Introduction

The advent of highly active antiretroviral therapy (HAART) has modified the prognosis of human immunodeficiency virus (HIV) infection by reducing the risk of death associated with the condition and improving the quality of life of people living with the infection [1]. Antiretroviral therapy (ART) slows down the development of acquired immunodeficiency syndrome (AIDS) and/or improves the condition of those with full-blown AIDS [2, 3]. Universal access to ART has led to a significant reduction in the morbidity and mortality associated with HIV/AIDS [2, 4].

That notwithstanding, ART causes varied changes in lipid metabolism resulting in hypertriglyceridaemia, hypercholesterolaemia, as well as other metabolic disorders such as insulin resistance, hyperglycaemia and redistribution of body fat which are all risk factors of cardiovascular diseases [5,6,7]. It has been posited that patients living with HIV/AIDS often have a set of body changes associated with fat redistribution, truncal obesity and peripheral wasting [8]. The changes may develop at any time ranging from about six weeks to several years following the initiation of HAART [9]. The changes are referred to as metabolic syndrome. Metabolic syndrome is therefore associated with antiretroviral therapy, as well as dyslipidaemic lipodystrophy associated with HIV and HAART [10,11,12].

The criteria for metabolic syndrome include five variables, namely, abdominal obesity, raised triglycerides, low levels of high-density lipoprotein (HDL), elevated blood pressure and a history of diabetes mellitus or impaired fasting blood glucose state. Several definitions have been suggested for metabolic syndrome. These include the United States of America National Cholesterol Education Program – Adult Treatment Panel (NCEP-ATP) III criteria, International Diabetes Federation (IDF) definition and the World Health Organization (WHO) definition [13]. The prevalence rate of metabolic syndrome among patients infected with HIV ranges from 17% to 45.4% [14, 15]. Additionally, the prevalence rate of metabolic syndrome among people using HAART ranges from 14 to 18% according to WHO criteria, IDF criteria and NCEP-ATP III criteria [16].

In Ghana, the prevalence rate of metabolic syndrome ranges from 24.5% to 48.3% according to WHO criteria, IDF criteria and NCEP-ATP III criteria. Specifically, it is 24.5% according to WHO criteria, 42.3% using the IDF criteria, and 48.3% using the NCEP‑ATP III criteria. However, there are limited research studies on the prevalence of metabolic syndrome in HIV-infected patients [9]. It is in view of this that this study aimed to determine the prevalence rate and clinical factors associated with metabolic syndrome among HIV-infected patients on antiretroviral therapy attending clinic at a district hospital in Ghana.

Methods

Research Design and Setting

This study used a quantitative descriptive cross-sectional research design to determine the prevalence rate and clinical factors associated with metabolic syndrome among HIV-infected patients on antiretroviral therapy. The study was conducted at Tema General Hospital, which is a district hospital and a major referral center in the Greater Accra Region of Ghana. It is the main and largest public health facility which serves the people of Tema metropolis and its environs. It is a 294-bed capacity health facility which provides general and specialist services. The daily average out-patient attendance of the hospital is 468. The hospital has 14 departments and units which include obstetrics and gynaecological unit, surgical unit, medical unit, paediatric unit, HIV/AIDS clinic, radiology, physiotherapy, ear, nose and throat clinic, and eye clinic amidst others. The HIV/AIDS clinic is managed by qualified and skilled health professionals who render care and health services to patients infected with HIV/AIDS.

Sample Size and Sampling Technique

The target population for the study included all patients with HIV/AIDS attending clinic at Tema General Hospital. Participants selected for the study included patients diagnosed with HIV/AIDS who were attending clinic at the HIV/AIDS clinic of Tema General Hospital, were 18 years and above, and had received ART at the HIV/AIDS clinic for at least 6 months. The study excluded patients with a documented medical history of comorbidities such as obesity, diabetes, tuberculosis and hypertension before the start of ART, those who were receiving HAART for other viral infections like Hepatitis B, patients who were very sick and pregnant women.

