Sage Journals: Discover world-class research

Abstract

Objective

The total cholesterol/high-density lipoprotein cholesterol ratio is a predictor of nonalcoholic fatty liver disease—a hepatic manifestation of metabolic syndrome. This study investigated the association between total cholesterol/high-density lipoprotein cholesterol ratio and metabolic syndrome among people living with human immunodeficiency virus on dolutegravir-based antiretroviral therapy in southwestern Uganda.

Methods

We conducted a secondary analysis of data obtained from a cross-sectional study of 377 adults who had been on dolutegravir-based antiretroviral therapy for at least 1 year at Ruhoko Health Centre IV, southwestern Uganda.

Results

The median total cholesterol/high-density lipoprotein cholesterol ratio was significantly higher in individuals with metabolic syndrome (3.92) than in those without (2.96, p < 0.001). A statistically significant association was observed between high total cholesterol/high-density lipoprotein cholesterol ratio and metabolic syndrome (adjusted odds ratios: 2.06, 95% confidence interval: 1.08–3.91, p = 0.028). The total cholesterol/high-density lipoprotein cholesterol ratio at an optimal cutoff of 3.30 had a significant ability (area under the curve = 0.696, 95% confidence interval: 0.642–0.750) to differentiate participants with metabolic syndrome from those without at a sensitivity of 73% and specificity of 60%.

Conclusion

The total cholesterol/high-density lipoprotein cholesterol ratio is a significant predictor of metabolic syndrome and serves as a potential blood-based biomarker.

Keywords

Total cholesterol/high-density lipoprotein cholesterol metabolic syndrome blood-based biomarker predictor dolutegravir

Introduction

With the increasing availability of antiretroviral therapy (ART), human immunodeficiency virus (HIV) infection is no longer a death sentence but has become a manageable chronic illness.¹ However, patients on ART seem to exhibit an increased prevalence of noncommunicable disorders such as metabolic syndrome (MetS).² MetS is associated with an increased risk of cardiovascular disease (CVD), type 2 diabetes mellitus, and nonalcoholic fatty liver disease (NAFLD)^3–5 and is characterized by a cluster of metabolic disorders, including central obesity, dyslipidemia, and hypertension, along with insulin resistance.⁶

Increased use of dolutegravir (DTG)-based ART among HIV-infected patients seems to have a strong association with a higher prevalence of MetS and its components.⁷ Studies have documented the prevalence of MetS among HIV-infected populations on ART in sub-Saharan Africa to be within the range of 16.9%–50.3% based on the International Diabetes Federation (IDF) criteria.⁸ Notably, higher prevalence rates have been observed among individuals on DTG-based regimens.⁹ In Ethiopia, a cross-sectional study among people living with HIV (PLWH) on DTG-based ART reported a MetS prevalence of 25.6%.¹⁰ Similarly, a study conducted in northern Tanzania found a MetS prevalence of 42.3% in the PLWH population.¹¹ In southwestern Uganda, Nzaramba et al.¹² reported a prevalence of 35.3% among PLWH on DTG-based ART.

In PLWH, undiagnosed or untreated MetS can lead to significant clinical problems. Gupta et al.⁵ noted that uncontrolled MetS can increase mortality, cause NAFLD, and accelerate CVD. NAFLD, which is increasingly problematic as a component of MetS, is a major hepatic manifestation in HIV infection and is often associated with liver cirrhosis and fibrosis, complicating the clinical management of HIV-infected patients.¹³ Due to these risks, early detection of MetS is critical to avert irreversible complications, particularly in resource-limited settings such as Uganda.

An increasing number of researchers have highlighted the potential importance of using the total cholesterol/high-density lipoprotein cholesterol (TC/HDL-C) ratio as a simple, cost-effective biomarker for estimating MetS and other metabolic disorders.^14,15 Increased TC/HDL-C ratios are related to insulin resistance, cardiovascular dysregulation, and dyslipidemia, which are all indicative of MetS.¹⁶ In addition, studies from other parts of the world, including eastern parts of Africa, have highlighted the TC/HDL-C ratio’s predictive value for NAFLD, thereby reinforcing the notion of its role as a marker of metabolic derangement.^17,18

Uganda, similar to many sub-Saharan African countries, faces a dual burden of infectious and noncommunicable diseases.¹⁹ Southwestern Uganda, in particular, has one of the highest HIV prevalence rates in the country,²⁰ coupled with a rapidly expanding rollout of DTG-based ART.^21,22 However, despite these trends, there are limited local data assessing the relationship between TC/HDL-C ratio and MetS among this population. The region’s high HIV burden, increasing use of newer ART regimens, and constrained health resources make it an important setting for evaluating simple biomarkers that can aid in early MetS detection and improve long-term outcomes in PLWH.

Therefore, our study aimed to assess the association between the TC/HDL-C ratio and MetS and determine the optimal TC/HDL‑C cutoff for discriminating individuals with and without MetS among PLWH on DTG-based ART in southwestern Uganda.

Material and methods

The reporting of this study conforms to the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) guidelines for observational studies.²³

Study design, population, and variables

We conducted a secondary data analysis using a dataset generated from a cross-sectional study¹² that investigated the association between low aspartate aminotransferase (AST)/alanine aminotransferase (ALT) ratio and MetS among 377 PLWH on DTG-based ART in southwestern Uganda. The secondary study was conducted between 20 January 2025 and 20 March 2025. In the primary study, patients aged 18 years and above who had been on DTG-based ART for at least 12 months and had provided informed consent were systematically enrolled. Individuals with missing clinical records, pregnant women, those on interrupted treatment, and those who visited due to an acute illness, including hepatitis, were excluded. Those who were on diabetes treatment, lipid-lowering medication, corticosteroids, or oral contraceptive pills were also excluded from the study. This primary study explored many independent variables, including anthropometric measurements (waist circumference (WC), hip circumference, neck circumference, mid-upper arm circumference (MUAC), weight, and height), liver enzymes (AST, ALT, gamma-glutamyl transferase (GGT), alkaline phosphatase (ALP)), and lipid profile (TC, triglycerides (TG), low-density lipoprotein (LDL)-cholesterol (LDL-C), and HDL-C)—some of which were used to calculate the TC/HDL-C ratio in our secondary study.

