Early Factors Associated With Prolonged Intensive Care Unit Stay in Polytrauma Patients With Long-Bone Fractures: A Retrospective Cohort Study.

BACKGROUND

Polytrauma patients with long-bone fractures often require intensive monitoring because the clinical course is influenced by injury burden, hemorrhage, systemic inflammation, and operative timing. This study evaluated early clinical, physiologic, operative, and laboratory factors associated with prolonged intensive care in polytrauma patients with operatively treated long-bone fractures.

METHODS

This was a retrospective cohort study including 115 adult polytrauma patients with an index long-bone fracture treated with definitive fixation. Baseline injury characteristics, admission physiology, transfusion exposure, operative timing, and early laboratory values were extracted from routinely collected hospital records. The primary endpoint was prolonged intensive care unit (ICU) length of stay, defined as an ICU stay of five days or longer, corresponding to the cohort median. Secondary outcomes included ICU admission, deep surgical-site infection within 90 days, and reoperation by six months. Continuous variables were compared using parametric or nonparametric tests according to distributional characteristics, and categorical variables were compared using Pearson chi-square tests.

RESULTS

The cohort included 115 patients with a mean age of 46.9 ± 12.6 years; 86 patients (74.8%) were male. ICU admission occurred in 97 patients (84.3%), and prolonged ICU stay occurred in 62 patients (53.9%). Patients with prolonged ICU stay had higher Injury Severity Score (ISS) (28.1 ± 4.5 vs. 22.3 ± 3.7; p < 0.001), higher admission lactate (median 3.5 vs. 2.4 mmol/L; p < 0.001), greater transfusion exposure (median 2.0 vs. 1.0 red blood cell units; p < 0.001), more frequent shock on admission (46.8% vs. 11.3%; p < 0.001), and longer time to definitive fixation (median 33.1 vs. 25.9 h; p = 0.005). Early inflammatory markers were also higher in the prolonged ICU group, including IL-6 at 24 hours and 72 hours, and CRP at 24 hours and 72 hours. In multivariable analysis, ISS remained independently associated with prolonged ICU stay (adjusted OR 4.65 per five-point increase, 95%CI 2.06-10.50; p < 0.001).

CONCLUSIONS

In this retrospective cohort, injury severity remained the dominant independent variable associated with prolonged ICU stay among polytrauma patients with long-bone fractures. Admission shock, lactate, transfusion burden, operative timing, and early inflammatory markers were associated with prolonged ICU stay in univariable analyses. However, inflammatory biomarkers were not included in the primary adjusted model because of incomplete data and concerns regarding model stability. These findings suggest that injury severity remains the central determinant of prolonged ICU stay in this cohort, while admission physiology, transfusion exposure, operative timing, and inflammatory biomarkers may provide additional descriptive information about early clinical severity. Larger studies with externally validated prediction models are needed before these variables can be incorporated into a formal risk-stratification tool.