We performed a cross-sectional study from January to September 2010 in a cohort of HIV-infected patients coming for their checkup at the day Hospital of the Central Hospital in Yaoundé. This hospital is a reference for taking care of HIV-infected patients. It is a structure assigned for the management of HIV-infected patients including counseling, ART administration, and biochemical analyses. All patients willing to participate in the study have been allowed to pass through all the steps of the study, since they are a vulnerable population, any frustration can affect them psychologically. But after that, only the patients meeting our criteria have been included in the study (the inclusion and exclusion criteria section gives more details on that). Data were collected at one point in time for each participant. After giving their informed consent, all participants were asked to complete a self-administered pre-tested questionnaire with information regarding age, sex, level of education, marital status, profession, and characteristics related to HIV infection (ART duration, being on treatment or not); confirmation was done using medical records. The study involved men and women aged ≥20 years old and the presence of metabolic disorders was described according to the BMI.

2.2. Sample Size Determination

At the end of the data collection, 492 patients were interviewed. The sample size was calculated based on the prevalence of HIV-infected patients in Cameroon using the Magnani (1997) [10] formula:

$n=\left(t^2\times p\left(1-p\right)\right)/m^2$

where n is the required sample size; t is the 95% confidence interval (standard value of 1.96); p is the estimated prevalence of HIV in Cameroon (4.3%) [11] , and m is the margin of error at 5% (standard value of 0.05). Thus, 64 subjects were obtained after calculations. However, considering a probable loss of about 10% of the sample and the huge population of HIV-infected patients hosted in the Central hospital in Yaoundé, a reference in taking care of HIV-infected patients, we focused on 500 participants, and 492 were obtained. The study was implemented in the Centre region of Cameroon.

Participants included in the study were HIV-infected patients coming to the Central Hospital of Yaoundé either for follow-up or to be enrolled as newly diagnosed HIV-infected patients. Each participant included in the study had met the following criteria: 1—had given informed consent; 2—was 20 years of age; 3—had come for a check-up at the day hospital of the central hospital of Yaoundé; 4—had lived in the city of Yaoundé and its surroundings. Those not included were: 1—Pregnant or breastfeeding women; 2—participants on medications that can affect carbohydrate and lipid metabolism; 3—seriously ill participants (bedridden or in terminal phase with AIDS); 4—with developmental disabilities.

The Cameroon National Ethics Committee for Research on Human Health approved the study at N˚138/CNE/SE/09 and the Ministry of Public Health at D30 47/AAR/MINSANTE/SG/DROS/CRC/CEAI. Jan 2010. An approval letter was obtained from the hospital authority (the General Manager). Informed written consent was obtained from the individual respondents before the beginning of the research. It was explained to them the voluntary nature of their participation, their freedom to withdraw from the study at any time, the confidentiality and privacy measures maintained by assigning a code to each participant, and venous blood collection. The participants with more severe conditions of obesity were referred to the endocrine unit of the same hospital. The study protocol and conduct adhered to the principles of the Declaration of Helsinki.

A structured questionnaire based on the WHO’s STEPwise instrument for chronic diseases v 2.1. was used to collect socio-demographic data (age, sex, educational level, marital status), HIV characteristics (ART status (on treatment or not) and ART duration), smoking status, alcohol consumption, and level of physical activity. The questionnaires were prepared in French and translated into English so English-speaking participants could understand. The pre-testing was done on five volunteer patients. The modifications were made after that to make the questions more understandable. Trained personnel administered the questionnaires through a face-to-face interview with the participants. The following socio-demographic data were obtained. Concerning physical activity, the participants auto-reported their level of physical activity as intense, moderate, or light activity, which had been put into two sections: current physical activity, and no physical activity. Alcohol consumption was categorized as no consumption (no consumer) and current consumption (consumer). The smoking status was also categorized as no smoking and current smoking. The education level was classified into four categories: no education, primary education, secondary education, and higher education level (university). In the marital status, five categories were observed: widower, divorced, single, married, and cohabiting. Concerning HIV characteristics, the patients self-reported their treatment status (being on treatment or not), and their duration of treatment, which was classified into four categories: on treatment for less than one year, between one and two years, between two and four years, and more than four years. These self-reported responses were confirmed using the patient’s medical reports.

