Admission Serum KL-6 Predicts Mechanical Ventilation Requirement and Mortality in Hospitalized Interstitial Lung Disease Patients: A Single-Center Prospective Study
摘要
Hospitalized patients with interstitial lung disease (ILD) are at high risk of requiring mechanical ventilation (MV), which is associated with substantial mortality. This study investigated the predictive value of admission serum KL-6, a marker of alveolar epithelial injury, for MV need and in-hospital mortality in this population.
MethodsWe conducted a single-center, prospective observational study of patients admitted with ILD. Serum KL-6 was measured at admission. The primary outcome was the requirement for MV during hospitalization, and the secondary outcome was all-cause in-hospital mortality. The predictive performance of KL-6 was assessed using receiver operating characteristic (ROC) curve analysis. Independent predictors were identified through multivariable logistic regression modeling.
ResultsAmong the 283 patients analyzed, 35 (12.4%) required MV. Admission serum KL-6 levels were significantly higher in patients who required MV compared to those who did not (median: 1420.5 vs. 839.8 U/mL, p < 0.001), and in non-survivors compared to survivors (1770.1 vs. 1068.0 U/mL, p < 0.01). KL-6 predicted the need for MV with a moderate area under the curve (AUC) of 0.70. The optimal predictive cut-off was 1016.5 U/mL. After multivariable adjustment for significant clinical variables, a KL-6 level ≥ 1016.5 U/mL remained an independent predictor of MV (adjusted odds ratios of 3.88 and 3.21 in two models, both p < 0.01). Furthermore, patients with high KL-6 levels (≥ 1016.5 U/mL) had significantly lower in-hospital survival (log-rank p < 0.001).
ConclusionAdmission serum KL-6 level serves as an independent biomarker for predicting the subsequent need for mechanical ventilation and is strongly associated with in-hospital mortality in patients hospitalized with ILD. Its measurement upon admission may facilitate early risk stratification and inform clinical management decisions.