Related Experiment Video
Updated: Mar 10, 2026
A Preclinical Controlled Cortical Impact Model for Traumatic Hemorrhage Contusion and Neuroinflammation
Published on: June 10, 2020
Clinical and Biochemical Predictors of 30-Day Mortality After Decompressive Craniectomy: A Retrospective Cohort Study
Ümit Akin Dere1, Yasemin Adali2,3, Batuhan Topal1
1Department of Neurosurgery, Pamukkale University, Faculty of Medicine, Denizli, Türkiye.
Background/Aim:
Decompressive craniectomy (DC) is a life-saving intervention performed in patients with severe neurological deterioration; however, early postoperative mortality remains high and reliable biochemical predictors are not well defined. This study aimed to identify early clinical and perioperative biochemical predictors of 30-day mortality in patients undergoing DC.
Patients And Methods:
This single-center retrospective cohort study included 102 patients who underwent DC between 2019 and 2024. Demographic, clinical, radiological, and biochemical parameters including sodium, potassium, hematocrit, international normalized ratio (INR), C-reactive protein (CRP), and neutrophil to lymphocyte ratio (NLR) were analyzed. The primary outcome was 30-day mortality. Univariable analyses were conducted using Chi-square and t-tests. Multivariable logistic regression models were constructed using a hierarchical approach to evaluate independent predictors. Model performance was assessed using likelihood ratio Chi-square, Hosmer-Lemeshow goodness-of-fit statistics, and area under the receiver operating characteristic curve (AUC).
Results:
Thirty-day mortality occurred in 70.6% of patients. In the multivariable model, preoperative INR >1.20 (OR=25.60; 95% CI=1.89-345.82; p=0.015) and discharge NLR >3.13 (OR=5.19; 95% CI=1.01-26.66; p=0.048) were identified as strong independent biochemical predictors of mortality. Conversely, preoperative GCS 8-13 was independently associated with improved short-term survival (OR=0.16; 95% CI=0.03-0.89; p=0.036). The predictive accuracy of the model significantly improved after adding biochemical variables, with Pseudo R² increasing from 0.115 to 0.252 and an AUC of 0.826, indicating good discrimination and calibration (Hosmer-Lemeshow p=0.765).
Conclusion:
Preoperative INR and discharge NLR are strong independent predictors of early mortality after decompressive craniectomy, while preoperative GCS remains a key indicator of improved survival. These routinely available biochemical markers may enhance postoperative risk stratification and inform clinical decision-making in critically ill neurosurgical patients.
