Blood glucose level and serum lipid profiles among people living with HIV on dolutegravir-based versus efavirenz-based cART; a comparative cross-sectional study

Abstract

Background

Antiretroviral therapy-linked metabolic abnormalities have become a growing concern among people living with HIV. There is limited data regarding the effects of dolutegravir-based treatment on blood glucose levels and serum lipid profiles in people living with HIV in Ethiopia. Thus, this study aimed to assess blood glucose levels and serum lipid profiles among people living with HIV on dolutegravir-based versus efavirenz-based therapy.

Method and materials

An institutional-based comparative cross-sectional study was conducted from 30 June 2021 to 30 August 2021. A total of 128 participants (64 in the dolutegravir-based group and 64 in the efavirenz-based group) were enrolled in the study. The Chi-square, independent t-test, Mann–Whitney U-test, and logistic regression were used as appropriate statistical tests using SPSS Version 26.0 for this study. A p-value of <0.05 was considered statistically significant.

Result

The prevalence of hyperglycemia and dyslipidemia were 17.2% (11/64) and 79.7% (51/64) in the dolutegravir group, and 9.4% (6/64) and 75% (48/64) in the efavirenz group, respectively. The efavirenz group had significantly higher mean values of total cholesterol (190.73 ± 44.13 vs. 175.27 ± 37.67 mg/dl, p = 0.035) and high-density lipoprotein (47.53 ± 14.25 vs. 40.92 ± 13.17 mg/dl, p = 0.007) than the dolutegravir group. For a Kg/m² increase in BMI and for each month’s increase in the duration of HIV, the patients were 66% (AOR = 1.66, 95% CI: 1.13, 2.44), and 13% (AOR = 1.13, 95% CI: 1.03, 1.23) more likely to have hyperglycemia, respectively. In contrast, female patients were 3.04 times more likely to have dyslipidemia (AOR = 3.03, 95% CI: 1.14, 8.05) as compared to male patients, and with an increase in CD4 cell count of 1 cell/mm³, the odds of dyslipidemia increased by 0.3% (AOR = 1.003, 95% CI: 1.001, 1.006).

Conclusion

Efavirenz-based therapy resulted in higher mean values of total cholesterol and high-density lipoprotein as compared with dolutegravir-based therapy. It is important to consider and evaluate high-density lipoprotein levels in HIV patients on dolutegravir-based therapy, and total cholesterol levels in people living with HIV on efavirenz-based therapy.

KEY MESSAGES

• The long-term use of ART is thought to be one of the potential causes of metabolic abnormalities such as dysregulation of glucose and lipid metabolism.

• The burden of DTG-based cART-related metabolic abnormalities in resource-limited settings has not been well characterized.

• This study aimed to address these gaps by assessing blood glucose levels and serum lipid profiles among people living with HIV on DTG-based versus EFV-based regimens and identifying factors associated with hyperglycemia and dyslipidemia.

Keywords:

• Blood glucose levels
• serum lipid profiles
• HIV
• dolutegravir
• efavirenz

Introduction

People living with HIV on antiretroviral drugs have been associated with a number of metabolic and anthropometric abnormalities, such as dyslipidemia, lipodystrophy, and insulin resistance, which may increase the risk of cardiovascular disease (CVD) [1,2]. To minimize the impact of this disease, the treatment guidelines are being updated [3]. The World Health Organization (WHO) recommends first-line cART, which consists of a core agent [integrase astrand inhibitors (INSIs), protease inhibitors (PIs), or non-nucleoside reverse transcriptase inhibitors (NNRTIs) in combination with two nucleoside/nucleotide reverse transcriptase inhibitors (NRTIs)] for people living with HIV who have not received any prior treatment [4].

Efavirenz (EFV) is mostly used in NNRTI combination therapy, but resistance, adverse drug events, and drug interactions have been reported in most patients treated with this combination [5,6]. As a result, dolutegravir (DTG)-based NNRTI regimens are becoming the preferred first-line treatment [7]. According to the updated WHO 2019 guidelines, a DTG-based regimen (50 mg) is the preferred first-line antiretroviral treatment, with an EFV-based regimen (400 mg) as an alternative [8]. This is also the case for Ethiopia; based on the current ART guideline, DTG is recommended as a first-line antiretroviral medication [6]. This change to DTG is a better option among ART for people living with HIV, whether they have already been treated or not, due to its safety, efficacy, high genetic barrier to resistance, and availability as a single pill regimen in some countries [9].

Protease inhibitors and, to a lesser extent, NRTIs and NNRTIs are known to cause abnormal glucose and lipid metabolism, resulting in hyperglycemia, dyslipidemia, and insulin resistance [10,11]. Recently, there have been several studies that show deranged glucose metabolism with DTG-based treatment [12–14].

