A Single-Center Review of Cerebrospinal Fluid Drainage Complications after Open and Complex Endovascular Thoracoabdominal Aortic Repair

Use of cerebrospinal fluid drainage (CSFD) has become common practice for prevention and treatment of spinal cord ischemia/infarction (SCI) during thoracoabdominal aortic (TAAA) repair at most centers. However, CSFD has inherent risks that can contribute to morbidity and mortality in these patients. Here we present a contemporary single-center review of CSF drain-related complications at a tertiary medical center. Retrospective single-center review of all open and endovascular TAAA repairs who had a CSFD placed at our institution from 2017 to 2019. We included all indications including elective and emergent aneurysm, dissection, penetrating atherosclerotic ulcer, and IMH. CSFD complications were graded as mild (did not require intervention of prolong length of stay), moderate (did not require intervention but prolonged length of stay) and severe (required intervention, resulted in significant morbidity or mortality). A total of 97 patients who had 118 CSF drains placed were included in this analysis. Of the 118 drains, 102 (86%) were placed for endovascular repairs (TEVAR and complex EVAR), 17 of the 118 (14%) were placed for open TAAA and pararenal aortic repairs. The majority of drains (107/118; 90.6%) were preoperative prophylactic drains as our site utilizes a preoperative prophylactic CSFD policy, 11 of 118 (9.3%) were placed emergently as SCI postoperative therapeutic drains. In patients who had CSFD, our overall SCI rate was 5.1% with a temporary SCI (complete resolution) rate of 7.2% (7/97) and a permanent SCI (permanent paralysis) rate of 3.1% (3/97). Of those with permanent paralysis, 2 of 3 (66%) had a preoperative prophylactic drain and 1 of 3 (33%) had a postoperative therapeutic drain placed. CSFD-related minor complications (including headache [6/97] and epidural hematoma at entry site not requiring intervention [2/97]) incidence was 8.2% (8/97). Moderate CSFD related complications (including any neuroaxial bleeding not requiring intervention) was 4 of 97 (4.1%) with 3 patients having subarachnoid hemorrhage (SAH) and 1 patient with cerebellar hemorrhage. Severe CSFD-related complications (including major bleeding and lumbar drain damage requiring intervention) incidence was 5/97 (5.1%); with 2.1% (2/97) related to subdural hemorrhage (SDH) and 3.1% (3/97) related to retained lumbar drains. Both major SDH events were in patients who had preoperative prophylactic drains placed and both developed symptoms of SDH >48 hours after removal of CSF drain. We had no CSFD-related mortalities. Rates of major CSFD-related complications at our institution parallel that of previously published data. CSFD-related complications are not insignificant and should be taken into consideration when determining the need for a prophylactic CSF drain.