Risk factors of rupture of internal carotid artery during surgical resection of carotid body tumor

Abstract

Objective: To investigate risk factors of rupture of internal carotid artery resection during carotid body tumor resection and to summarize our treatment experience. Methods: During the period from 1991 to 2016, rupture of internal carotid artery occurred in 27 patients (28 tumors) during surgical resection of carotid body tumor in the First Affiliated Hospital of Sun Yat-sen University. Their clinical and follow-up data were retrospectively collected and analyzed. For all patients underwent surgical resection during this period, Logistic regression analysis was used to investigate the risk factors of intraoperative rupture of internal carotid artery. Results: Of these 28 tumors, there were 15 (53.6%) tumors with diameter≥5 cm and 20 (71.4%) Shamblin Ⅲ tumors. Intraoperatively, shunt was applied for 8 (28.6%) cases. Thirteen (46.4%) patients underwent ligation of external carotid artery, while 2 (7.1%) patients accepted resection of cranial nerves. Direct closure/patchplasty, autologous vessels or graft reconstruction was used in 16, 10 and 2 cases, respectively. Postoperatively, stroke occurred in 4(14.3%) cases and cranial nerve deficit in 15 (53.6%) cases. During a median length of 36 (14-125) months, cranial nerve deficit persisted in 5 cases. Follow-up radiologic examination indicated 3 (10.7%) cases of targeted vessel occlusion. However, no new-onset stroke was identified. Among all patients underwent surgical resection of carotid body tumor, female (OR=3.650, P=0.012), age≤25 years old (OR=3.710, P=0.013) and Shamblin Ⅲ tumor (OR=4.631, P=0.008) increase the risks of intraoperative carotid artery rupture. Conclusions: Shamblin Ⅲ tumor is the predictor of rupture of internal carotid artery. Intraoperative, properly increased blood pressure, intraoperative heparinization and use of shunt for those cases without well-compensated cranial collateral arteries are likely to decreasing the incidence of stroke.