Abnormal lipid levels often occurs in people living with HIV who receive ART [15]. Among ART medications, EFV-based cART has been associated with the development of dyslipidemia, specifically increases in total cholesterol/high-density lipoprotein cholesterol (TC/HDL-C) ratio, low-density lipoprotein cholesterol (LDL-C), and triglycerides (TG) [16]. These profiles are mechanistically linked with insulin resistance and fat redistribution syndromes [17,18]. Indeed, studies have shown that switching from EFV to DTG improves lipid profiles and potentially reduces the risk of cardiovascular disease (CVD) [19,20]. However, a recent study showed that DTG is associated with lower HDL-C values, and this could increase the risk for CVD [21].

Susceptibility to the development of these metabolic defects varies with the individual and may be influenced by genetic differences [22]. Genome-wide association studies as well as target gene studies have recognized single nucleotide polymorphisms that could account for variations in blood glucose and lipid levels [23–25]. In addition, gender, age, body mass index, diet, duration of HIV infection, duration of ART, and treatment regimen type, among others, have been shown to be responsible for the high burden of metabolic abnormalities in people living with HIV [26–29]. Identification and modification of these risk factors have significant benefits for people living with HIV on ART. Despite these facts, the burden of ART-related metabolic abnormalities in resource-limited settings has not been well characterized. In addition, information on the burden and risk factors associated with blood glucose and lipid profile abnormalities among people living with HIV on DTG and EFV-based cART in Sub-Saharan African countries is scarce [22].

The long-term use of ART is thought to be one of the potential causes of metabolic abnormalities such as dysregulation of glucose and lipid metabolism. As a result, these ART-related metabolic abnormalities have been associated with a higher risk of death from CVD. However, the detection of blood glucose and lipid profiles among people living with HIV treated with DTG-based cART has not yet been investigated in Ethiopia or in the study area, p in particular. Therefore, the objective of this study was to address the gaps by assessing blood glucose levels and serum lipid profiles among people living with HIV on DTG-based versus EFV-based regimens, identifying factors associated with hyperglycemia and dyslipidemia in the study area, and giving recommendations to reduce metabolic complications.

Methods and materials

Study design, setting, population

An institutional-based comparative cross-sectional study was conducted among people living with HIV on DTG-based and EFV-based cART who were attending Dessie Comprehensive Specialized Hospital (DCSH) between 30 June 2021 and 30 August 2021. The hospital has been serving more than 6350 people living with HIV. In addition to the general services, the hospital provides HIV/AIDS interventions including diagnosis, treatment and monitoring. We included all people living with HIV aged 18 years or older who were either on DTG or EFV-based regimens (along with NNRTI or NRTI) for more than six months. Patients with mental health problems requiring urgent emergency treatment, pregnant and lactating women, patients taking anti-tuberculosis and lipid-altering drugs (corticosteroids, antihyperlipidemics), and known diabetes mellitus, renal disease, thyroid disease, hypertension, and chronic liver disease in both groups (the DTG- and EFV-based groups) were excluded from this study.

Sample size determination and sampling technique

Using the G*Power version 3.1 software, the sample size for the study was calculated by choosing a t-test and taking into account the following parameters: alpha = 0.05, power (1-Beta) =0.8 (80%), DTG to EFV-based regimens ratio of 1:1, and effect size (d) = 0.5. A total of 128 participants (64 in the DTG-based group and 64 in the EFV-based group) were enrolled in the study. In this study, a purposive sampling method was applied to recruit the study participants.

Data collection procedures

The data was collected using a questionnaire adapted from the WHO STEP-wise approach to chronic disease risk factor surveillance [20] as a data collection tool. Following the study participants’ informed consent, we collected socio-demographic data such as age, sex, place of residence, marital status, level of education, and occupational status. Besides, behavioral factors such as consumption of fruits and vegetables and physical activity were collected directly from participants using a questionnaire. Using the checklist, clinical data including viral load, CD4 cell count, WHO clinical stage of AIDS, current ART regimen type, duration of treatment, and duration of HIV were also collected from the patient’s medical record.

Furthermore, data on anthropometric parameters (weight, height, waist circumference (WC), and hip circumference (HC)) and blood pressure (BP) were collected from the study participants. The weight and height readings of the study participants were taken to the nearest 100 g and 1 mm, respectively, using a digital balance with an attached height measurement. WC and HC of the participants were measured to the nearest 0.1 cm using a flexible inelastic tape and waist-to-hip circumference ratio (WHR) was determined by dividing the WC by the HC in centimeters. Moreover, blood pressure was taken by nurse professionals using a mercury sphygmomanometer. After the participant had rested for at least 5 min, the blood pressure was taken twice from the left arm at 5-min intervals. The mean of the two measurements was estimated and used in the analysis. All these data collection and measurements were made by clinical nurses with strict supervision of the investigators. Finally, about 5 ml of the blood sample was taken from each study subject after an overnight fasting by a laboratory technologist. After the separation of serum from the whole blood, FBS and serum lipid profiles such as serum TC, TG, and HDL-C were analyzed at the central laboratory of DCSH Following standard protocol, a clinical chemistry analyzer, the Siemens Dimension EXL 200 System, was utilized.

