Impact of estimated glomerular filtration rate (eGFR) on in-hospital mortality: an age- and HIV status-specific retrospective cohort study in Uganda.

BACKGROUND

Limited studies have explored the relationship between estimated Glomerular Filtration Rate(eGFR) and in-hospital mortality(IHM) in low-income sub-Saharan African countries. This study aimed to explores this association, offering insights into its impact in resource-limited settings.

METHODS AND RESULTS

We retrospectively included 226 patients(age 45.35 ± 18.85yrs, 54.4% women) admitted to Naguru-referral hospital between January 1st and June 30th, 2024. Baseline demographics and clinical variables, including eGFR, were recorded at admission. Patients were followed from date of admission to discharge and primary outcome was IHM. Multivariable Hazard regression analysis assessed the association between eGFR and IHM, respectively. During follow-up, 45(19.9%) of patients died. Per-standard deviation(SD) increase in eGFR(48.60 mL/min/1.73m) was associated with Hazard Ratio(HR) of 0.46[95%CI: 0.282-0.759, p = 0.002, β = -0.77] for IHM in fully adjusted models. When stratified by eGFR quartiles, using highest quartile(≥ 120 mL/min/1.73m) as reference, HR was 1.08[95%CI: 0.276-4.226, p = 0.912, β =  + 0.08] for 99.0-120 mL/min/1.73m; 4.08[95%CI: 1.284-12.954, p = 0.017, β =  + 1.41] for 66.8-99.0 mL/min/1.73m, and 4.08[95%CI: 1.284-12.954, p = 0.037, β =  + 1.25] for < 66.8 mL/min/1.73m. Among age stratification-subgroups: age < 40yrs: 0.93[95%CI: 0.89-0.97, p < 0.001, β = -0.07]; 40-60yrs: 0.98[95%CI: 0.966-0.999, p = 0.039, β = -0.02]; ≥ 60yrs, p < 0.005 with p-value for age = 0.046; and HIV-positive: 0.94[95%CI: 0.905-0.974, p < 0.001, β = -0.06] with p-value = 0.021. Significant Pearsons-correlation(r) was observed only in: [< 40yrs, HIV(-)] with p = 0.016, r = -0.275; [40-60yrs, HIV( +)] with p = 0.020, r = -0.397; and [≥ 60yrs,HIV( +)] with p = 0.003, r = -0.997.

CONCLUSIONS

We report that eGFR was associated with in-hospital mortality, with a stronger association observed in HIV-negative patients(< 40yrs) and HIV-positive patients (aged ≥ 60yrs yrs). Further research is warranted to validate these findings.