In the secondary analysis, the primary dependent variable of interest was MetS, which was defined based on the National Cholesterol Education Programme Adult Treatment Panel III (NCEP ATP III) criteria as fulfilling three or more of the following conditions: (a) WC >102 cm for men and >88 cm for women; (b) TG level ≥150 mg/dL; (c) HDL-C level <40 mg/dL for men and <50 mg/dL for women; (d) blood pressure (BP) >130/85 mmHg or hypertension treatment; and (e) fasting plasma glucose (FPG) level ≥110 mg/dL or being on diabetes treatment.²⁴ However, the TC/HDL-C ratio was the main independent variable in the secondary analysis.

Other independent variables considered in this secondary analysis included sociodemographic variables (such as sex, age, marital status, educational level, residence, religion, employment status, and having a ventilated kitchen), lifestyle factors (such as smoking status, alcohol consumption, and vegetable and fruit intake), medical history (reported comorbidities such as hypertension, family history of dementia, and other chronic conditions such as hypertension, kidney disease, CVDs, and dyslipidemias), biochemical parameters (such as FPG; TC; TG; LDL; liver enzymes such as serum AST, ALT, ALP, and GGT; and serum electrolytes such as sodium, potassium, and chloride), sleep-related factors (such as poor sleep quality, obstructive sleep apnea (OSA), and sleep duration); anthropometric parameters (such as BP, body mass index (BMI), and WC), and ART-related variables (such as DTG-based ART duration).

The TC/HDL-C ratio was calculated by dividing the TC levels in mg/dL by the HDL-C levels in mg/dL, and a high ratio was defined as ≥5 according to the NCEP ATP III criteria.^14,17,25,26 OSA is a condition characterized by recurrent episodes of respiratory pause with durations greater than or equal to 10 s. OSA was evaluated using the Stop Bang scoring model tool.²⁷ Sleep quality is defined as an individual’s self-satisfaction with all aspects of sleep experience and was determined by the Pittsburgh Sleep Quality Index (PSQI), which encompasses factors such as prolonged sleep latency, frequent awakenings, insufficient sleep duration, daytime sleepiness, and overall dissatisfaction with sleep. A good sleep quality is defined as a global PSQI score ≤5, and a poor sleep quality is defined as a global PSQI score ≥6.²⁸ Sleep duration was categorized as <5, 5–6, 6–7, and >7 h.²⁹ BMI, which is the ratio of weight to the square of height in kg/m², was categorized as <25, 25–29.9 (overweight), and ≥30 (obesity).³⁰ Low fruit and vegetable intake was defined as consuming less than five servings of fruit and vegetables per day.³¹

Sample size determination

For the secondary analysis, a sample size calculation was based on the prevalence (35.3%) of MetS¹² among PLWH on DTG-based ART in a study conducted in southwestern Uganda. We used the Kish Leslie formula (1965) with a 5% margin of error and a 95% confidence interval (CI).

N = 1 . 96^{2} \times 0.353 (1 - 0.353) / 0.0 5^{2} = 351

Since the primary study enrolled 377 participants, this sample size was adequate for the secondary data analysis.

Statistical analysis

Data were analyzed using STATA version 17. Categorical variables were summarized using frequencies and proportions. Median (interquartile range (IQR)) was utilized to summarize the continuous variable “age” as it was not normally distributed across the study participants. The median TC/HDL-C ratio was compared between participants with and without MetS using Mann–Whitney U test. The prevalence of MetS was also compared between participants with a high TC/HDL-C ratio and those with a low TC/HDL-C ratio using chi-square test. A p-value <0.05 was considered to indicate statistical significance.

Logistic regression was used to assess the association between the TC/HDL-C ratio and MetS. MetS was the binary dependent variable (1 = present, 0 = absent). At the bivariate level, crude associations between MetS and each independent variable, including the TC/HDL-C ratio, were evaluated. Associations were quantified using crude odds ratios (cORs), with corresponding 95% CIs, and statistical significance was determined at a p-value <0.05. The primary exposure and other variables that were clinically and/or statistically significant at this level were included in the multivariable model to adjust for potential confounding. Adjusted ORs (aORs) with 95% CIs and p values were reported. The model’s goodness of fit was assessed using the Hosmer–Lemeshow test, with a p-value >0.05 indicating a good fit (p = 0.6617 in our final model). In the final multivariable model, associations were considered significant at a p value <0.05.

To evaluate the predictive ability of the TC/HDL-C ratio for identifying participants with MetS, we conducted a receiver operating characteristic (ROC) curve analysis. ROC analysis was selected because it is an established method for evaluating the diagnostic or discriminative performance of a continuous biomarker in classifying a binary outcome. The area under the curve (AUC) was computed to quantify the overall discriminative performance of the TC/HDL-C ratio. An AUC value closer to 1.0 indicates better predictive power, whereas an AUC of 0.5 indicates that the predictive performance is due to chance. Predictive performance was considered statistically significant if the 95% CI for the AUC did not include the null value of 0.5. Additionally, the interpretation of AUC values is summarized as follows: AUC of 0.90–1.00 =excellent; 0.80–0.89 = good; 0.70–0.79 = fair; 0.60–0.69 = poor; and 0.50–0.59 = failure.³²

Ethical approvals and consent to participate

Ethical clearance was obtained from the Research Ethics Committee (REC) of the Mbarara University of Science and Technology (REC number, MUST-2024-1575) on 10 June 2024 to conduct the primary study.¹² All participants provided written informed consent prior to enrollment. The consent forms were translated into the local language (Runyankore) to ensure comprehension. Formally educated participants gave consent by signing the written forms. For participants with no formal education, the written informed consent form was read aloud in Runyankore, followed by administration of the comprehension screening tool approved by the MUST REC. Only participants who demonstrated understanding were allowed to consent by placing a thumbprint on the informed consent form. Participants were also explicitly asked to provide consent for the use of their data in the secondary analysis, for which they approved. This study was conducted in accordance with the Declaration of Helsinki (1975) as revised in 2024. Confidentiality of the study participants was observed by giving each participant a study code that was not traceable to them. To protect patient privacy, all personal identifiers were removed, and the data were fully deidentified before analysis. Permission to perform secondary analysis was also obtained from the principal investigator of the primary study.