Hyperglycemia and Dyslipidemia resulting from ART are the main outcome variables assessed in this study.

Operational definitions

Dyslipidemia was defined based on the National Cholesterol Education Program Adult Treatment Panel III guideline as LDL-C ≥ 130 mg/dl, TC ≥200 mg/dl, TG ≥150 mg/dl, and HDL-C < 40 mg/dl for men, <50 mg/dl for women, occurring in isolation or combination. TC/HDL-C ratio was considered as raised if it was ≥5 [23].

Hyperglycemia was defined as FBS ≥110 mg/dl [23]. Grade 1 hyperglycemia (110–125 mg/dL); Grade 2 hyperglycemia (>125–250 mg/dL); Grade 3 hyperglycemia (>250–500 mg/dL), and grade 4 hyperglycemia (≥500 mg/dL) [24].

Hypertension was defined as systolic blood pressure ≥140 mmHg and/or diastolic blood pressure ≥90 mmHg [25].

BMI classified as underweight (BMI <18.5 kg/m²), normal weight (18.5–24.9 kg/m²), overweight (25–29.9 kg/m²), and obese (≥30 kg/m²) [26].

Waist circumference: cut-off point for females >80 cm and >94 cm for males [27].

Waist to hip ratio: cut-off point for females ≥0.85 and ≥0.9 for males [27].

Regarding physical activity: Activities that require a lot of physical effort and significantly increase breathing or heart rate for at least 10 min straight, including running, carrying or lifting heavy loads, digging, and construction work, are considered vigorous-intensity activities [28]. Activities of moderate intensity result in slight increases in breathing or heart rate for at least ten uninterrupted minutes [28].

Sufficient physical exercise: Adults should engage in at least 150–300 min of moderate-intensity aerobic exercise per week, 75–150 min of high-intensity aerobic exercise, or the equivalent amount of vigorous exercise [29].

Low fruit and vegetable intake; defined as consuming less than five servings of fruit and vegetables per day [30]. For raw green leafy vegetables, 1 serving = one cup; for cooked or chopped vegetables, 1 serving = ½ cup; for fruit (banana, orange etc…), 1 serving = 1 medium size piece; for chopped, cooked and canned fruit, 1 serving = ½ cup; and for juice from fruit, 1 serving = ½ cup [31].

Statistical data analysis

The collected data were checked and entered into Epi data version 4.6, and then exported to SPSS version 26.0 program for analysis. Descriptive analysis was carried out and results were presented using tables. The categorical variables were explored using frequency and percentage, and the continuous variables were expressed as mean ± standard deviation (SD) and median (interquartile range (IQR)). Using the chi-square test, categorical variables were compared. The independent t-test was performed to assess differences between groups on continuous variables that showed normality, while Mann–Whitney U-test was performed on continuous variables that did not show normality. To assess the normality of the data distribution, the Shapiro–Wilk test was used. To investigate the relationships between outcome variables and potential associated factors, bivariable and multivariable binary logistic regression models were used. Variables with p-values <0.25 in bivariable logistic regression were fitted into the multivariable logistic regression model for the final analysis [32]. Crude odds ratio (COR) and adjusted odds ratio (AOR) with 95% confidence intervals (CI) were reported. A p-value of <0.05 was considered statistically significant. Multicollinearity among selected independent variables was checked, and the variance inflation factor was found to be acceptable (less than 5). The Hosmer and Lemeshow test was used to determine the goodness of fit of the model.

Ethics approval and consent to participate

Ethical approval and clearance were obtained by the ethical committee of the School of Medicine, College of Medicine and Health Sciences, the University of Gondar, with a protocol number of Ref: 666/06/2021. An official support letter was obtained with protocol number Ref: Bioc-90/09/2013 from the Department of Biochemistry, School of Medicine, University of Gondar, and submitted to DCSH. After discussing the purpose and method of the study, written permission was obtained from DCSH before data collection. The study participants were informed about the study before collecting any data and samples. Accordingly, written informed consent was obtained from the participants who were able to read/write while verbal informed consent was obtained from the participants who could not read/write. Participants have a full right to continue or withdraw from the study. The data were kept in codes instead of any personal identifiers and the confidentiality of all participants was maintained throughout the study.