Ethical clearance for this study was granted by the Committee on Human Research, Publication and Ethics of Kwame Nkrumah University of Science and Technology (CHRPE/AP/406/19) and the head of the HIV/AIDS clinic of the district hospital where the study was conducted. Patients with HIV/AIDS attending clinic at the HIV/AIDS clinic of Tema General Hospital were informed about the study and the requirements for participating in it. They were then encouraged to ask questions and clarify all doubts they had about the study. After answering all questions to their satisfaction and clarifying all doubts, those who met the inclusion criteria and willingly agreed and consented to be part of the study were recruited as study participants. Guided by the inclusion criteria and Yamane’s formula for determining a sample size, 240 HIV‑infected patients aged 18 years and above were selected for the study. Convenience sampling technique was used to recruit participants for the study.

Data Collection

A structured questionnaire was used to collect data from the 240 participants. The questionnaire was initially piloted on 10 patients diagnosed with HIV/AIDS and on ART, who were attendants of an HIV/AIDS clinic at a district hospital in Ghana. None of the 240 study participants dropped out of the study. The questionnaires were administered by the researchers. The researchers made sure that the questionnaire was developed in line with the objectives of this study. Data collected included participants’ demographic characteristics, alcohol intake, tobacco use, usage of recreational drugs, as well as their personal medical history, focusing on obesity, diabetes and hypertension before the use of ART. Also, three consecutive blood pressure measurements of each participant was recorded using a mercury sphygmomanometer with appropriate cuff sizes and a stethoscope, after he or she was seated for at least 15 min. An average of the three readings was then used in the data analysis. In addition, participants’ weights were checked using a weight measuring scale and recorded to the nearest 0.5 kg. Furthermore, their heights were checked without them being in any footwear using a well‑calibrated wall‑mounted rule and recorded to the nearest 0.1 m.

Body mass index (BMI) was calculated based on the participants’ weights in kilograms divided by the square of their heights in meters. Each participant’s waist circumference (WC) was also measured to the nearest 0.1 cm horizontally at the narrowest point between the lower end of the rib cage and iliac crest. Again, hip circumference was measured to the nearest 0.1 cm at the greatest horizontal circumference below the iliac crest at the level of the greater trochanter (the widest portion on the buttocks). Waist and hip circumference were measured with an inelastic tape measure. Waist to hip ratio (WHR) was then calculated from the waist and hip circumference.

Data Analysis

Data collected was analyzed with Statistical Package for Social Sciences, version 22.0. Baseline characteristics were summarized using medians and inter-quartile range (IQR) for continuous variables. Chi-square test was used to test for statistically significant associations between categorical variables. Bivariate logistic regression was used to evaluate characteristics associated with the presence of metabolic syndrome. A two-tailed p-value of less than 0.05 was considered statistically significant.

Results

Demographic Characteristics of Participants

The demographic characteristics of participants are shown in Table 1. The findings showed that 65.4% of the participants were females (n = 157) whiles 34.6% were males (n = 83). The ages of the respondents ranged from 21 to 69 years with a mean age of 40.9 years (SD = 10.9 years). Most of the participants (40.4%) were in the 31 to 40 year group with a modal age of 34 years. The median age of the participants was 39.5 years.

Table 1 Demographic characteristics of participants
Full size table

More than half of the participants were single (52.5%), while less than a quarter of them were married (22.1%), with a few of them divorced (18.3%). Regarding their educational level, 37.1% of the participants had had up to senior high school education (n = 89), 33.7% had had up to junior high school education (n = 81), 13.7% had had up to the primary school level (n = 33), 11.3% had had up to the tertiary education level (n = 27), while only 4.2% of them had no formal education (n = 10). The majority of participants were employed (64.2%) and reported employment as their source of income, while 35.8% mentioned that they were unemployed and had family as their source of income.

Prevalence of Metabolic Syndrome Among Patients on ART

The anthropometric biochemical measurements of participants and the prevalence of metabolic syndrome among them have been revealed in Table 2. The prevalence of metabolic syndrome among the study population was 17.1%, 27.9% and 20.4% according to the WHO, NCEP‑ATP III and IDF criteria respectively.

