Novel inflammatory metabolic parameters as predictors of metabolic dysfunction-associated fatty liver disease in people living with HIV receiving antiretroviral therapy: a retrospective cohort study.
Metabolic dysfunction-associated fatty liver disease (MAFLD) has emerged as a major non-communicable comorbidity in people living with HIV (PLWH). The role of inflammatory metabolic parameters-including the lymphocyte-to-high-density lipoprotein cholesterol ratio (LHR), platelet to high-density lipoprotein cholesterol ratio (PHR), and aggregate index of systemic inflammation (AISI)-in predicting MAFLD in PLWH remains unclear. This study aimed to evaluate the prognostic value of inflammatory metabolic parameters for predicting MAFLD in PLWH.
We conducted a retrospective cohort study of 814 PLWH receiving stable antiretroviral therapy (ART) between 2018 and 2025 at a tertiary care center. Baseline demographic, clinical, and laboratory data were collected. LHR, PHR, and AISI were calculated and categorized into tertiles. The primary outcome was incident MAFLD, diagnosed according to Chinese guidelines. Kaplan-Meier survival analysis, Cox proportional hazards regression, restricted cubic spline (RCS) models, and time-dependent receiver operating characteristic (ROC) curves were applied to assess associations and predictive performance. Sensitivity analyses were performed across subgroups.
During a median follow-up of 2.8 years, 116 participants (14.3%) developed MAFLD, corresponding to an incidence rate of 50.4 per 1,000 person-years. Higher LHR and PHR tertiles were significantly associated with increased MAFLD risk, whereas AISI showed no predictive value. In fully adjusted Cox models, the high LHR tertile remained an independent risk factor (HR = 2.315, 95% CI: 1.208-4.436, p = 0.011), while only the middle PHR tertile retained significance. RCS analyses showed no significant non-linear associations for LHR, PHR, or AISI. Time-dependent ROC analyses demonstrated that LHR had the strongest short-term predictive ability (AUC = 0.713 at 1 year), followed by PHR (AUC = 0.644), while AISI consistently performed poorly (AUC < 0.600). Subgroup analyses confirmed the robustness of LHR and PHR associations across demographic, clinical, and metabolic subgroups.
In this large cohort of PLWH on ART, LHR and PHR were independent predictors of incident MAFLD, with LHR demonstrating the strongest and most consistent predictive value. AISI was not associated with MAFLD. LHR and PHR, as simple, low-cost indices derived from routine laboratory tests, may serve as practical tools for identifying high-risk individuals who could benefit from early monitoring and targeted intervention.
We conducted a retrospective cohort study of 814 PLWH receiving stable antiretroviral therapy (ART) between 2018 and 2025 at a tertiary care center. Baseline demographic, clinical, and laboratory data were collected. LHR, PHR, and AISI were calculated and categorized into tertiles. The primary outcome was incident MAFLD, diagnosed according to Chinese guidelines. Kaplan-Meier survival analysis, Cox proportional hazards regression, restricted cubic spline (RCS) models, and time-dependent receiver operating characteristic (ROC) curves were applied to assess associations and predictive performance. Sensitivity analyses were performed across subgroups.
During a median follow-up of 2.8 years, 116 participants (14.3%) developed MAFLD, corresponding to an incidence rate of 50.4 per 1,000 person-years. Higher LHR and PHR tertiles were significantly associated with increased MAFLD risk, whereas AISI showed no predictive value. In fully adjusted Cox models, the high LHR tertile remained an independent risk factor (HR = 2.315, 95% CI: 1.208-4.436, p = 0.011), while only the middle PHR tertile retained significance. RCS analyses showed no significant non-linear associations for LHR, PHR, or AISI. Time-dependent ROC analyses demonstrated that LHR had the strongest short-term predictive ability (AUC = 0.713 at 1 year), followed by PHR (AUC = 0.644), while AISI consistently performed poorly (AUC < 0.600). Subgroup analyses confirmed the robustness of LHR and PHR associations across demographic, clinical, and metabolic subgroups.
In this large cohort of PLWH on ART, LHR and PHR were independent predictors of incident MAFLD, with LHR demonstrating the strongest and most consistent predictive value. AISI was not associated with MAFLD. LHR and PHR, as simple, low-cost indices derived from routine laboratory tests, may serve as practical tools for identifying high-risk individuals who could benefit from early monitoring and targeted intervention.