Results

Socio‑demographic and behavioral characteristics of participants

A total of 128 (64 DTG and 64 EFV-based) cART treated HIV positive people participated in this study. The mean age for EFV and DTG-treated patients was 42.39 ± 10.91 and 45.42 ± 12.46 years old, respectively. 64.1% (41/64) of the DTG and 59.4% (38/64) of the EFV-treated groups were female. About 54.7% (35/64) of DTG and 57.8% (37/64) of EFV-treated patients were married. 65.6% (42/64) of DTG and 62.5% (40/64) of EFV-treated patients were living in an urban area Whereas about 22 (34.4%) of DTG and 42.2% (27/64) of EFV-treated patients had insufficient physical activity, 81.3% (52/64) of DTG and 82.8% (53/64) of EFV-treated patients had low fruit and vegetable intake per day (Table 1).

Table 1. Socio-demographic and behavioral characteristics of study participants among people living with HIV on DTG-based versus EFV-based cART at DCSH, Northeast Ethiopia, 2021 (n = 128).

Variables	Category	DTG-based (n = 64) N (%)	EFV-based (n = 64) N (%)	p-Value
Age (Mean ± SD)		45.42 ± 12.46	42.39 ± 10.91	0.146
Gender	Male	23 (35.9)	26 (40.6)	0.585
	Female	41 (64.1)	38 (59.4)
Residence	Urban	42 (65.6)	40 (62.5)	0.713
	Rural	22 (34.4)	24 (37.5)
Education	Can’t read and write	19 (29.7)	15 (23.4)	0.227
	Read and write only	10 (15.6)	7 (10.9)
	Primary school	16 (25)	28 (43.8)
	Secondary school	11 (17.2)	10 (15.6)
	College and above	8 (12.5)	4 (6.3)
Marital status	Single	5 (7.8)	9 (14.1)	0.478
	Married	35 (54.7)	37 (57.8)
	Divorced	15 (23.4)	13 (20.3)
	Widow/widower	9 (14.1)	5 (7.8)
Occupation	Farmer	8 (12.5)	12 (18.8)	0.841
	Merchant	10 (15.6)	9 (14.1)
	Employee	13 (20.3)	8 (12.5)
	Labor	10 (15.6)	13 (20.3)
	Student	2 (3.1)	3 (4.7)
	Homemaker	13 (20.3)	12 (18.8)
	Other	8 (12.5)	7 (10.9)
Level of physical	Insufficient	22 (34.4)	27 (42.2)	0.363
Activity	Recommended	42 (65.6)	37 (57.8)
Fruit and vegetable	Low	52 (81.3)	53 (82.8)	0.818
intake per day	Recommended	12 (18.8)	11 (17.2)
No. of days consuming fruit	≤4 Days	48 (75)	46 (71.9)	0.689
and vegetable per week	>4 Days	16 (25)	18 (28.1)

Notes: Continuous variables are reported as mean (SD), categorical variables are reported as number and percentage; Other, refers to driver, pensioners.

Abbreviations: DTG: dolutegravir; EFV: efavirenz; SD: standard deviation; No.: number.

Clinical and anthropometric related characteristics of participants

The clinical features obtained from the patient’s medical records showed that the median (IQR) CD4 cell count was 411.5 (IQR = 613.25–326.25) in the DTG and 422.5 (IQR = 575.25–319.75) in the EFV-treated patients. Concerning the viral load count, 7.8% (5/64) of DTG and 18.8% (12/64) of EFV-treated patients had a viral load of more than 1000. Of the total proportion of study participants, 4.7% (3/64) of DTG and 9.4% (6/64) of EFV-treated patients were classified as clinical WHO Stage III. The EFV-treated patients had a longer duration of HIV infection and ART use than the DTG-treated patients. About 28.1% (18/64) of DTG-treated patients and 15.6% (10/64) of EFV-treated patients were either overweight or obese. Regarding the regional distribution of fat, 34.4% (22/64) of DTG and 25% (16/64) of EFV-treated patients had WC above the cut-off value; in addition, about 54.7% (35/64) of DTG and 43.8% (28/64) of EFV-treated patients had elevated WHR, which is a better indicator of regional fat distribution. Moreover, 7.8% (5/64) of DTG and 10.9% (7/64) of EFV-treated patients were hypertensive (Table 2).

Table 2. Clinical and anthropometric related characteristics of study participants among people living with HIV on DTG based versus EFV-based cART at DCSH, Northeast Ethiopia, 2021 (n = 128).

