The lactate-to-albumin ratio for mortality risk stratification in sepsis-induced myocardial injury: a retrospective cohort study

Sepsis-induced myocardial injury (SIMI) is associated with a high fatality rate, yet specific prognostic indicators remain scarce. The lactate-to-albumin ratio (LAR), which integrates markers of tissue hypoperfusion and systemic inflammation, may offer valuable prognostic information. We aimed to assess the value of LAR for mortality risk stratification in SIMI. In this retrospective analysis, we extracted data from the MIMIC-IV database for 836 critically ill adults with SIMI, defined as cardiac troponin T > 0.01 ng/mL. LAR was calculated as serum lactate (mmol/L) divided by serum albumin (g/dL), using the first measurements obtained within 24 h of ICU admission. Patients were divided into four groups based on LAR quartiles. The primary endpoint was 28-day all-cause mortality; the secondary endpoint was 90-day mortality. Multivariable Cox proportional-hazards regression (reporting hazard ratios, HR) and logistic regression (reporting odds ratios, OR) were used to evaluate the association between LAR and mortality. Restricted cubic splines, receiver operating characteristic curves, and decision-curve analysis were employed to assess predictive performance. Of the 836 patients, 273 (32.7%) died within 28 days and 349 (41.7%) within 90 days. In the fully adjusted Cox model, the highest LAR quartile (Q4) versus the lowest (Q1) was associated with significantly increased mortality risk (HR = 3.331, 95% CI: 1.917–5.787 for 28-day; HR = 2.907, 95% CI: 1.855–4.556 for 90-day; both p < 0.001). Logistic regression yielded consistent findings (28-day OR = 4.78, 95% CI: 2.55–9.46; 90-day OR = 4.32, 95% CI: 2.43–7.89). Restricted cubic splines confirmed a graded, monotonically increasing dose-response relationship. In the test set, the multiple-marker model incorporating LAR demonstrated discriminative performance comparable to LAR alone (AUC: 0.694 vs. 0.693 for 28-day mortality) but consistently outperformed conventional severity scores. Decision-curve analysis showed that the multiple-marker model provided superior net clinical benefit across a wide range of threshold probabilities. Subgroup analyses revealed that the prognostic value of LAR remained robust across most clinical strata, including older age, liver disease, respiratory failure, and higher severity score categories. The lactate-to-albumin ratio is a robust, independent, and dose-dependent predictor of short-and medium-term mortality in patients with SIMI. Its performance is comparable to established ICU severity scores, and it provides additional net clinical benefit when incorporated into a multi-marker model. As a low-cost, readily available biomarker, LAR may facilitate early risk stratification and personalized management. Prospective studies are warranted to externally validate these findings and to determine whether LAR-guided strategies can improve clinical outcomes.