Association between Lactate dehydrogenase and 28-day all-cause mortality in patients with non-traumatic Intracerebral hemorrhage: A retrospective analysis of the MIMIC-IV database.

Abstract

Lactate dehydrogenase (LDH), a nonspecific inflammatory biomarker, has been used in the assessment of acute myocardial infarction, acute hepatitis, acute lung injury, and other severe diseases. However, no studies have evaluated the prognostic value of LDH in patients with non-traumatic intracerebral hemorrhage (ICH). This cohort study aims to assess the association between LDH levels and 28-day all-cause mortality in patients with non-traumatic ICH. Data for this retrospective cohort analysis were obtained from the MIMIC-IV (v2.2) database, and the study included patients with non-traumatic ICH as defined by the International Classification of Diseases, 9th and 10th editions. Patients were categorized into four distinct groups based on their LDH levels. The primary outcome of interest was the 28-day mortality rate. To analyze these associations and assess the consistency of interactions, subgroup analyses, Cox regression analysis, Kaplan-Meier (KM) curves, and nonlinear analysis were conducted. A total of 406 patients with non-traumatic ICH were enrolled in the study and were divided into quartiles based on LDH levels. The KM curve indicated that the 28-day all-cause mortality rate of patients in the Q4 group (LDH > 287.25) was significantly higher than in the Q1 (LDH < 194.7) (P < 0.001) and Q2 (194.7 < LDH < 233.0) (P < 0.001) groups, though not significantly different from Q3 (P = 0.140). Multivariate Cox proportional hazards analysis revealed that patients in the highest LDH quartile had a significantly increased risk of mortality compared to those in the lowest quartile across three models: unadjusted [HR, 3.401; 95% CI, 1.719-6.731; P < 0.001], partially adjusted [HR, 2.422; 95% CI, 1.211-4.846; P = 0.012], and fully adjusted [HR, 3.054; 95% CI, 1.522-6.126; P = 0.002]. Restricted cubic spline (RCS) models revealed an L-shaped association between LDH levels and the 28-day all-cause mortality rate, indicating a non-linear relationship (P < 0.001). No significant interactions were observed between LDH levels and other factors in the subgroup analyses (all P for interaction > 0.05). Our findings indicate a significant association between 28-day all-cause mortality and LDH levels in patients with non-traumatic intracerebral hemorrhage. Specifically, patients with elevated LDH levels within the first 24 hours of ICU admission are at a higher risk of mortality.