Blood pressure was assessed using a validated electronic device (OMRON) after the subjects were seated and rested for at least 10 minutes. The tight-fitting clothing of the upper left arm was removed, and the arm was positioned so that the cuff was leveled with the heart. The measure was then automatically taken. The higher blood pressure was evaluated according to the International Diabetes Federation (IDF) definition, as systolic blood pressure was greater and equal to 130 mmHg, and diastolic blood pressure was greater and equal to 85 mmHg. Height was determined using a portable stadiometer to the nearest 0.1 cm. The patient was made to remove his/her footwear, and stand feet together and arms at the sides. The heels, buttocks, and upper back were against the straight edge in a completely upright position. Measurements were taken into centimeters and then expressed in meters. Weight and body fat were measured with an electronic Tanita impedance device (Tanita UM073). Participants were standing on the weighing scale wearing barefoot and light clothes. The weight in kilograms was obtained. Three measurements were done and the average was taken. Body mass index (BMI) was calculated as weight in kilograms (kg) divided by the square of the height in meters (m). Patients were categorized according to their nutritional status as underweight (<18.5 kg/m²), normal weight (18.5 - 24.9 kg/m²), overweight (25 - 29.9 kg/m²), and obese (≥30 kg/m²). From these four groups, we divided our study population into two groups as these are our targeted values: the underweight/normal weight category and the overweight/obese category. The waist circumference was measured with a flexible inelastic tape placed at the midpoint between the lower rib margin and the iliac crest in a perpendicular plane to the long axis of the body without restrictive garments. The measurements were recorded to the nearest cm using a flexible non-expandable tape measure. The presence of abdominal obesity was evaluated using IDF definitions: according to IDF criteria, we have abdominal obesity when waist circumference is >80 cm in women and >94 cm in men [12] .

The blood samples were taken after a fast of 12 hours by the Central hospital personnel. The venous blood was collected at the fold of the elbow using a Vacutainer (5 ml per tube), without anticoagulants, and two others with anticoagulants (heparin and citrate) for the plasma. Each tube before blood collection was assigned a code for each of the patients involved in the study. The plasma and serum were obtained after centrifugation at 3500 tours/minutes for 10 minutes. The biological material was fractionated into aliquots of 200 µl. One was used to immediately evaluate blood glucose level and the other was kept for further use. The collected aliquots were frozen at −20˚C for further biochemical analyses.

Glucose was measured using the Glucose Oxidase-Peroxidase (GOP-POD) method [13] . Total cholesterol [14] , and triglycerides [15] were quantified by standard enzymatic spectrophotometric methods using ChronoLab Diagnostic Kits in the laboratory. HDL cholesterol was measured using a heparin manganese precipitation of Apo B-containing lipoproteins [14] . LDL cholesterol was calculated using the Friedwald equation [16] . CD4 cell count was obtained with flux cytometry.

Dyslipidemia was defined as total cholesterol ≥ 200 mg/dl and/or triglyceride ≥ 150 mg/dl, and/or HDL Cholesterol < 40 mg/dl in men, and <50 mg/dl in women, and/or LDL ≥ 160 mg/dl in men and ≥150 mg/dl in women [17] . Immunodepressed patients were those with a CD4 cell count of less than 350 cells/mm³ and immunocompetent patients were those with a CD4 cell count equal to and greater than 350 cells/mm³. The metabolic syndrome was evaluated using the International Diabetes Federation (IDF) as one of the appropriate definitions for the Cameroonian population [18] . According to IDF criteria [12] , waist circumference (≥80 cm in women and ≥94 cm in men) is a prerequisite in addition to ≥2 of the following components: fasting triglyceride levels ≥ 150 mg/dL, high-density lipoprotein (HDL) cholesterol level < 40 mg/dL for men and <50 mg/dL for women, fasting glucose levels ≥ 100 mg/dL, or hypertension (blood pressure ≥ 130/85mmHg or current receipt of medication for hypertension). Nutritional status was defined as follows: underweight (BMI < 18.5 kg/m²), normal (BMI: 18.5 - 24.9 kg/m²), overweight (BMI: 25.0 - 29.9 kg/m²), and obese (BMI ≥ 30.0 kg/m²) [17] .