Variables	DTG-based (n = 64) N (%)	EFV-based (n = 64) N (%)	p-Value
Duration of HIV infection in	21.63 ± 8.32	31.69 ± 7.98	<0.001*
months mean±(SD)
Duration of taking ART in	12 ± 2.51	22.19 ± 4.45	<0.001*
months mean±(SD)
WHO clinical stage of AIDS
Stage I	48 (75)	42 (65.6)	0.425
Stage II	13 (20.3)	16 (25)
Stage III	3 (4.7)	6 (9.4)
CD4 cell count (cells/mm^3)	411.5 (613.25-326.25)	422.5 (575.25-319.75)	0.502
median (IQR)
Viral load count (copies/ml)
Undetectable	44 (68.8)	32 (50)	0.064
20–1000	15 (23.4)	20 (31.3)
≥1000	5 (7.8)	12 (18.8)
Body mass index (Kg/m^2)
Underweight	9 (14.1)	12 (18.8)	0.355
Normal	37 (57.8)	42 (65.6)
Overweight	15 (23.4)	9 (14.1)
Obese	3 (4.7)	1 (1.6)
Waist circumference (cm)
High	22 (34.4))	16 (25)	0.246
Normal	42 (65.6)	48 (75)
Waist-to-hip ratio
High	35 (54.7))	28 (43.8)	0.216
Normal	29 (45.3)	36 (56.3)
Blood pressure (mm/Hg)
Normotensive	59 (92.2)	57 (89.1)	0.544
Hypertensive	5 (7.8)	7 (10.9)

Notes: Continuous variables are reported as mean (SD) and median (IQR), and categorical variables are reported as numbers and percentages.

Abbreviations: HIV: human immunodeficiency virus; ART: antiretroviral therapy; AIDS: acquired immune deficiency syndrome; DTG: dolutegravir; EFV: efavirenz; SD: standard deviation; IQR: interquartile range; *significant difference at p < 0.05.

Blood glucose levels of study participants

Overall, 13.28% (17/128) of the study participants were found to have hyperglycemia. The prevalence of hyperglycemia was 17.2% (11/64) in the DTG-treated patients and 9.4% (6/64) in the EFV-treated patients. Grade 2 hyperglycemia was found in 6.3% (4/64) of DTG and 3.1% (2/64) of EFV-treated patients. Grade 1 hyperglycemia was found in 10.9% (7/64) of DTG and 6.3% (4/64) of EFV-treated patients. The independent t-test revealed a statistically insignificant difference between the mean value of FBS in both groups (p > 0.05) (Table 3).

Table 3. Comparison of fasting blood glucose and serum lipid profiles using independent t-test/Mann–Whitney U-test between patients treated with DTG and EFV at DCSH, Northeast Ethiopia, 2021 (n = 128).

Parameters	Study group	Mean ± SD/Median (IQR)	Mean difference	95% CI	p-Value
FBS (mg/dl)	EFV-based	81.55 ± 20.91	−4.41	−12.03, 3.21	0.225
	DTG-based	85.95 ± 22.62
TC (mg/dl)	EFV-based	190.73 ± 44.13	15.47	1.12, 29.82	0.035*
	DTG-based	175.27 ± 37.67
HDL-C (mg/dl)	EFV-based	47.53 ± 14.25	6.61	1.81, 11.41	0.007*
	DTG-based	40.92 ± 13.17
LDL-C (mg/dl)	EFV-based	110.34 ± 36.56	7.38	−4.19, 18.94	0.209
	DTG-based	102.97 ± 29.13
TG (mg/dl)^a	EFV-based	152.5 (189.5–120)	–	–	0.488
	DTG-based	145 (179.5–119.25)
TC/HDL-C ratio^a	EFV-based	4.01 (4.9–3.28)	–	–	0.098
	DTG-based	4.5 (5.17–3.63)

FBS: fasting blood sugar; TC: total cholesterol; HDL-C: high-density lipoprotein cholesterol; LDL-C: low-density lipoprotein cholesterol; TG: triglyceride; DTG: dolutegravir; EFV: efavirenz; SD: standard deviation; IQR: interquartile range; ^aMann–Whitney U test was used; *significant difference at p < 0.05.

Serum lipid levels of study participants

Among the study participants involved, 77.34% (99/128) had dyslipidemia, of which 79.7% (51/64) were DTG-treated patients and 75% (48/64) were EFV-treated patients. The number of patients who had raised TC, low HDL-C, raised TG, raised LDL-C and raised TC/HDL-C was 26.6% (17/64), 64.1% (41/64), 42.2% (27/64), 20.3% (13/64) and 31.3% (20/64) among the DTG-treated group, and 37.5% (24/64), 39.1% (25/64), 50% (32/64), 31.3% (20/64) and 21.9% (14/64) among the EFV-treated group, respectively. The independent t-test revealed a significant difference in the mean values of TC and HDL-C between the DTG-treated group and the EFV-treated group. The mean value of HDL-C was reduced in DTG-treated patients as compared to EFV-treated patients (p = 0.007). Besides, the mean value of TC was elevated in EFV-treated patients as compared to DTG-treated patients (p = 0.035). However, there were statistically insignificant differences between the mean value of LDL-C, and the median values of TG and TC/HDL-C ratio in both groups (Table 3).

