A propensity score-matched comparison of deep versus mild hypothermia during thoracoabdominal aortic surgery.

OBJECTIVE: By using deep hypothermic circulatory arrest and non-deep hypothermic circulatory arrest approaches, we examined the impact of distal ischemia time and temperature on intra-abdominal reversible adverse outcomes and permanent adverse outcomes during descending thoracic aortic and thoracoabdominal aortic aneurysm operations. METHODS: A retrospective review of all patients who underwent descending thoracic aortic and thoracoabdominal aortic aneurysm repair between January 2002 and December 2008 was undertaken, including relevant preoperative, intraoperative, and postoperative data, and followed by a propensity score-matched analysis. Of the total of 262 patients, 240 had data complete enough to permit analysis, and 90 were suitable for the propensity-matched study. Reversible adverse outcomes included renal failure, liver failure, and temporary hemodialysis. Permanent adverse outcomes included paraplegia, permanent hemodialysis, and 30-day mortality. RESULTS: Thirty-day mortality was 7.1% (17/240). Overall, reversible adverse outcomes developed in 40.8% of patients and permanent adverse outcomes developed in 10% of patients. The propensity score analysis identified statistically significant decreased odds of developing reversible adverse outcomes in patients undergoing deep hypothermic circulatory arrest (odds ratio, 0.32; confidence interval, 0.12-0.85). Specifically, significantly lower rates of acute renal failure (22% vs 46.4%, P = .03) and liver failure (17.8% vs 34.3%, P = .04) were observed in the deep hypothermic circulatory arrest group compared with the non-deep hypothermic circulatory arrest group. In addition, there were decreased odds of reversible adverse outcomes (odds ratio, 0.22; confidence interval, 0.06-0.79) developing in patients with a stage II elephant trunk procedure. CONCLUSIONS: During descending thoracic aortic and thoracoabdominal aortic aneurysm repairs, the use of deep hypothermic circulatory arrest results in improved postoperative adverse outcome rates compared with non-deep hypothermic circulatory arrest techniques. The development of reversible adverse outcomes is strongly associated with the development of permanent adverse outcomes.

Staged repair significantly reduces paraplegia rate after extensive thoracoabdominal aortic aneurysm repair.

OBJECTIVE: Paraplegia remains a devastating, and still too frequent, complication after repair of extensive thoracoabdominal aortic aneurysms. Strategies to prevent ischemic spinal cord damage after extensive segmental artery sacrifice-or occlusion, essential for endovascular repair-are still evolving. METHODS: Ninety patients who underwent extensive segmental artery sacrifice (median, 13; range, 9-15) during open surgical repair from June 1994 to December 2007 were reviewed retrospectively. Fifty-five patients (mean age, 65 +/- 12 years; 49% were male), most with extensive Crawford type II thoracoabdominal aortic aneurysms, had a single procedure (single-stage group). Thirty-five patients (mean age, 62 +/- 14 years; 57% were male) had 2 procedures (2-stage group), usually Crawford type III or IV repair after operation for Crawford type I descending thoracic aneurysm. The median interval between the 2-stage procedures was 5 years (3 months to 17 years). There were no significant differences between the groups with regard to age, gender, cause of the aneurysm, hypertension, chronic obstructive pulmonary disease, urgency, previous cerebrovascular accidents, year of procedure, or cerebrospinal fluid drainage. In single-stage procedures, hypothermic circulatory arrest was used in 29% of patients, left-sided heart bypass was used in 40% of patients, and partial cardiopulmonary bypass was used in 27% of patients. Somatosensory-evoked potentials were monitored in all patients, and motor-evoked potentials were monitored in 39% of patients. Cerebrospinal fluid was drained in 84% of patients. RESULTS: Overall hospital mortality was 11.1%. There were no significant differences in mortality, stroke, postoperative bleeding, infection, renal failure, or pulmonary insufficiency between the groups. However, 15% of patients in the single-stage group had permanent spinal cord injury versus none in the 2-stage group (P = .02). The significantly lower rate of paraplegia and paraparesis in the 2-stage group occurred despite a significantly higher number of segmental arteries sacrificed in this group: a median of 14 (11-15) versus 12 (9-15) (P < .0001). CONCLUSION: A staged approach to extensive thoracoabdominal aortic aneurysm repair may reduce the incidence of spinal cord injury. This is of particular importance in designing strategies involving hybrid or entirely endovascular procedures.