Data collected from the study participants were entered into an Excel spreadsheet and double-checked for errors. Data were analyzed using IBM SPSS (Statistical Package for Social Analysis) Statistics version 21.0 for Windows. The data were expressed as frequencies. The main dependent variable was body mass index (BMI), which showed the nutritional status and was divided into classes: 1 = underweight/normal weight, 2 = overweight/obese. The independent variables were socio-demographic parameters (age, gender, marital status, and educational level), behavioral risk factors (physical activity, alcohol consumption, and smoking status), and HIV-related characteristics (being on treatment or not, ART duration, CD4 cell counts). Age was categorized as less than forty years and greater and equal to forty years. Males and females constitute our study population. Before the analysis, the continuous variables were tested for normality. Visual inspection of the histogram was done. The study participants were characterized using descriptive statistics regarding socio-demographic factors, and HIV-related clinical factors using frequency. The association between nutritional status, the dependent variable and the other socio-demographic and clinical variables, and waist circumference were assessed using the Chi-square test (categorical variables), and their P-values were presented. The logistic regression model (bivariate analysis) was applied to measure the independent roles of different predictors or risk factors associated with overweight/obesity (calculating the crude odds ratio (OR) at 95% CI). The statistical significance was set as a P value of 0.05.

In this study, we had 492 patients, of which 338 were women (74.7%) and 117 men (25.3%); the most represented age group was 30 to 39 years old (41.9%). The prevalence of overweight and obesity in our study population was 27.5% and 8.1%, respectively, and 40.5% of patients had a high waist circumference. According to CD4 count cells, the patients who had a count of <350 cells/mm³ were mostly represented. Frequent alcohol drinking was observed in 32.7% (n = 69) of patients. While 73.1% (n = 337) of patients had a low rate or had never done physical exercise. Regarding the educational level, secondary education was mostly represented (59.9%; n = 270). The married patients represented one-third of the study population (35.3%, n = 160). These are shown in Table 1 .

Table 2 shows the relationship between BMI and selected risk factors. There was a significant association between nutritional status and abdominal obesity

BMI: Body Mass Index; LDL-c: Low-density lipoprotein- cholesterol, HDL-c: High-density lipoprotein- cholesterol, MetS: Metabolic Syndrome *p < 0.05.

(p = 0.0001), systolic blood pressure (p = 0.007), diastolic blood pressure (p = 0.012), metabolic syndrome (p = 0.0001), and CD4 count cells (p = 0.007).

It emerges from Figure 1 that abdominal obesity, hyper-TG, hypo-HDL, elevated blood pressure, and hyperglycemia were more prevalent in overweight/obese patients than underweight/normal patients in the study population. The metabolic syndrome was also highly observed in patients with overweight/obesity compared to those who were underweight and of normal weight (55.5% and 17.8% respectively).

Table 3 shows independent predictors of overweight and obesity in our study population. The presence of abdominal obesity [p = 0.0001; OR: 5.820 (3.409 - 9.935)], elevated SBP [p = 0.014; OR: 1.955 (1.144 - 3.343)], CD4 count cells ≥ 350 [p = 0.016; OR: 2.078 (1.144 - 3.777)] and MetS [p = 0.0001; OR: 5.768 (3.344 - 9.948)], were positively associated with overweight/obesity. These factors may significantly increase overweight and obesity in our study population. We noted also that elevated DBP [p = 0.026; OR: 0.563 (0.340 - 0.933)] doesn’t increase the risk of being overweight and obese, it is instead the absence of elevated diastolic increased risk, which is not common). Also, these factors have been analyzed

Independent predictors of overweight/obesity in our study population are abdominal obesity, high systolic, and low diastolic blood pressure, CD4/count cells, and metabolic syndrome; *p < 0.05.

using logistic binary regression.