Results

Study flowchart

Narrative of the study flowchart

Data from all 377 PLWH on DTG-based ART from the primary study, who met the eligibility criteria, were included in this study. Of these, a significant number of patients (133 (35.28%)) were diagnosed with MetS. Among those with MetS, more than half (31 (52.5%)) had a high TC/HDL-C ratio.

Demographic characteristics of the study participants

A total of 377 participants were analyzed in the study, with a median age of 44 (IQR: 30–59) years. The study predominantly comprised women, constituting 56.2% of the study population. Married (57.3%) and single (27.9%) participants were identified with regard to marital status; 14.9% of the participants were separated. Overall, 40.1% of the participants had tertiary education, 19.9% secondary education, 18.0% primary education, and 22.0% had no formal education. A significant proportion (57.3%) of the participants resided in urban areas, whereas 42.7% were from rural regions. Religious affiliation was diverse, with Protestants forming the largest group (36.6%), followed by Catholics (29.7%), other religious groups (25.2%), and Muslims (8.5%). Employment status indicated that a majority of the participants (84.1%) were employed, whereas 15.9% were unemployed, as shown in Table 1.

Table 1.

Characteristics of the study participants.

Variables	Total (N = 377), frequency (%)
Sociodemographic characteristics
Age in years, median (IQR)	44 (30–59)
Sex
Male	165 (43.8%)
Female	212 (56.2%)
Marital status
Single	105 (27.9%)
Married	216 (57.3%)
Separated	56 (14.9%)
Educational level
No education	83 (22.0%)
Primary	68 (18.0%)
Secondary	75 (19.9%)
Tertiary	151 (40.1%)
Residence
Urban	216 (57.3%)
Rural	161 (42.7%)
Religion
Catholic	112 (29.7%)
Protestant	138 (36.6%)
Muslim	32 (8.5%)
Others	95 (25.2%)
Employment status
Unemployed	60 (15.9%)
Employed	317 (84.1%)
MetS and its components
MetS
Absent	244 (64.7%)
Present	133 (35.3%)
High fasting plasma glucose
Absent (<110 mg/dL)	322 (85.4%)
Present (≥110 mg/dL)	55 (14.6%)
Central obesity (high waist circumference)
Absent (<88 cm for females and <102 cm for males)	278 (73.7%)
Present (≥88 cm for females and ≥102 cm for males)	99 (26.3%)
Elevated triglycerides
Absent (<150 mg/dL)	112 (29.7%)
Present (≥150 mg/dL)	265 (70.3%)
Low high-density lipoprotein levels
Absent (≥50 mg/dL for females and ≥40 mg/dL for males)	196 (52.0%)
Present (<50 mg/dL for females and <40 mg/dL for males)	181 (48.0%)
High blood pressure (SBP ≥ 130 mmHg and DBP ≥ 85 mmHg)
Absent	145 (38.5%)
Present	232 (61.5%)
Clinical characteristics and biomarkers
Smoking
Nonsmoker	332 (88.1%)
Ever smoked	45 (11.9%)
Alcohol status
Never consumed	143 (37.9%)
Ever consumed	234 (62.1%)
Diagnosed hypertension
No	315 (83.6%)
Yes	62 (16.4%)
Vegetable and fruit intake
<5 servings per day	367 (97.3%)
≥5 servings per day	10 (2.7%)
Global PSQL score, median (IQR)	4 (2–6)
OSA
Low risk of OSA	302 (80.1%)
High risk of OSA	75 (19.9%)
Sleep duration
>7 h	72 (19.1%)
6–7 h	251 (66.6%)
5–6 h	38 (10.1%)
<5 h	16 (4.2%)
Family history of hypertension
No	314 (83.3%)
Yes	63 (16.7%)
Family history of diabetes mellitus
No	285 (75.6%)
Yes	92 (24.4%)
Family history of CVD
No	368 (97.6%)
Yes	9 (2.4%)
Family history of dyslipidemia
No	361 (95.8%)
Yes	16 (4.2%)
Family history of kidney disease
No	315 (83.6%)
Yes	62 (16.4%)
ART duration
≤5 years	115 (30.5%)
>5 years	262 (69.5%)
DTG-based ART duration
≤2 years	112 (29.7%)
>2 years	265 (70.3%)
MUAC (cm)	28 (27–29)
BMI (kg/m²)	22.72 (20.32–25.86)
Total cholesterol (mg/dL), median (IQR)	168 (144–208.8)
AST (IU/L), median (IQR)	21 (15–30)
ALT (IU/L), median (IQR)	23 (17–31)
AST/ALT ratio, median (IQR)	0.90 (0.66–1.28)
ALP (IU/L), median (IQR)	101 (82–119)
GGT (IU/L), median (IQR)	19 (13–29)
Na⁺ (mmol/L), median (IQR)	137.8 (135.40–141)
K⁺ (mmol/L), median (IQR)	4.03 (3.81–4.41)
Cl⁻ (mmol/L), median (IQR)	99.40 (97.30–102.7)
TC/HDL-C ratio, median (IQR)	3.36 (2.55–4.34)
TC/HDL-C ratio
Low (<5)	318 (84.4%)
High (≥5)	59 (15.6%)

This table presents the sociodemographic, MetS and it is components, and clinical characteristics of the study participants, including age, sex, marital status, educational level, residence, religion, and employment status. Data are expressed as median (IQR) for continuous variables and as frequencies (percentages) for categorical variables.

ALT: alanine aminotransferase; ALP: alkaline phosphatase; ART: antiretroviral therapy; AST: aspartate aminotransferase; BMI: body mass index; Cl⁻: chloride ion; DBP: diastolic blood pressure; DTG: dolutegravir; GGT: gamma-glutamyl transferase; HDL-C: high-density lipoprotein cholesterol; IQR: interquartile range; K⁺: potassium ion; MUAC: mid-upper arm circumference; Na⁺: sodium ion; N: number of participants; OSA: obstructive sleep apnea; PSQI: Pittsburgh sleep quality index; SBP: systolic blood pressure; TC: total cholesterol; TG: triglyceride; MetS; metabolic syndrome.