Direct spinal cord perfusion pressure monitoring in extensive distal aortic aneurysm repair.

BACKGROUND: Although maintenance of adequate spinal cord perfusion pressure (SCPP) by the paraspinal collateral network is critical to the success of surgical and endovascular repair of descending thoracic and thoracoabdominal aortic aneurysms, direct monitoring of SCPP has not previously been described. METHODS: A catheter was inserted into the distal end of a ligated thoracic segmental artery (SA) (T6 to L1) in 13 patients, 7 of whom underwent descending thoracic and thoracoabdominal aortic aneurysm repair using deep hypothermic circulatory arrest. Spinal cord perfusion pressure was recorded from this catheter before, during, and after serial SA sacrifice, in pairs, from T3 through L4, at 32 degrees C. Somatosensory and motor evoked potentials were also monitored during SA sacrifice and until 1 hour after cardiopulmonary bypass. Target mean arterial pressure was 90 mm Hg during SA sacrifice and after nonpulsatile cardiopulmonary bypass, and 60 mm Hg during cardiopulmonary bypass. RESULTS: A mean of 9.8 +/- 2.6 SAs were sacrificed without somatosensory and motor evoked potential loss. Spinal cord perfusion pressure fell from 62 +/- 12 mm Hg (76% +/- 11% of mean arterial pressure) before SA sacrifice to 53 +/- 13 mm Hg (58% +/- 15% of mean arterial pressure) after SA clamping. The most significant drop occurred with initiation of nonpulsatile cardiopulmonary bypass, reaching 29 +/- 11 mm Hg (46% +/- 18% of mean arterial pressure) before deep hypothermic circulatory arrest. Spinal cord perfusion pressure recovered during rewarming to 40 +/- 14 mm Hg (51% +/- 20% of mean arterial pressure), and further within the first hour of reestablished pulsatile flow. Somatosensory and motor evoked potentials returned in all patients intraoperatively. Recovery of SCPP began intraoperatively, and in 5 patients with prolonged monitoring, continued during the first 24 hours postoperatively. All but 1 patient, who had remarkably low postoperative SCPPs and experienced paraparesis, regained normal spinal cord function. CONCLUSIONS: This study supports experimental data showing that SCPP drops markedly but then recovers gradually during the first several hours after extensive SA sacrifice. Direct monitoring may help prevent a fall of SCPP below levels critical for spinal cord recovery after surgery and endovascular repair of descending thoracic and thoracoabdominal aortic aneurysms.

Thoracic and thoracoabdominal aneurysm repair: is reimplantation of spinal cord arteries a waste of time?

BACKGROUND: The impact of different strategies for management of intercostal and lumbar arteries during repair of thoracic and thoracoabdominal aortic aneurysms (TAA/A) on the prevention of paraplegia remains poorly understood. METHODS: One hundred consecutive patients with intraoperative monitoring of motor evoked potentials (MEP) and somatosensory evoked potentials (SSEP) during TAA/A repair involving serial segmental artery sacrifice (October 2002 to December 2004) were reviewed. RESULTS: Operative mortality was 6%. The median intensive care unit stay was 2.5 days (IQ range: 1-4 days), and the median hospital stay 10.0 days (IQ range: 8-17 days). Potentials remained unchanged during the course of serial segmental artery sacrifice, or could be returned to baseline levels by anesthetic and blood pressure manipulation, in 99 of 100 cases. An average of 8.0 +/- 2.6 segmental artery pairs were sacrificed overall, with an average of 4.5 +/- 2.1 segmental pairs sacrificed between T7 and L1, where the artery of Adamkiewicz is presumed to arise. Postoperative paraplegia occurred in 2 patients. In 1, immediate paraplegia was precipitated by an intraoperative dissection, resulting in 6 hours of lower body ischemia. A second ambulatory patient had severe paraparesis albeit normal cerebral function after resuscitation from a respiratory arrest. CONCLUSIONS: With monitoring of MEP and SSEP, sacrifice--without reimplantation--of as many as 15 intercostal and lumbar arteries during TAA/A repair is safe, resulting in acceptably low rates of immediate and delayed paraplegia. This experience suggests that routine surgical implantation of segmental vessels is not indicated, and that, with evolving understanding of spinal cord perfusion, endovascular repair of the entire thoracic aorta should ultimately be possible without spinal cord injury.