Association of Lactate‐to‐Albumin Ratio With 28‐Day Mortality in Critically Ill Patients With Obstructive Sleep Apnea: A Retrospective Cohort Study Based on the MIMIC‐IV Database
Jian Zhou, Chuntian Chi, Fanghao Yu, Junpei Wu, Hanjian LouBackground
Obstructive sleep apnea (OSA) is not uncommon among critically ill patients, especially in those with acute hypercapnic respiratory failure. Its high mortality rate remains a major challenge in the intensive care setting. The frequent apnea and respiratory failure during sleep in OSA patients, along with the intermittent hypoxemia caused by it, are accompanied by a series of systemic pathological and physiological changes. Metabolic imbalance is a key characteristic of the underlying pathological mechanism of this disease. However, the association between lactate‐to‐albumin ratio (LAR) and the prognosis of OSA patients remains to be fully elucidated. To date, a paucity of studies has evaluated the prognostic capabilities of these indicators. The majority of these studies have focused on the neutrophil‐to‐lymphocyte ratio (NLR) and platelet‐to‐lymphocyte ratio (PLR), and the results are contradictory, insufficient to fully clarify their clinical significance. Relying solely on lactate may not fully reflect patients’ physiological conditions, whereas integrating LAR can provide a more comprehensive and accurate description of the metabolic state. This study aims to explore the relationship between LAR and 28‐day mortality rates in ICU patients with OSA, using data from the MIMIC‐IV database.
Methods
Patients diagnosed with OSA were selected from the Medical Information Mart for Intensive Care IV (v.2.2) database, identified by specific ICD‐9 and ICD‐10 codes (detailed in the Methods section). To assess the predictive power of LAR, receiver operating characteristic (ROC) curve analysis, Kaplan–Meier (K‐M) survival curve analysis, logistic regression analysis, and subgroup analysis were employed. LAR was analyzed both as a categorical variable (based on the optimal cutoff) and as a continuous variable.
Results
The sample population comprised 807 patients diagnosed with OSA who were admitted to the intensive care unit for the first time. The 28‐day mortality rate was 20.44%. The area under the ROC curve (AUC) for LAR was 0.730 (95% CI: 0.686–0.772), which was comparable with OASIS (0.736; 95% CI: 0.695–0.776) and superior to SOFA (0.637; 95% CI: 0.591–0.683; DeLong test p = 0.500). The optimal cutoff value for LAR was 0.791. The high‐LAR group demonstrated a poor prognosis in the K‐M analysis ( p < 0.001). Multivariate logistic regression demonstrated a significant association between LAR and unfavorable outcomes, with an odds ratio (OR) of 3.00 (95% CI: 1.83–4.90, p < 0.001) when analyzed as a dichotomous variable, and an OR of 1.52 per unit increase (95% CI: 1.27–1.82, p < 0.001) when analyzed as a continuous variable. Subgroup analysis revealed no statistically significant interaction between LAR and any subgroup ( p for interaction: 0.052–0.906).
Conclusion
LAR is a rational and easily accessible biomarker that is significantly associated with 28‐day mortality in ICU patients with OSA. These findings are only applicable to OSA patients who received lactate and albumin testing during ICU admission, and the results cannot be generalized to the entire population of ICU patients with OSA.