Clinical characteristics of the participants

The majority of the participants were nonsmokers (88.1%) and had a history of alcohol consumption (62.1%). Most participants reported low fruit and vegetable intake, with 97.3% consuming fewer than five servings per day. Only 16.4% had been diagnosed with hypertension, while 19.9% were at high risk of OSA. Sleep duration varied, with most participants (66.6%) reporting 6–7 h of sleep per night. Family history data revealed that 24.4% of the participants had a history of diabetes, 16.7% had a history of hypertension, and 16.4% had a history of kidney disease. Histories of CVD (2.4%) and dyslipidemia (4.2%) were less common. The majority had been on ART for more than 5 years (69.5%), with a similar proportion on DTG-based ART for more than 2 years (70.3%). The median values for the anthropometric and biochemical markers were as follows: BMI, 22.72 kg/m²; MUAC, 28 cm; and TC level, 168 mg/dL; with liver enzymes remaining within normal limits.

MetS and its components and biomarkers

MetS was identified in 35.3% of the participants, indicating a notable prevalence within the cohort. Among its components, elevated TG level was the most common abnormality, affecting 70.3% of the participants. Low HDL levels were observed in 48.0%, while high BP was noted in 61.5% of the participants, further underscoring the metabolic risks within this population. Central obesity, defined by a high WC, was found in 26.3%, and elevated FPG level was detected in 14.6% of the participants. These findings reflect a clustering of cardiovascular and metabolic risk factors. Biomarker evaluation showed a median TC level of 168 mg/dL and a TC/HDL-C ratio of 3.36, with 15.6% of the participants exhibiting a high TC/HDL-C ratio (>5), a marker of atherogenic risk. Liver function markers (AST, ALT, ALP, and GGT) were largely within normal limits, with an AST/ALT ratio median of 0.90, suggesting no overt hepatic dysfunction. Electrolyte balance was maintained, with the median sodium, potassium, and chloride levels within physiological ranges.

Association of high TC/HDL-C ratio with MetS

The median TC/HDL-C ratio was significantly higher in individuals with MetS (3.92, IQR: 3.15–4.87) than in those without (2.96, IQR: 2.29–3.87; p < 0.001). Additionally, the prevalence of MetS differed significantly by TC/HDL-C status (p = 0.003); more than half (31/59 (52.5%)) of the individuals with a high TC/HDL-C ratio had MetS, compared with only 102/318 (32.1%) individuals showing MetS among those with a low TC/HDL-C ratio (Table 2).

Table 2.

Distribution of MetS by TC/HDL-C status.

Variable	Total 377	MetS
Variable	Total 377	Absent (N = 244)	Present (N = 133)	p value
TC/HDL-C ratio: median (IQR)	3.36 (2.55–4.34)	2.96 (2.29–3.87)	3.92 (3.15–4.87)	<0.001
TC/HDL-C ratio				0.003
Low (<5)	318 (84.4%)	216 (67.9%)	102 (32.1%)
High (≥5)	59 (15.6%)	28 (47.5%)	31 (52.5%)

This table presents the distribution of MetS across participants with low and high TC/HDL-C ratios. Results are stratified by MetS presence, with medians (IQR) and frequency distributions reported. Statistical significance (indicated by bold values) is determined using appropriate tests at p < 0.05.

HDL-C: high-density lipoprotein cholesterol; IQR: interquartile range; N: number of participants; TC: total cholesterol; TC: total cholesterol; HDL-C: high-density lipoprotein cholesterol; MetS; metabolic syndrome.

At the bivariate level (Table 3), participants with a high TC/HDL-C ratio were more likely to have MetS (cOR: 2.34, 95% CI: 1.34–4.12, p = 0.003) compared with those with a low TC/HDL-C ratio. Several other factors were also significantly associated with MetS at this level. Female sex was associated with higher odds of MetS than male sex (cOR: 2.44, 95% CI: 1.56–3.82, p < 0.001). Advancing age showed a positive association with MetS (cOR: 1.03, 95% CI: 1.01–1.04, p < 0.001). Those with a BMI of 25–29 kg/m² (cOR: 1.72, p = 0.033) and ≥30 kg/m² (cOR: 4.52, p < 0.001) were more likely to have MetS than those with a BMI of <25 kg/m². Ever smoking (cOR: 1.90, p = 0.044), high MUAC (cOR: 1.17, p < 0.001), high risk of OSA (cOR: 2.24, p = 0.002), family history of hypertension (cOR: 2.18, p = 0.005), duration of DTG-based ART >2 years (cOR: 2.24, p = 0.002), and lower AST levels (cOR: 0.97, p = 0.013) were also significantly associated with MetS.

Table 3.

Association between high TC/HDL-C ratio and MetS.

