Evaluation of Determinants of Intraoperative Blood Loss During Craniotomy for Brain Tumors: A Retrospective Single-Center Cohort Study.

BACKGROUND: Inefficient blood product management in craniotomy for tumor removal poses clinical and economic burdens. This study evaluates predictive factors for significant intraoperative blood loss to establish a robust perioperative risk stratification model. MATERIALS AND METHODS: This retrospective cohort study examined adults undergoing elective brain tumor resections between 2020 and 2024 at a national tertiary referral center. Our primary outcome was significant blood loss, defined as an estimated loss of at least 20% of the patient's total blood volume (EBV). Independent predictors were evaluated using multivariate logistic regression, specifically backward-elimination methods. RESULTS: Out of 378 patients, 35% of patients experienced significant intraoperative blood loss. Bivariate analysis showed female gender, meningioma pathology, cerebral sinus involvement, tumor size ≥ 5 cm, and operation time ≥ 300 minutes associated with higher blood loss (p < 0.05). In the final multivariate model, tumor size ≥ 5 cm was the primary independent predictor (adjusted Odds Ratio [aOR] = 2.736; 95% Confidence Interval [CI]: 1.746-4.285; p < 0.001), followed by meningioma pathology (aOR = 2.104; 95% CI: 1.292-3.426; p = 0.003). Operation time ≥ 300 minutes in the final model (aOR = 1.703) was retained due to its clinical relevance in the operating room. CONCLUSION: Tumor size ≥ 5 cm and meningioma pathology were independently associated with significant intraoperative blood loss. Utilizing these factors as a preoperative screening tool allows clinical teams to anticipate significant blood loss, optimize cross-matching, and reduce institutional resource wastage.