Table 2 Prevalence of metabolic syndrome using the WHO, NCEP-ATP III, and IDF criteria
Full size table

Factors Associated with the Development of Metabolic Syndrome

Table 3 summarizes the baseline characteristics of all the participants, as well as a comparison of the participants with and without metabolic syndrome. The prevalence of metabolic syndrome was 27.9% as per the NCEP-ATP III criteria (n = 67). All of the findings in this section were based on the NCEP-ATP III criteria.

Table 3 Significant variables associated with metabolic syndrome
Full size table

The findings of the study revealed that the median age of participants with metabolic syndrome was 43 years (IQR: 38 – 48 years, p = 0.001), and 36 years for those without metabolic syndrome (IQR: 30 – 42 years, p = 0.001). Per participants’ demographic characteristics, 30.5% of the female participants (n = 48, p < 0.001) had metabolic syndrome, whereas 22.9% of the males had it (n = 19, p < 0.001). Hence, among the participants with metabolic syndrome, females (r = 0.54, p < 0.001) recorded a higher frequency compared to their male counterparts.

Among the participants with metabolic syndrome, most were not married (n = 54, r = 0.387, p = 0.014). The majority of participants with metabolic syndrome (58.8%) reported tobacco use (r = 0.693, p = 0.019), while 23.0% of them reported current alcohol consumption (n = 14, r = -0.239, p = 0.000). A negative association was found between alcohol use and metabolic syndrome (r = -0.239). The median duration of ART among participants with metabolic syndrome was higher at 64 months (IQR: 50 – 68) compared to those without metabolic syndrome at 42 months (IQR: 38 – 49) (r = 0.341, p = 0.010).

The type of drug regimen adopted by participants and its association with metabolic syndrome revealed that, most of those with metabolic syndrome used ART drug combinations of nucleoside reverse transcriptase inhibitors (NRTIs) and protease inhibitors (PIs) (46.4%, n = 13), whereas just a little over a quarter of them with metabolic syndrome (25.5%, n = 54) were using a combination of NRTIs and non-nucleoside reverse transcriptase inhibitors (NNRTIs) regimen (r = 0.434, p = 0.002).

Clinical and Biochemical Parameters of Participants with Metabolic Syndrome

Table 4 summarizes the clinical and biochemical parameters of participants who had metabolic syndrome. The findings revealed that most of them had hypertension (85.1%), diabetes (80.6%), and high triglyceride levels (50.7%).

Table 4 Clinical and biochemical parameters of participants with metabolic syndrome
Full size table

Determinants of Metabolic Syndrome

A bivariate analysis to identify the determinants of metabolic syndrome has been presented in Table 5. The results of this study showed that high triglycerides [OR = 6.44, 95%CI: (0.44–9.51), p = 0.002], high cholesterol [OR = 4.52, 95%CI: (0.21–9.32), p = 0.039], fasting blood glucose or diabetes mellitus [OR = 4.15, 95%CI: (1.13–8.69), p = 0.001] and duration of ART for at least 60 months [OR = 2.92, 95%CI: (0.76–7.67), p = 0.031] were associated with the development of metabolic syndrome.

Table 5 Bivariate analysis of the variables associated with metabolic syndrome
Full size table

The findings showed that participants who were not married were 1.13 times more likely to develop metabolic syndrome compared to their married counterparts [OR = 1.13, 95%CI: (0.45–1.67), p = 0.019].

The use of a PI-based ART regimen increased the odds of developing metabolic syndrome by about 2 folds [OR = 1.98, 95%CI: (0.29–7.02), p = 0.001]. No significant association was found for current alcohol use [OR = 0.36, 95%CI: (0.27–0.47), p = 0.216] and hypertension [OR = 6.13, 95%CI: (0.49–9.17), p = 0.063].