Variable	Bivariate analysis		Multivariate analysis
Variable	cOR (95% CI)	p value	aOR (95% CI)	p value
TC/HDL-C ratio
Low (<5)	1.00		1.00
High (≥5)	2.34 (1.34–4.12)	0.003	2.06 (1.08–3.91)	0.028
Age (years)	1.03 (1.01–1.04)	<0.001	1.01 (0.99–1.03)	0.161
Sex
Male	1.00		1.00
Female	2.44 (1.56–3.82)	<0.001	3.18 (1.83–5.55)	<0.001
Educational level
No education	1.00
Primary	0.72 (0.38–1.37)	0.314
Secondary	0.61 (0.32–1.15)	0.128
Tertiary	0.32 (0.18–0.57)	<0.001
Residence
Urban	1.00
Rural	1.16 (0.76–1.78)	0.486
Religion
Catholic	1.00
Protestant	1.06 (0.63–1.79)	0.817
Muslim	1.46 (0.65–3.24)	0.357
Others	0.86 (0.48–1.55)	0.622
Employment status
Unemployed	1.00
Employed	0.72 (0.41–1.27)	0.260
BMI
<25 kg/m²	1.00		1.00
25–29 kg/m²	1.72 (1.05–2.84)	0.033	1.07 (0.59–1.93)	0.830
≥30 kg/m²	4.52 (2.01–10.16)	<0.001	1.51 (0.49–4.66)	0.470
Smoking
Nonsmoker	1.00		1.00
Ever smoked	1.90 (1.02–3.57)	0.044	2.85 (1.35–6.02)	0.006
Alcohol status
Never consumed	1.00		1.00
Ever consumed	0.69 (0.45–1.07)	0.094	0.66 (0.40–1.08)	0.100
MUAC (cm)	1.17 (1.09–1.25)	<0.001	1.09 (0.99–1.20)	0.078
Vegetable and fruit intake
<5 servings per day	1.00
≥5 servings per day	0.45 (0.09–2.15)	0.318
Number of days per week of vegetable and fruit intake
≤4	1.00
>4	2.16 (0.79–5.93)	0.134
Poor sleep quality (global PSQI score)
Good quality (≤5)	1.00		1.00
Poor quality (>5)	1.19 (0.74–1.91)	0.465	0.68 (0.36–1.27)	0.227
OSA
Low risk of OSA	1.00		1.00
High risk of OSA	2.24 (1.34–3.74)	0.002	1.70 (0.78–3.74)	0.184
Sleep duration (h)
>7	1.00		1.00
6–7	0.86 (0.50–1.48)	0.594
5–6	0.49 (0.20–1.18)	0.112
<5	1.22 (0.41–3.66)	0.720
Family history of hypertension
No	1.00		1.00
Yes	2.18 (1.26–3.76)	0.005	1.61 (0.77–3.38)	0.205
Family history of diabetes mellitus
No	1.00		1.00
Yes	1.32 (0.82–2.15)	0.255	0.82 (0.44–1.51)	0.519
Family history of cardiovascular diseases
No	1.00
Yes	3.80 (0.93–15.43)	0.062
Family history of dyslipidemia
No	1.00
Yes	1.11 (0.39–3.11)	0.849
Family history of kidney disease
No	1.00
Yes	1.01 (0.57–1.79)	0.970
DTG-based ART duration
≤2 years	1.00		1.00
>2 years	2.24 (1.36–3.70)	0.002	1.30 (0.69–2.44)	0.412
Na⁺ (mmol/L)	0.97 (0.94–1.01)	0.132	0.98 (0.94–1.02)	0.395
K⁺ (mmol/L)	0.78 (0.54–1.13)	0.187	0.85 (0.55–1.31)	0.461
CL⁻ (mmol/L)	0.97 (0.93–1.00)	0.048	0.96 (0.92–1.01)	0.087
AST	0.97 (0.95–0.99)	0.013	0.97 (0.95–1.00)	0.070
ALT	1.00 (0.98–1.02)	0.920	1.01 (0.98–1.04)	0.435
ALP	1.00 (1.00–1.01)	0.521	1.00 (1.00–1.01)	0.385
GGT	1.00 (0.98–1.02)	0.873	1.00 (0.98–1.02)	0.719

Hosmer–Lemeshow test: p = 0.6617.

This table examines the association between a high TC/HDL-C ratio and MetS using bivariate and multivariate logistic regression analyses. Results are expressed as crude odds ratios (cOR) and adjusted odds ratios (aOR) with 95% confidence intervals (CIs). Key variables include demographic factors, clinical indicators, metabolic markers, and sleep quality. Statistical significance (indicated by bold values) is determined at p < 0.05.

aOR: adjusted odds ratio; ALT: alanine aminotransferase; ALP: alkaline phosphatase; AST: aspartate aminotransferase; BMI: body mass index; CI: confidence interval; cOR: crude odds ratio; DTG: dolutegravir; GGT: gamma-glutamyl transferase; HDL-C: high-density lipoprotein cholesterol; MUAC: mid-upper arm circumference; N: number of participants; OSA: obstructive sleep apnea; PSQI: Pittsburgh sleep quality index; TC: total cholesterol; TG: triglyceride.

After adjustment for confounding variables, a high TC/HDL-C ratio was significantly associated with MetS, with an aOR of 2.06 (95% CI: 1.08–3.91, p = 0.028) (Table 3). The TC/HDL-C ratio at an optimal cutoff of 3.30 demonstrated a significant predictive ability (AUC = 0.696, 95% CI: 0.642–0.750) to differentiate participants with MetS from those without MetS, with a sensitivity of 73% and specificity of 60% (Figure 1). Other important variables in the multivariate analysis that were associated with MetS included female sex (aOR: 3.18, 95% CI: 1.83–5.55, p < 0.001) and a history of smoking (aOR: 2.85, 95% CI: 1.35–6.02, p = 0.006).

Figure 1.

ROC curve showing the predictive performance of the TC/HDL-C ratio for MetS.

Discussion

In this Ugandan cohort of PLWH, a higher TC/HDL‐C ratio strongly indicated MetS. Patients with MetS had a significantly higher median TC/HDL‐C ratio (3.92 vs. 2.96; p < 0.001), and over half (52.5%) of those with a “high” ratio (≥5) had MetS versus 32.1% with a low ratio. After adjusting for confounders, a high TC/HDL‐C ratio was associated with a two‐fold increase in MetS risk (aOR = 2.06, p = 0.028). These findings are clinically meaningful—a raised TC/HDL‐C ratio is a known atherogenic indicator. In the context of HIV, it may identify patients with multiple metabolic derangements. For example, van Rooyen et al.³³ found that among PLWH in Africa, a TC/HDL‐C ratio ≥5.4 identified individuals at greater cardiovascular risk. Our results suggest that even at modest elevations (optimal cutoff = 3.30, AUC = 0.696), the ratio significantly distinguishes patients with MetS (sensitivity: 73%, specificity: 60%). In practice, routinely calculating TC/HDL‐C ratios during HIV care could help identify patients who require early lifestyle or pharmacologic interventions to forestall diabetes, hypertension, and CVD.

