September 1996

Penetrating Trauma of the Internal Carotid Artery

Author Affiliations

From the Division of Vascular Surgery, Department of Surgery, University of Southern California School of Medicine, Los Angeles.

Arch Surg. 1996;131(9):942-948. doi:10.1001/archsurg.1996.01430210040008

Objective:  To assess management of penetrating internal carotid artery (ICA) injuries.

Design:  Retrospective review of institutional protocol.

Setting:  Level 1 trauma center in a major urban area.

Patients:  Sixty-one patients with penetrating ICA injuries.

Interventions:  In the period 1975 to 1987 (group 1; n=36), management was based on individual surgeons' preferences. Between 1988 and 1995 (group 2; n=25), an algorithm was employed: (1) hemodynamically stable patients with suspected ICA injuries underwent a diagnostic angiography; (2) surgically accessible injuries were reconstructed regardless of neurologic status with 2 exceptions: (a) neurologically intact patients with ICA occlusion were treated by anticoagulation and mild pharmacological hypertension and (b) minimal nonocclusive injuries were managed nonoperatively and followed up by serial angiography or duplex ultrasonography; and (3) heparinization, shunting, and completion angiography were employed.

Main Outcome Measures:  Neurologic status at admission and discharge were compared by the Fisher exact test.

Results:  In group 1, 24 patients (67%) presented neurologically intact, and 12 (33%) with a deficit. Sixteen injuries were managed nonoperatively, 14 were repaired, and 6 were ligated. At discharge 6 (17%) were improved, 24 (66%) were unchanged, 6 (17%) were worse. Four patients (11%) died of cerebrovascular causes. In group 2, 19 patients (76%) presented neurologically intact, and 6 (24%) with a deficit. Eleven injuries were managed nonoperatively, 12 were repaired, and 2 were ligated. A death occurred in a patient who arrested, was admitted to the hospital in a coma, and died before ICA repair.

Conclusions:  Neurologic outcome after ICA injury is enhanced by an algorithm predicated on the liberal use of angiography, a predefined surgical approach, and selective observation.Arch Surg. 1996;131:942-948