VFA was found with HTN and MS while subjects with BMI in the obese category revealed a significant association only with MS not with HTN or DM (table 5). In 2015, Shah et al carried out adiposity assessments in the MESA study participants and found that significant HR and OR were seen in serial readings of VFA with incidence of MS.27 This was further reiterated in a population based, large sample size study, carried out by Lee et al in which they found a significant association of VFA with MS as compared with WC and BMI in the Korean population.28 These findings were previously validated in 2007, on a sample of Japanese Americans in which intra-abdominal fat adiposity estimated on CT had a greater receiver operating characteristic value than other central obesity measures.10
HTN has also been postulated as having a significant association with central obesity. Multiple studies have established the relationship of obesity and BMI with HTN. In 2020, Sigit et al found HTN to be the most prominent component in a very large sample size of Indonesian and Dutch population cohorts. However, they failed to establish a relationship between HTN and abdominal obesity.29 In this study, 29.7% (49) subjects had HTN with a highly significant association of HTN with visceral adiposity (p value 0.0001). Superficial adiposity also revealed significance in its association with HTN. Sironi et al further reiterated that in newly diagnosed males with HTN, the body fat distribution had a disproportionate distribution in the visceral compartment, that is, the intrathoracic and intra-abdominal compartments resulting in raised systolic pressures and insulin levels.30 Another study conducted on the Chinese population in 2021 revealed the association of visceral adiposity index with the degree of arterial stiffness in individuals with normal weight and BMI.31 In India, Goswami et al found that a significantly high percentage of visceral body fat was associated with HTN and dyslipidaemia.32
Goswami et al also found a strong association between diabetes and visceral fat in their study with a higher prevalence of dyslipidaemia in the diabetic cohort.32 Ishihara et al revealed that individuals with a higher visceral fat content were thrice more likely to have type 2 diabetes as compared with individuals with lower VFA.33 In this study 23% of the patients had diabetes and had a significantly higher amount of visceral body fat as compared with non-diabetics. In a Japanese cohort, there was a significant association of VFA with type 2 diabetes in males however significant association of the third and fourth quartile of SFA was found in females in addition to the higher amount of VFA in diabetic females.34
The main limitation of the study is a small sample size. As a proper health screening programme does not exist in the country participants could only be contacted through the database where they were acquiring unenhanced CT examinations for other clinical indications. Despite that care was taken to evaluate the variables for normal distribution and appropriate statistical tests to ensure validity of results.
As this was the first study for evaluation of central adiposity, a number of research questions have risen. Considering the higher prevalence of MS in women, a gender-based sample size warrants evaluation for adiposity measurements and their associations. Validation of anthropometric measurements with volumetric analysis may serve as a useful tool for evaluation of adiposity in the field settings.
Conclusion
In conclusion, VFA which is a central adiposity measure estimated on CT volumetric analysis shows a significant association with MS, DM and HTN. The prevalence of MS was 29.7% and approximately 50% of female subjects had MS. The highest tertile of VFA revealed a significant association with MS in male patients. In addition, SFA reveals significant association with DM only.
This post was originally published on https://bmjopen.bmj.com