Biologically, the TC/HDL‐C ratio captures the dyslipidemia underlying MetS. A high ratio reflects both excess TC (often from high LDL or TG‐rich lipoproteins) and low HDL‐C levels. Reduced HDL‐C level is a core feature of MetS and insulin resistance, impairing reverse cholesterol transport and promoting endothelial dysfunction. Insulin resistance itself drives an atherogenic lipid profile (elevated very-low-density lipoprotein (VLDL)‐TG and small dense LDL levels and suppressed HDL level).¹⁶ Indeed, population studies show that the TC/HDL‐C ratio is independently associated with insulin resistance and multiple MetS components.¹⁶ In a large Japanese community sample, both TG/HDL‐C and TC/HDL‐C ratios were closely associated with the number of MetS features and homeostatic model assessment of insulin resistance (HOMA‐IR).¹⁶ Thus, the TC/HDL‐C ratio likely serves as an integrated marker of the underlying pathophysiology, combining excess atherogenic lipoproteins with deficient protective HDL that promotes MetS.

Our findings not only align in part with prior literature but also highlight differences. Fernández‐Aparicio et al.³⁴ performed a cross‐sectional study of 981 Spanish adolescents and showed that the TC/HDL ratio robustly discriminated MetS (ROC AUC > 0.83) in both sexes. Similarly, Zhang et al.³⁵ conducted a population-based study of 1669 Chinese adults and reported a significant association between the TC/HDL‐C ratio and insulin resistance, a core MetS component. More recently, Babaahmadi‐Rezaei et al.³⁶ analyzed 9809 Iranian adults and found significantly higher TC/HDL ratios among MetS cases; their ROC analysis yielded AUCs of 0.676 (95% CI: 0.661–0.690) in men and 0.743 (95% CI: 0.732–0.754) in women. Additional Iranian studies have consistently identified the TC/HDL‐C ratio as a key diagnostic marker for MetS.^37,38 In Korea, the TC/HDL‐C ratio showed a strong predictive value for MetS,³⁹ whereas the U.S. data associated an elevated TC/HDL‐C ratio with peripheral artery disease,⁴⁰ further highlighting its cardiometabolic relevance. However, Rezapour et al.³⁸ reported that although the TC/HDL‐C ratio had a high AUC (0.79), the TG/HDL‐C ratio (AUC = 0.85) was a stronger predictor of MetS, a finding echoed by Saeedi et al.,¹⁴ who showed that the TG/HDL‐C ratio outperformed the TC/HDL‐C ratio in an elderly Iranian cohort. These studies concluded that the TG/HDL‐C ratio was superior to the TC/HDL‐C ratio, in contrast to our emphasis on TC/HDL‐C. Possible reasons include insufficient sample size and population differences; our PLWH on ART may exhibit greater perturbation of TC relative to TG or differences in age and comorbidities. Although our AUC estimate of 0.696 (95% CI: 0.642–0.750) did not reach the conventional threshold for strong discrimination, the CI excludes the null value of 0.5, indicating statistically significant discriminatory ability of the TC/HDL-C ratio for MetS beyond chance. This suboptimal predictive estimate with lower sensitivity may have been limited by the relatively small sample size and nonrepresentative sample in our study. van Rooyen et al,³³ who analyzed 140 black South Africans with HIV infection, identified a TC/HDL‐C ratio ≥5.4 as a marker of cardiovascular risk, a threshold higher than ours, suggesting differing lipid norms due to ART or diet. Although Muyanja et al.⁴¹ reported a high MetS prevalence in Ugandans on ART, the TC/HDL‐C ratio was not specifically assessed. Overall, our findings support the TC/HDL‐C ratio as a relevant MetS marker in HIV-infected populations, whereas discrepancies with non-HIV cohorts emphasizing the TG/HDL‐C ratio may be attributed to variations in demographic characteristics, comorbid conditions, and lipid metabolism.

Importantly, our study provides new insights into cardiometabolic risk in sub‐Saharan Africa. It is among the first studies to identify the TC/HDL‐C ratio as a predictor of MetS in an HIV‐infected Ugandan cohort, filling a research gap. Most African HIV-infected cohorts focus on individual lipids or classic MetS criteria, rarely assessing composite lipid indices. By demonstrating that even a moderately elevated TC/HDL‐C ratio (cutoff = 3.30) doubles the risk of MetS, we provide evidence that this simple measure could improve early risk detection in HIV care. In our resource‐limited setting, the TC/HDL‐C ratio readily computed from standard lipid panels could serve as an affordable screening tool for MetS, prompting targeted counseling or lipid‐lowering interventions. Thus, our findings contribute to global knowledge by extending it to a high‐HIV‐burden, resource‐constrained environment and underscore the need to employ cardiometabolic screening strategies, such as using the TC/HDL‐C ratio, in the local patient population.

Recommendations

We recommend integrating routine TC/HDL-C ratio screening into HIV care protocols for individuals on DTG-based ART in Uganda. For healthcare providers, this can enhance the detection of MetS and associated cardiovascular risk early, enabling early intervention and individualized lifestyle modifications to foster improved health outcomes among this vulnerable group. Policymakers should consider updating national HIV treatment guidelines to include lipoprotein ratio monitoring and support capacity-building for lipid profiling within HIV clinics, ensuring sustainable implementation and improved long-term health outcomes for this high-risk population.

Limitations

There are some limitations in our study that must be kept in mind when interpreting the results. First, as a secondary analysis of a cross-sectional study, it provides only a snapshot of associations between the TC/HDL-C ratio and MetS, and we cannot establish causal directions or temporality of events easily. Additionally, some of the variables, including family history of hypertension, CVD, kidney disease, and diabetes mellitus, were reported by the participants themselves; therefore, there is a likelihood of incorrect information being provided. We also recognize the influence of unmeasured confounding factors, including lifestyle, other comorbid conditions, and medication use, which were not controlled for during analysis and can influence the observed associations. Even after controlling for established confounders, residual confounding could still exist. Finally, the generalizability of our results is restricted to PLWH on DTG-based ART, and the findings might not be relevant to other populations or to individuals on other ART regimens.