The study also found that 76.1% of participants who were obese (n = 51) developed metabolic syndrome as compared to 23.9% of those who were not obese (n = 16) [OR = 3.67 (95%CI; (0.05–6.99), p = 0.019]. It was revealed that there was about fourfold increased odds of metabolic syndrome among participants who were obese.

Among the variables studied, the odds for developing metabolic syndrome were significantly higher among participants who used ART regimens with PIs and NRTIs combinations [OR = 1.98, 95%CI: (0.29–7.02)], those who were at least 41 years [OR = 1.98 (95%CI; (0.29–1.78)], not married [OR = 1.13 (95%CI; (0.45–1.67)], were currently smoking [OR = 0.59 (95%CI; (0.18–2.13)], and were of the female gender [OR = 0.57 (95%CI; (0.19–00.99)].

Discussion

Prevalence of Metabolic Syndrome Among Patients on ART

The prevalence of metabolic syndrome in this study ranged from 17.1% to 27.9%. This is in contrast to the findings of another study which found that the prevalence of metabolic syndrome among persons using HAART was between 14% (IDF definition) and 18% (ATP III definition) [17]. The current study therefore revealed a slight increase in prevalence rates which can be attributed to the use of ART. The difference in findings was because the participants of the present study were not having metabolic syndrome before the initiation of ART. The strength of this study is the comparative use of the three different criteria to assess the prevalence of metabolic syndrome.

The prevalence of metabolic syndrome in this study per the NCEP‑ATP III criteria was 27.9% (n = 67). This was high compared to the prevalence rates of 17% reported by Jericó et al. [18] and 18% reported by Samaras et al. [16]. In contrast, studies by Strufaldi, Da Silva and Puccini [19], and Brown et al. [20] found prevalence rates of 13% and 10% respectively among their study participants. However, the prevalence of 27.9% per the NCEP-ATP III criteria found in this study was relatively lower compared to the prevalence of 30.2% per NCEP-ATP III criteria found by Dimodi et al. [21] among Cameroonian patients, and a prevalence of 48.3% per NCEP-ATP III criteria found by Obirikorang et al. [9] in a study done among Ghanaian patients.

Again, some studies reported a prevalence of 14% [16] and 11.4% [22] per the IDF criteria, which was far lower compared to the prevalence of 20.4% revealed in this study. A similar disparity was observed in other research studies [9, 21]. Using the WHO criteria in assessing metabolic syndrome among patients with HIV, this study found a prevalence rate of 17.1%. A study conducted by Obirikorang et al. [9] reported a higher prevalence rate of metabolic syndrome of 24.5% among their study participants using the WHO criteria. The possible reason for the varying prevalence rate of metabolic syndrome can be explained by the different criteria, duration of exposure to HAART, and the different sample size used. The difference in prevalence rates from this study and other studies, particularly those that were done in high-income countries, can also be attributed to the lifestyle differences and the generally higher burden of obesity in those countries.

Factors Associated with the Development of Metabolic Syndrome

The findings of this study revealed that age, ART duration, ART regimen, smoking status and gender were positively associated with the occurrence of metabolic syndrome. This finding corroborates with a study by Norris and Dreher [23] who indicated that factors that contribute to the development of metabolic syndrome in patients with HIV include advanced age. Similarly, Grabar, Weiss and Costagliola [24] also established that age is a risk factor associated with HAART-induced metabolic syndrome. In line with a study by Termizy and Mafauzy [25], increasing age is a risk factor of metabolic syndrome. It has also been posited that age is a significant risk factor for the development of metabolic syndrome among asymptomatic and antiretroviral treatment-naive adults with an odds ratio of 1.10 [95% CI (1.04–1.16), p < 0.01] [26].

The present study also highlighted a moderate association between the gender of participants and metabolic syndrome state, with a marked higher prevalence in women. This finding is in line with a study by Dimodi et al. [21] who espoused that the higher prevalence of metabolic syndrome in women can be explained by the higher frequency of HIV-infected women compared to men. The findings of this study was partly in tandem with the findings of a study by Ford, Giles and Dietz [27] who reported a metabolic syndrome prevalence of 20.5% in Mexican–American men and 35.5% in women. The higher predisposition of women who developed metabolic syndrome could be due to biological, psychological and environmental factors as indicated by Villamar, Albuja and Salas [28]. However, this finding is in contrast to the findings of a study by Norris and Dreher [23] who found that the male gender was significantly associated with the development of metabolic syndrome.