Factors associated with hyperglycemia in people living with HIV on ART

A multivariable logistic regression analysis was carried out and two variables namely; duration of HIV infection and BMI were significantly associated with hyperglycemia at a p-value <0.05. Accordingly, the duration of HIV infection was significantly associated with hyperglycemia in people living with HIV on ART. For each month’s increase in the duration of HIV infection, the patients were 13% more likely to have hyperglycemia (AOR = 1.13, 95% CI: 1.03, 1.23). In addition, BMI was significantly associated with hyperglycemia in people living with HIV on ART. For a Kg/m² increase in BMI, the patients were 66% more likely to have hyperglycemia (AOR = 1.66, 95% CI: 1.13, 2.44) (Table 4).

Table 4. Factors associated with hyperglycemia in people living with HIV on ART (n = 128).

	Hyperglycemia
Variables	No	Yes	COR (95% CI)	p-Value	AOR (95% CI)	p-Value
Age^#			1 (0.96–1.04)	0.922	0.99 (0.94–1.05)	0.78
Sex^#
Male	41	8	1		1
Female	70	9	0.66 (0.24–1.84)	0.43	0.279 (0.05–1.46)	0.13
Place of residence
Urban	74	8	0.44 (0.16–1.25)	0.123	0.6 (0.16–2.238)	0.446
Rural	37	9	1		1
Physical activity
Insufficient	40	9	2 (0.71–5.58)	0.187	1.69 (0.43–6.58)	0.45
Recommended	71	8	1		1
Duration of HIV infection (months)			1.04 (0.99–1.10)	0.113	1.13 (1.03–1.23)	0.006*
Duration of taking ART (months)			0.94 (0.86–1.04)	0.22	0.85 (0.67–1.08)	0.176
Current regimen type
EFV	58	6	1		1
DTG	53	11	2.01 (0.69–5.80)	0.199	1.42 (0.15–13.75)	0.76
BMI (Kg/m^2)			1.27 (1.10–1.48)	0.002	1.66 (1.13–2.44)	0.009*
WC (cm)			1.06 (1.01–1.12)	0.029	0.88 (0.76–1.02)	0.08
Systolic BP (mm/Hg)			1.03 (0.99–1.06)	0.144	1.01 (0.97–1.06)	0.52

COR: crude odd ratio; AOR: adjusted odds ratio; CI: confidence interval; HIV: human immunodeficiency virus; ART: antiretroviral therapy; BMI: body mass index; WC: waist circumference; BP: blood pressure; DTG: dolutegravir; EFV: efavirenz; ^#included in the multivariable logistic analysis by clinical justification; *p value < 0.05; Hosmer Lemeshow test p value = 0.541.

Factors associated with dyslipidemia in HIV-infected patients on ART

In multivariable logistic regression analysis, sex and CD4 cell count were significantly associated with dyslipidemia in people living with HIV on ART. Female patients were 3.03 times more likely to have dyslipidemia (AOR = 3.03, 95% CI: 1.14, 8.05) as compared to male patients. In addition, CD4 cell count was significantly associated with dyslipidemia in people living with HIV on ART. An increase in CD4 cell count by 1 cell/mm³, the odds of dyslipidemia increase by 0.3% (AOR = 1.003, 95% CI: 1.001, 1.006) (Table 5).

Table 5. Association of independent variables with dyslipidemia in people living with HIV on ART at DCSH, Northeast Ethiopia, 2021 (n = 128).

	Dyslipidemia
Variables	No	Yes	COR (95% CI)	p-Value	AOR (95% CI)	p-Value
Age			1.03 (0.99–1.07)	0.115	1.029 (0.99–1.07)	0.165
Gender
Male	17	32	1		1
Female	12	67	2.97 (1.27–6.94)	0.012	3.03 (1.14–8.05)	0.026*
Place of residence
Urban	16	66	1.63 (0.7–3.77)	0.249	1.23 (0.47–3.24)	0.68
Rural	13	33	1		1
BMI (Kg/m^2)			1.132 (0.99–1.294)	0.071	1.13 (0.97–1.31)	0.114
Current regimen type^#
EFV	16	48	1		1
DTG	13	51	1.31 (0.57–3)	0.527	1.06 (0.20–5.59)	0.948
Duration of taking ART (months)^#			0.997 (0.933–1.066)	0.938	1.02 (0.89–1.17)	0.788
Fruit and vegetable intake per day
Low	21	84	2.13 (0.8–5.70)	0.131	2.07 (0.67–6.44)	0.208
Recommended	8	15	1		1
No. of days consuming fruit and vegetable per week
≤4 days	18	76	2.02 (0.84–4.88)	0.119	1.45 (0.53–3.99)	0.469
>4 days	11	23	1		1
CD4 cell count (cell/mm^3)			1.004 (1.001–1.006)	0.007	1.003 (1.001–1.006)	0.016*