In light of the limitations identified in our study, we recommend several directions for future research to strengthen the understanding of the relationship of the TC/HDL-C ratio and other lipoprotein ratios with MetS among PLWH on DTG-based ART. First, longitudinal studies are essential to establish the temporal and causal relationships between an elevated TC/HDL-C ratio and the development of MetS, which helps address the inherent limitations of the cross-sectional design. Additionally, future research should incorporate objective clinical and biochemical assessments rather than self-reports for variables such as family history of chronic diseases to reduce the possibility of incorrect information. It is also critical to account for a broader range of potential confounding factors, including detailed information on lifestyle behaviors, concurrent comorbidities, and use of other medications, through comprehensive data collection and multivariable adjustment. Interventional studies may further explore whether targeted lipid-lowering or lifestyle interventions in patients with an elevated TC/HDL-C ratio can mitigate MetS risk. Finally, to enhance generalizability, future studies should include diverse HIV-positive populations across various ART regimens and geographic settings.

Conclusion

Our study demonstrates that a high TC/HDL-C ratio is strongly associated with MetS among PLWH on DTG-based ART in southwestern Uganda. This study highlights the potential utility of lipoprotein ratio profiling for the early identification of metabolic risk, enabling timely interventions to prevent cardiovascular complications in this vulnerable population.

Footnotes

Acknowledgments

We express our deepest gratitude to the principal investigator for granting permission to conduct this secondary data analysis. We also acknowledge the study participants for accepting to participate in the primary study and providing consent for use of the data for secondary analysis. Additionally, we used AI tools to support language refinement and clarity during manuscript preparation.

Author contributions

CNB, CN, and ES participated in conceptualization of the study. CNB performed data analysis and result interpretation. CNB, CN, ES, MJM, SO, FS, DN, RK, FS, BM, WA, DT, LOO, and SPR contributed to writing the first draft of the manuscript with CNB, ES, MJM, DT, and SPR providing critical revisions. All authors have read and approved the first draft of the manuscript.

Availability of data and materials

The datasets for this study have been submitted to the journal.

Consent for publication

All authors have consented to the publication of this work.

Declaration of conflicting interests

The authors declare no conflict of interest.

Funding

The study did not receive any funding.

ORCID iDs

Charles Nkubi Bagenda

Michael Junior Mugisa

References

Mojola

Angotti

Denardo

, et al. The end of AIDS? HIV and the new landscape of illness in rural South Africa. Glob Public Health 2022; 17: 13–25.

Trachunthong

Tipayamongkholgul

Chumseng

, et al. Burden of metabolic syndrome in the global adult HIV-infected population: a systematic review and meta-analysis. BMC Public Health 2024; 24: 2657.

Woldu

Minzi

Engidawork

Prevalence of cardiometabolic syndrome in HIV-infected persons: a systematic review. J Diabetes Metab Disord 2020; 19: 1671–1683.

Masenga

Elijovich

Koethe

, et al. Hypertension and metabolic syndrome in persons with HIV. Curr Hypertens Rep 2020; 22: 78–78.

Gupta

Tyagi

Mathews

, et al. Non-alcoholic fatty liver disease and metabolic syndrome in patients with HIV/AIDS and its correlation with antiretroviral therapy and severity of disease. SL J Med 2023; 32: 18–25.

Kumari

Kumar

Kant

An update on metabolic syndrome: metabolic risk markers and adipokines in the development of metabolic syndrome. Diabetes Metab Syndr 2019; 13: 2409–2417.

Yusufu

Nagu

Ottaru

, et al. The prevalence of metabolic syndrome and associated factors among adults on antiretroviral therapy in Dar es Salaam, Tanzania. Res Sq 2024; rs.3.rs: 4372785.

Asgedom

Kebede

Gebrekidan

, et al. Prevalence of metabolic syndrome among people living with human immunodeficiency virus in sub-Saharan Africa: a systematic review and meta-analysis. Sci Rep 2024; 14: 11709.

Amutuhaire

Mulindwa

Castelnuovo

, et al. Prevalence of cardiometabolic disease risk factors in people with HIV initiating antiretroviral therapy at a high-volume HIV clinic in Kampala, Uganda. Open Forum Infect Dis 2023; 10: ofad241.

10.

Jemal

Ashenef

Sinamaw

, et al. Metabolic syndrome among people living with HIV on dolutegravir and efavirenz-based antiretroviral therapy in Ethiopia: a comparative cross-sectional study. J Int Assoc Provid Aids Care 2024; 23: 23259582241303305.

11.

Mirai

Kilonzo

Sadiq

, et al. Metabolic syndrome and associated factors among people living with HIV on dolutegravir-based antiretroviral therapy in northern Tanzania. J Int Assoc Provid Aids Care 2024; 23: 23259582241306492.

12.

Nzaramba

Bagenda

Mudondo

, et al. Low aspartate aminotransferase/alanine aminotransferase ratio as an indicator of metabolic syndrome among HIV patients on dolutegravir therapy in southwestern Uganda. Cureus 2025; 17: e77166.

13.

Heyens

Busschots

Koek

, et al. Liver fibrosis in non-alcoholic fatty liver disease: from liver biopsy to non-invasive biomarkers in diagnosis and treatment. Front Med (Lausanne) 2021; 8: 615978.

14.

Saeedi

Baqeri

Bidokhti

, et al. Clinical utility of lipid ratios as potential predictors of metabolic syndrome among the elderly population: Birjand Longitudinal Aging Study (BLAS). BMC Geriatr 2023; 23: 403.

15.

Kosmas

Rodriguez Polanco

Bousvarou

, et al. The triglyceride/high-density lipoprotein cholesterol (TG/HDL-C) ratio as a risk marker for metabolic syndrome and cardiovascular disease. Diagnostics 2023; 13: 929.

16.

Kawamoto

Tabara

Kohara

, et al. Relationships between lipid profiles and metabolic syndrome, insulin resistance and serum high molecular adiponectin in Japanese community-dwelling adults. Lipids Health Dis 2011; 10: 79.

17.

Ren

Shi

Ding

, et al. Total cholesterol to high-density lipoprotein cholesterol ratio is a significant predictor of nonalcoholic fatty liver: Jinchang cohort study. Lipids Health Dis 2019; 18: 47–47.

18.

Zhou

Tian

Xie

WN.

Total cholesterol to high-density lipoprotein ratio and nonalcoholic fatty liver disease in a population with chronic hepatitis B. World J Hepatol 2022; 14: 791–801.

19.

