Prevalence and Predictors of Dyslipidemia and Diabetes Mellitus Among cART-Treated HIV Patients in Western Ethiopia: A Cross-Sectional Study
Nano Debele Gadisa, Mekdes Alem Baye, Halima Abdulkadir Hussein, Abebe Edao Nagesso, Mintewab Hussein, Afework Habtamu Aseme, Bezawit Fekadu, Demeke Jabesa Aga, Guta Tuna Edao, Gutama Diriba Tucho, Mikias Shitaw Tadese, Mitiku Deresa, Gobena Dedefo Dekebo, Mistire Wolde Gebre, Samuel Kinde BirruBackground
Combination of antiretroviral therapy (cART) has improved the life expectancy of HIV patients. Yet, it can also increase the risk of metabolic dysfunction, including dyslipidemia and diabetes mellitus (DM). Despite this, local data on the prevalence and risk factors for dyslipidemia and DM are limited. In addition, prognostic tools for regular monitoring and thereby the early detection of metabolic dysfunction have yet to be validated among cART-treated patients.
Methods
A hospital-based cross-sectional study was conducted at Nekemte Specialized Hospital (Western Ethiopia) from February to April 2025. A total of 176 cART-treated HIV patients were purposively selected for the study. Socio-demographic, anthropometric, and metabolic parameters were assessed, and their associations with dyslipidemia and DM were evaluated using univariate (bivariate) and multivariable regression analyses. The predictive performances of lipid index-based prognostic tools were assessed using receiver operating characteristic curve analysis.
Results
Dyslipidemia and previously undiagnosed DM were observed among 54.5% and 9.7% of participants, respectively. Significant risk factors included family history of DM (AOR = 4.5), frequent alcohol use (AOR = 5.2), age greater than 40 years (AOR = 3.4), sedentary lifestyle (AOR = 5.6), frequent consumption of traditional spiced milk butter (AOR = 5.2), elevated TyG-WC (AOR = 6.7), and high VAI (AOR = 8.7). TyG-WC had the highest predictive performance for dyslipidemia (AUC = 0.795) with an optimal cutoff of 434.1, followed by TyG-BMI (cutoff 128.5), LAP (cutoff 34.5), and VAI (cutoff 1.91).
Conclusion
Dyslipidemia and undiagnosed diabetes remain highly prevalent among cART-treated HIV patients compared to the general population. In addition to commonly recognized risk factors, frequent consumption of a high-fat, carbohydrate-restricted diet like Nitter kibbeh may influence metabolic outcomes in this sub-population, highlighting the need for further research. Lipid indices, particularly TyG-WC, appear to be feasible tools for routine monitoring and early prediction of metabolic dysfunction. Revising population-specific thresholds could further improve early detection in this group.