RÉSUMÉ
OBJECTIVE: This study examined the relationship between two new variables, tumor distance to base of skull (DTBOS) and tumor volume, with complications of carotid body tumor (CBT) resection, including bleeding and cranial nerve injury. METHODS: Patients who underwent CBT resection between 2004 and 2014 were studied using a standardized, multi-institutional database. Demographic, perioperative, and outcomes data were collected. CBT measurements were determined from computed tomography, magnetic resonance imaging, and ultrasound examination. RESULTS: There were 356 CBTs resected in 332 patients (mean age, 51 years; 72% female); 32% were classified as Shamblin I, 43% as Shamblin II, and 23% as Shamblin III. The mean DTBOS was 3.3 cm (standard deviation [SD], 2.1; range, 0-10), and the mean tumor volume was 209.7 cm3 (SD, 266.7; range, 1.1-1642.0 cm3). The mean estimated blood loss (EBL) was 257 mL (SD, 426; range, 0-3500 mL). Twenty-four percent of patients had cranial nerve injuries. The most common cranial nerves injured were the hypoglossal (10%), vagus (11%), and superior laryngeal (5%) nerves. Both Shamblin grade and DTBOS were statistically significantly correlated with EBL of surgery and cranial nerve injuries, whereas tumor volume was statistically significantly correlated with EBL. The logistic model for predicting blood loss and cranial nerve injury with all three variables-Shamblin, DTBOS, and volume (R2 = 0.171, 0.221, respectively)-was superior to a model with Shamblin alone (R2 = 0.043, 0.091, respectively). After adjusting for Shamblin grade and volume, every 1-cm decrease in DTBOS was associated with 1.8 times increase in risk of >250 mL of blood loss (95% confidence interval, 1.25-2.55) and 1.5 times increased risk of cranial nerve injury (95% confidence interval, 1.19-1.92). CONCLUSIONS: This large study of CBTs demonstrates the value of preoperatively determining tumor dimensions and how far the tumor is located from the base of the skull. DTBOS and tumor volume, when used in combination with the Shamblin grade, better predict bleeding and cranial nerve injury risk. Furthermore, surgical resection before expansion toward the base of the skull reduces complications as every 1-cm decrease in the distance to the skull base results in 1.8 times increase in >250 mL of blood loss and 1.5 times increased risk of cranial nerve injury.