COR: crude odd ratio; AOR: adjusted odds ratio; CI: confidence interval; BMI: body mass index; ART: antiretroviral therapy; CD4: cluster of differentiation 4; No.: number; DTG: dolutegravir; EFV: efavirenz; *p value < 0.05; #included in the multivariable logistic analysis by clinical justification; Hosmer Lemeshow test p value = 0.265.

Discussion

This study was designed to assess blood glucose levels and serum lipid profiles among people living with HIV on dolutegravir-based versus efavirenz-based cART and identify factors associated with hyperglycemia and dyslipidemia. This is the first study in Ethiopia, given that the drug is relatively new for HIV management in Ethiopia.

The prevalence of dyslipidemia was found to be 79.7% (51/64) in DTG and 75% (48/64) in EFV-treated patients, which is higher than the study conducted in Cameroon [33]. The mean TC level in people living with HIV who received EFV (190.73 ± 44.13 mg/dl) was higher than in people living with HIV who received DTG (175.27 ± 37.67 mg/dl) (p = 0.035). The present study revealed that the proportion of raised TC level was 37.5% in EFV and 26.6% in DTG-treated patients. Similar to our findings, the SINGLE and START clinical studies showed that EFV-treated patients had higher mean TC level increases compared with DTG-treated patients [14,34]. In addition, the ADVANCE clinical study also revealed that EFV-treated patients had a higher increase in median differences in serum TC levels as compared to DTG-treated patients [35]. In contrast to our finding, a study from Cameroon showed that there was a statistically insignificant difference between the mean serum TC levels in both groups. In addition, the proportion of hypercholesterolemia was higher in the DTG (18.2%) users compared to the EFV (9.1%) users [33]. The observed variation could be due to differences in the duration of treatment and sample size, which mean the study from Cameroon was conducted with a smaller sample size than the current study and was done among participants who had been on ART for a longer period of time, with the median duration of treatment being 144 weeks (144–165) [33]. In addition, EFV can efficiently activate a unique liver protein called pregnane X-receptor (PXR) and stimulate its target gene expression that mediates lipid uptake and cholesterol synthesis, thereby causing hypercholesterolemia. In contrast, INSTI didn’t activate PXR [31].

The study also showed that the mean serum HDL-C level in people living with HIV who received DTG was lower (40.92 ± 13.17 mg/dl) than in people living with HIV who received EFV (47.53 ± 14.25 mg/dl) (p = 0.007). Low HDL-C levels were present in 64.1% of the DTG patients compared to 39.1% of the EFV patients. The START clinical study found that patients receiving EFV-based cART experienced greater increases in HDL-C levels than those receiving INSTI-based cART, which is consistent with our findings [27]. Furthermore, the ADVANCE clinical study also revealed that EFV-treated patients had a higher increase in median differences in HDL-C levels as compared to DTG-treated patients [28]. Moreover, a study from Zambia found that patients receiving DTG-based cART had a low proportion of HDL-C, which was significantly higher than that of patients receiving NNRTIs (33%) [18]. In contrast to our study, a comparative analysis of phase IIb-IIIb clinical studies found that DTG had a safer lipid profile (HDL-C, LDL-C, and TC) in combination with NRTIs at 48 weeks [36]. Similarly, a study conducted in Italy showed that switching patients to a DTG-based regimen improves lipid profiles after 48 weeks of treatment [20]. The inconsistency in results could be attributed to racial disparities as the previous studies were predominantly among whites, and behavioral factors such as diet could be implicated in the observed differences [30]. Hence there is need to have geographical-specific data on various ART-regimens. Furthermore, available data indicates that long-term EFV therapy and its concentration are directly proportional to HDL-C levels [37]. The possible molecular mechanism by which EFV increases HDL-C levels was through the down-regulation of plasma cholesterol ester transfer protein expression through transcription factor [38].

In addition, this study revealed that there were statistically insignificant differences in the mean values of LDL-C and median values of TG between two groups. This is supported by the result of the previous study [39]. Furthermore, a statistically insignificant difference between the two groups was also observed in the median TC/HDL-C ratios. This is corroborated by the results of SINGLE and START clinical studies [22,27].

