How glycemic variability trajectories impact clinical outcomes in critically ill patients with severe pneumonia: a...

While glycemic variability (GV) affects outcomes in critically ill patients, its temporal patterns and prognostic implications in severe pneumonia remain unclear. This study aimed to identify distinct GV trajectory patterns and their associations with clinical outcomes.

This prospective cohort study enrolled 315 patients with severe pneumonia admitted to our intensive care unit (ICU) at Taicang Hospital of Nanjing University of Chinese Medicine, from January 2021 to December 2024. Using group-based trajectory modeling (GBTM), we analyzed the coefficient of variation (CV) trajectories during the first 5 ICU days. We compared clinical characteristics between trajectory groups, conducted exploratory landmark survival analysis, and assessed the performance of clinical parameters for identifying trajectory classification.

GBTM identified two distinct GV trajectories: Dynamic Variant (DV, 12.1%) characterized by initially high variability that gradually decreased but remained persistently elevated, and Stable Control (SC, 87.9%) with consistently low variability. While early outcomes were comparable, after day 14, the DV group demonstrated significantly higher mortality (70.3% vs. 60.0%, p = 0.034). DV patients had higher procalcitonin (PCT, 2.35 vs. 0.76 ng/ml, p = 0.030) and glycated hemoglobin A1c (HbA1c, 7.1% vs. 6.3%, p = 0.002) levels on ICU admission, despite similar disease severity scores and other clinical parameters. Combined prediction using PCT and HbA1c showed good performance in identifying trajectories (Area Under the Receiver Operating Characteristic curve: 0.87, 95% Confidence Interval: 0.80-0.94).

Distinct GV trajectories in severe pneumonia patients are associated with differential subsequent mortality risks, with admission PCT and HbA1c levels serving as predictors of trajectory group classification.

Not applicable.