Comparison of Prognostic Accuracy of Albumin–Bilirubin Score and Child–Pugh Score in Cirrhosis

Abstract

Background: Accurate prognostication in cirrhosis guides triage, transplant referral and end-of-life discussions. The Child–Pugh (CTP) score, although widely used, incorporates subjective variables. The albumin–bilirubin (ALBI) score is an entirely laboratory-based alternative, but Indian data comparing the two scores are scarce.

Aim: To compare short-term (7- and 30-day) prognostic performance of the ALBI score with CTP and MELD scores in cirrhosis.

Methods: In this single-centre observational study, 150 consecutive adults (>18 years) with ultrasound-confirmed cirrhosis admitted between January 2023 and June 2024 were prospectively enrolled. Serum albumin, bilirubin, creatinine and INR were measured within 24 h; ALBI, CTP and MELD scores were calculated. Primary outcome was all-cause 7-day mortality; secondary outcome was 30-day mortality. Discrimination was assessed with receiver-operating-characteristic (ROC) analysis and compared with DeLong’s test.

Results: Mean age was 40.0 ± 8.8 years; 69 % were male. Aetiologies were alcohol (42 %), hepatitis B (31 %) and hepatitis C (19 %). Twenty-one patients (14 %) died within 7 days and 31 (21 %) within 30 days. For 7-day mortality, ALBI showed an AUC 0.888 (95 % CI 0.80–0.97) versus CTP 0.718 and MELD 0.724 (p < 0.001 for ALBI vs both). A cut-off > –0.72 gave 81 % sensitivity and 88 % specificity. For 30-day mortality, ALBI again out-performed CTP and MELD (AUC 0.940 vs 0.741 and 0.751; p < 0.001).

Conclusions: In this Indian cohort the ALBI score provided superior, objective prediction of short-term mortality compared with CTP and MELD scores. Its simplicity and bedside applicability support routine use.