Association of BMI with mortality in drug-induced liver injury

Background To clarify the associations between BMI and the incidences of all-cause death or liver-related death (LRD)/liver transplantation (LT) in drug-induced liver injury (DILI). Methods DILI patients from three hospitals were retrospectively retrieved and follow-up from 2009 to 2021. They were categorized into underweight (BMI < 18.5 kg/m 2 ), normal weight (BMI of 18.5–23.9 kg/m 2 ), overweight (BMI of 24–27.9 kg/m 2 ) and obese (BMI ≥ 28 kg/m 2 ) groups. Cox regression models were conducted to reveal the effect of BMI on all-cause death or LRD/LT. Results A total of 1469 eligible DILI patients were included: underweight 73 (4.97%), normal weight 811 (55.21%), overweight 473 (32.20%) and obese 112 (7.62%). Eighty-nine patients (6.06%) had all-cause death, of which 66 patients (4.49%) had LRD/LT. The median age was 52 years old, and females were 1039 (70.73%). The associations between BMI and all-cause mortality ( nonlinear test P < 0.01) or liver-related mortality/LT ( nonlinear test P = 0.01) were J-shaped. Multivariate Cox regression analysis showed that underweight (HR: 3.02, 95% CI: 1.51–6.02) was significantly associated with all-cause mortality after adjusting for age and sex. Furthermore, obese males were significantly associated with liver-related mortality/LT (HR: 3.49, 95% CI: 1.13–10.72) after additional adjustment for serological indices and comorbidities. Conclusion Association between BMI and mortality is a J-shape. The overall mortality was significantly higher in underweight and obese group. Male obesity is independently associated with LRD/LT. These findings indicate that DILI patients with extreme BMI would have a high risk of dismal outcomes, which warrants extra medical care.