Abstract / Summary
Abstract Background Older adults constitute a growing and heterogeneous proportion of intensive care unit (ICU) admissions, and early risk stratification in this group remains difficult. The lactate-to-albumin ratio (LAR) is a composite biomarker that combines a marker of acute metabolic derangement with one of physiological reserve, and has shown prognostic value in several critical care settings. Its performance in geriatric ICU patients, and specifically whether it adds information to established severity scores, has not been established. Methods We conducted a retrospective, single-centre cohort study of consecutive patients aged ≥ 65 years admitted to the mixed medical–surgical ICU of a tertiary referral hospital in Türkiye between January 2024 and July 2025. LAR was calculated as serum lactate (mmol/L) divided by serum albumin (g/dL), from samples obtained within the first 6 h of ICU admission. The primary outcome was all-cause 28-day mortality. Discrimination was assessed by area under the receiver operating characteristic curve (AUC), and curves were compared using the DeLong test. Multivariable logistic regression assessed whether LAR was independently associated with mortality after adjustment for age, sex, Clinical Frailty Scale (CFS), SOFA, APACHE II and C-reactive protein (CRP). Results Of 296 screened admissions, 294 were analysed; 98 patients (33.3%) died within 28 days. Non-survivors had higher admission LAR than survivors [1.37 (IQR 0.84–2.50) vs. 0.62 (IQR 0.46–0.99), p < 0.001]. LAR discriminated 28-day mortality with an AUC of 0.793 (95% CI 0.740–0.846), significantly better than serum lactate alone (AUC 0.746; ΔAUC 0.047, 95% CI 0.021–0.074, p = 0.0005) and serum albumin alone (AUC 0.656; ΔAUC 0.137, 95% CI 0.061–0.213, p = 0.0004). APACHE II discriminated considerably better than LAR (AUC 0.924, 95% CI 0.893–0.955; ΔAUC 0.131, 95% CI 0.080–0.181, p < 0.001). Adding LAR to APACHE II did not significantly improve discrimination (AUC 0.929 vs. 0.924; ΔAUC 0.006, 95% CI −0.0001 to 0.0115, p = 0.055; likelihood ratio test p = 0.098). On univariable analysis LAR was strongly associated with mortality (OR 2.17, 95% CI 1.63–2.88, p < 0.001), but after full adjustment this association was no longer statistically significant (adjusted OR 1.29, 95% CI 1.00–1.66, p = 0.052). APACHE II (adjusted OR 1.36, 95% CI 1.21–1.52, p < 0.001) and CRP (adjusted OR 1.06 per 10 mg/L, 95% CI 1.01–1.11, p = 0.015) remained independently associated with mortality. Conclusions In critically ill older adults, admission LAR discriminates 28-day mortality better than either of its components alone and is strongly associated with mortality on unadjusted analysis. However, it did not add significant discriminatory information to the APACHE II score, and it was not independently associated with mortality after adjustment for illness severity. LAR may be useful as a rapidly available early indicator where a full severity score is not yet computable, but on these data it does not complement APACHE II.