The median duration of ART among participants with metabolic syndrome in this study was higher at 64 months compared to those without metabolic syndrome at 42 months. HAART duration has been found to be associated with lipoatrophy [29]. The findings of this study was consistent with the findings of a study by Worm et al. [15] who indicated that, the possible explanation for the varying factors that lead to the development of metabolic syndrome among HIV-positive patients on antiretroviral therapy could be due to the duration of exposure to HAART.

Again, in this study, metabolic syndrome state was associated with the use of ART drug combinations of NRTIs and PI. It has been established that PI-based ART increases the total HDL cholesterol [30]. The findings of this study is consistent with the findings of a study by Reaven [31] who suggested that HAART plays a central role in the development of lipodystrophy and associated metabolic alterations.

Determinants of Metabolic Syndrome

This study found that high triglycerides levels, high cholesterol levels, fasting blood glucose levels or diabetes mellitus, and duration on ART for at least 60 months were associated with the development of metabolic syndrome. The findings of this study were in tandem with a study by Alencastro et al. [32] which posited that the highest prevalence of abdominal obesity, hypertriglyceridemia and low levels of HDL were associated with the development of metabolic syndrome.

The current study also found that participants who were not married were more likely to develop metabolic syndrome than their married counterparts. This finding is in consonance with the findings of other studies which have revealed that being unmarried increases one’s chances of having metabolic syndrome, cardiovascular diseases and poor health outcomes [33, 34]. This is because people who are married often have healthy lifestyles or better health behaviours than those who are not married. Hence, the increased risk of metabolic syndrome among unmarried people may be due to their poor or unhealthy lifestyle [34]. Again, the use of a PI-based ART regimen, hypertension, diabetes mellitus, obesity, increased cholesterol levels and high triglycerides levels resulted in increased odds of developing metabolic syndrome. In line with the findings of studies by Myong et al. [35], Sales et al. [26], and Termizy and Mafauzy [25], the odds for metabolic syndrome development increased with poor health habits and poor lifestyles such as smoking, unhealthy diet, lack of exercise and unhealthy alcohol consumption. The findings of this study is in tandem with the findings of a study by Sidorenkov, Nilssen and Grjibovski [36] which indicated that smoking is positively associated with metabolic syndrome.

Additionally, the findings of this study is in consonance with the assertion by IDF that, central obesity is a pre-requisite in the definition of metabolic syndrome. This study however found no association between metabolic syndrome and severity of obesity. Hence, it can be postulated that the severity of obesity does not influence the risk of having metabolic syndrome. However, once obesity sets in, the probability of having metabolic syndrome can be said to be higher than that of individuals who are not obese.

Limitation of the Study

This study was carried out at a district hospital in the capital region of Ghana with a unique setting and practices in the management of HIV/AIDS which might be different from other settings.

Conclusions

The prevalence of metabolic syndrome in this study was quite high, affecting more than a quarter of the participants. The factors associated with the development of metabolic syndrome were the female gender, long duration of treatment, being unmarried, being over 41 years of age, high cholesterol levels, fasting blood glucose levels, and high triglycerides levels. The most significant factors were being on NRTIs and PIs containing regimens. These findings suggest that patients on antiretroviral treatment with the above characteristics should be monitored regularly for metabolic syndrome. Since regimens containing NRTIs, NNRTIs and PIs seem to increase the risk of metabolic syndrome, patients on regimens containing these drugs should be monitored regularly. Also, healthcare professionals should incorporate metabolic syndrome assessment as part of the treatment and management plan for patients receiving ART and HAART. Finally, further prospective studies are needed to determine exactly when the metabolic syndrome starts.