Ministry of Health. Non-Communicable Disease Risk Factor Baseline Survey. Uganda 2014 report, https://cdn.who.int/media/docs/default-source/ncds/ncd-surveillance/data-reporting/uganda/steps/uganda_2014_steps_report.pdf?sfvrsn=95eba2e8_3&download=true (2014, accessed 12 April 2025).

20.

Betunga

Atuhaire

Nakasiita

, et al. Factors influencing the use of multiple HIV prevention services among transport workers in a city in southwestern Uganda. PLOS Glob Public Health 2023; 3: e0001350.

21.

Lamorde

Atwiine

Owarwo

, et al. Dolutegravir-associated hyperglycaemia in patients with HIV. Lancet Hiv 2020; 7: e461–e462.

22.

Nabitaka

Nawaggi

Campbell

, et al. High acceptability and viral suppression of patients on dolutegravir-based first-line regimens in pilot sites in Uganda: a mixed-methods prospective cohort study. PloS One 2020; 15: e0232419.

23.

Von Elm

Altman

Egger

; STROBE Initiativeet al. The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement: guidelines for reporting observational studies. Lancet 2007; 370: 1453–1457.

24.

Grundy

Cleeman

Daniels

; National Heart, Lung, and Blood Instituteet al. Diagnosis and management of the metabolic syndrome: an American Heart Association/National Heart, Lung, and Blood Institute scientific statement. Circulation 2005; 112: 2735–2752.

25.

Mach

Baigent

Catapano

; ESC Scientific Document Groupet al. 2019 ESC/EAS guidelines for the management of dyslipidaemias: lipid modification to reduce cardiovascular risk: The Task Force for the management of dyslipidaemias of the European Society of Cardiology (ESC) and European Atherosclerosis Society (EAS). Eur Heart J 2020; 41: 111–188.

26.

National Cholesterol Education Program (NCEP) Expert Panel on Detection, Evaluation, and Treatment of High Blood Cholesterol in Adults (Adult Treatment Panel III). Third report of the National Cholesterol Education Program (NCEP) expert panel on detection, evaluation, and treatment of high blood cholesterol in adults (Adult Treatment Panel III) final report. Circulation 2002; 106: 3143–3421.

27.

Nagappa

Liao

Wong

, et al. Validation of the STOP-Bang questionnaire as a screening tool for obstructive sleep apnea among different populations: a systematic review and meta-analysis. PloS One 2015; 10: e0143697.

28.

Avelino

Da Silva

Chaves

, et al. Triglyceride-glucose index is associated with poor sleep quality in apparently healthy subjects: a cross-sectional study. Arch Endocrinol Metab 2023; 67: 73–91.

29.

Bertolazi

Fagondes

Hoff

, et al. Validation of the Brazilian Portuguese version of the Pittsburgh sleep quality index. Sleep Med 2011; 12: 70–75.

30.

World Health Organization. Obesity: preventing and managing the global epidemic: report of a WHO consultation, https://iris.who.int/handle/10665/42330 (2000, accessed 15 April 2025).

31.

Hannon

Khan

Teran-Garcia

Nutrigenetic contributions to dyslipidemia: a focus on physiologically relevant pathways of lipid and lipoprotein metabolism. Nutrients 2018; 10: 1404.

32.

Çorbacıoğlu

ŞK

Aksel

Receiver operating characteristic curve analysis in diagnostic accuracy studies: a guide to interpreting the area under the curve value. Turk J Emerg Med 2023; 23: 195–198.

33.

van Rooyen

Fourie

Steyn

, et al. Cardiometabolic markers to identify cardiovascular disease risk in HIV-infected black South Africans. S Afr Med J 2014; 104: 195–199.

34.

Fernández-Aparicio

Perona

Schmidt-RioValle

, et al. Assessment of different atherogenic indices as predictors of metabolic syndrome in Spanish adolescents. Biol Res Nurs 2022; 24: 163–171.

35.

Zhang

Chen

Deng

, et al. Association between lipid ratios and insulin resistance in a Chinese population. PloS One 2015; 10: e0116110.

36.

Babaahmadi‐Rezaei

Raeisizadeh

Zarezade

, et al. Comparison of atherogenic indices for predicting the risk of metabolic syndrome in southwest Iran: results from the Hoveyzeh Cohort Study (HCS). Diabetol Metab Syndr 2024; 16: 112.

37.

Abbasian

Delvarianzadeh

Ebrahimi

, et al. Lipid ratio as a suitable tool to identify individuals with MetS risk: a case- control study. Diabetes Metab Syndr 2017; 11: S15–S19.

38.

Rezapour

Shahesmaeili

Hossinzadeh

, et al. Comparison of lipid ratios to identify metabolic syndrome. Arch Iran Med 2018; 21: 572–577.

39.

Chu

Jung

Park

, et al. Risk assessment of metabolic syndrome in adolescents using the triglyceride/high-density lipoprotein cholesterol ratio and the total cholesterol/high-density lipoprotein cholesterol ratio. Ann Pediatr Endocrinol Metab 2019; 24: 41–48.

40.

Aday

Lawler

Cook

, et al. Lipoprotein particle profiles, standard lipids, and peripheral artery disease incidence: prospective data from the Women’s Health Study. Circulation 2018; 138: 2330–2341.

41.

Muyanja

Muzoora

Muyingo

, et al. High prevalence of metabolic syndrome and cardiovascular disease risk among people with HIV on stable ART in southwestern Uganda. AIDS Patient Care Stds 2016; 30: 4–10.

Total cholesterol/high-density lipoprotein cholesterol ratio is a significant predictor of metabolic syndrome among people on dolutegravir-based antiretroviral therapy: A cross-sectional study in southwestern Uganda

Abstract

Objective

Methods

Results

Conclusion

Keywords

Introduction

Material and methods

Study design, population, and variables

Sample size determination

Statistical analysis

Ethical approvals and consent to participate

Results

Study flowchart

Narrative of the study flowchart

Demographic characteristics of the study participants

Clinical characteristics of the participants

MetS and its components and biomarkers

Association of high TC/HDL-C ratio with MetS

Discussion

Recommendations

Limitations

Conclusion

Footnotes

Acknowledgments

Author contributions

Availability of data and materials

Consent for publication

Declaration of conflicting interests

Funding

ORCID iDs

References