The present study also compares blood glucose level among people living with HIV receiving DTG and EFV-based regimens. The prevalence of hyperglycemia was 17.2% (11/64) in DTG and 9.4% (6/64) in EFV-treated patients. This finding is higher compared with the SINGLE clinical study conducted in the USA, which observed the prevalence of 8.2% (34/414) and 4.8% (20/419) in DTG and EFV-treated patients, respectively [14,40]. In addition, it is also higher compared with the study done in Uganda; the study found that 0.47% of people in the case group (DTG-based cART) had new-onset hyperglycemia, compared to 0.03% in the control group [41]. The discrepancy in the result might be due to in clinical studies the participants may be withdrawn due to treatment side effects, and the prevalence of grade 1 hyperglycemia was not reported in a SINGLE clinical study. The difference could also be due to differences in lifestyle; and clinical, anthropometric, and metabolic-related factors [42,43]. This study revealed that there was a statistically insignificant difference in the mean values of FBS between the DTG and the EFV-treated group. This is supported by a study conducted in Cameroon [33].

In our study, we found dyslipidemia to be significantly associated with gender and CD4 cell count. Female patients were 3.04 times more likely to have dyslipidemia as compared to male patients. This finding is in line with a study done in Ethiopia [44]. This finding in women is probably due to menopause, which is associated with a decrease in circulating estrogen levels, which may increase the risk of dyslipidemia [45]. In addition, CD4 cell count was significantly associated with dyslipidemia in people living with HIV on ART. With an increase in CD4 cell count by l cell/mm³, the odds of dyslipidemia increase by 0.3%. Many studies also reported a similar finding [46,47]. This could be explained because ART suppresses viral load, resulting in an increase in CD4 cell count. This allows for immune recovery and health restoration, which leads to lipid elevation in addition to the ongoing effect of ART on lipid elevation [47,48].

The present study also revealed that BMI and the duration of HIV infection were significantly associated with dyslipidemia. For a kg/m² increase in BMI, the patients were 66% more likely to have hyperglycemia. This significant association was observed in other similar studies [49,50] in which high BMI was significantly associated with hyperglycemia in people living with HIV on ART. This is due to higher BMI being strongly related to insulin resistance and decreasing insulin-stimulated glucose disposal [51]. Non-esterified fatty acids, glycerol, cytokines, proinflammatory agents, and other substances that are involved in the development of insulin resistance are present in higher concentrations in obese people. Hyperglycemia develops as a result of insulin resistance and impaired beta-cell function [52]. Besides, the duration of HIV infection was significantly associated with hyperglycemia in people living with HIV on ART. For each month’s increase in the duration of HIV infection, the patients were 13% more likely to have hyperglycemia. This finding is in line with other studies [49,53]. This could be explained because of a significant association between the duration of HIV infection and insulin resistance [54]. Persistent HIV infection has been associated with insulin resistance through an increase in inflammatory chemokines involved in insulin regulation, followed by hyperglycemia [49,55].

This study has limitations mainly related to the nature of the cross-sectional study design; it might not describe the cause and effect relationship between the factors and the outcomes under study. In addition, the study did not incorporate a control group of HIV-negative individuals, which would have provided better insight into the role of HIV infection and antiretroviral therapies. Moreover, due to financial constraints, this study did not include hemoglobin A1C as a diagnostic tool for hyperglycemia.

Conclusion

The study found that EFV-treated patients had higher mean values of TC and HDL-C as compared with DTG-treated patients. But there were statistically insignificant differences in the mean values of FBS and LDL-C, as well as the median values of TG and the TC/HDL-C ratio in both groups. Therefore, it is important to consider and evaluate HDL levels in people living with HIV on DTG-based therapy, and TC levels in HIV patients on EFV-based therapy. The duration of HIV infection and BMI were significantly associated with hyperglycemia, whereas sex and CD4 cell count were significantly associated with dyslipidemia in people living with HIV on ART. There is a need for baseline and follow-up evaluation of blood glucose levels and serum lipid profiles of people living with HIV on ART (DTG and EFV). Moreover, further studies should be conducted with a larger sample size from different ART centers using a prospective cohort study design to investigate the effects of DTG and EFV on blood glucose levels and serum lipid profiles.

Author contribution

MJ and TA wrote the main manuscript text. TS, MT, and EC analyzed the data and reviewed the draft. All authors read and approved the final manuscript.

Acknowledgment

The authors are indebted to acknowledge data collectors, the University of Gondar and Dessie Comprehensive Specialized Hospital, and Debre Markos University for their unreserved contributions. No funding was received for this study.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Data availability statement

All the necessary materials can be found in the text. Due to the privacy policy, confidential data materials could be obtained from the corresponding author upon request.

Additional information

Funding

The author(s) reported there is no funding associated with the work featured in this article.