Complications of Spinal Fluid Drainage in Thoracic and Thoracoabdominal Aortic Aneurysm Surgery in 724 Patients Treated From 1987 to 2013

Complications of Spinal Fluid Drainage in Thoracic and Thoracoabdominal Aortic Aneurysm Surgery in 724 Patients Treated from 1987 to 2013 Martha M. Wynn, MD,* Joshua Sebranek, MD,* Erich Marks, MD,* Travis Engelbert, MD,† and Charles W. Acher, MD† Objective: To study complications from spinal fluid drainage in open thoracic/thoracoabdominal and thoracic endovascular aortic aneurysm repairs to define risks of spinal fluid drainage. Design: Retrospective, prospectively maintained, institutionally approved database. Setting: Single institution university center. Participants: 724 patients treated from 1987 to 2013 Interventions: The authors drained spinal fluid to a pressure r6 mmHg during thoracic aortic occlusion/reperfusion in open and r8 mmHg after stent deployment in endovascular procedures. Low pressure was maintained until leg strength was documented. If bloody fluid appeared, drainage was stopped. Head computed tomography (CT) and, if indicated, spine CT and magnetic resonance imaging (MRI) were performed for bloody spinal fluid or neurologic deficit. Measurements and Main Results: Spinal fluid drainage was studied for bloody fluid, CT/MRI-identified intracranial and spinal bleeding, neurologic deficit, and death. Seventythree patients (10.1%) had bloody fluid; 38 (5.2%) had

intracranial blood on CT. One patient had spinal epidural hematoma. Higher volume of fluid drained and higher central venous pressure during proximal clamping were associated with intracranial blood. Most patients with intracranial blood were asymptomatic. Six patients had neurologic deficits: of the 6, 3 died (0.4%), 1 (0.1%) had permanent hemiparesis, and 2 recovered. Three of the six deficits were delayed, associated with heparin anticoagulation. Conclusions: 10% of patients had bloody spinal fluid; half of these had intracranial bleeding, which was almost always asymptomatic. In these patients, immediately stopping drainage and correcting coagulopathy may decrease the risk of serious complications. Neurologic deficit from spinal fluid drainage is uncommon (0.8%), but has high morbidity and mortality. & 2014 Elsevier Inc. All rights reserved.

E

thoracic aorta and visceral artery reimplantation. Spinal drains were placed in patients with acute presentation unless they were extremely unstable. All elective and acute patients having Z15 cm descending thoracic aortic coverage with TEVAR had SFD. Six cardiac anesthesiologists cared for all patients during the time of the study. All patients had standardized anesthetic and postoperative management. Anesthetic technique used fentanyl, benzodiazepines, amnestic volatile agent, low-dose naloxone, barbiturates, and moderate systemic hypothermia (33oC-34oC).18,19 Nitroglycerin, beta-blockers, dopamine, epinephrine, and norepinephrine were used as indicated for hemodynamic control. Standardized guidelines for mean arterial pressure (MAP), cardiac index, spinal fluid pressure (SFP), hemoglobin, and preventing coagulopathy were followed, with the goals of optimizing volume status, cardiac function, cardiac index and MAP; reducing tissue oxygen demand and increasing tissue oxygen delivery; maximizing direct and collateral network perfusion to the spinal cord; and preventing coagulopathy.20,21 TEVAR and open surgery patients received intravenous mannitol (before proximal clamping or graft deployment), and patients with Crawford type-II, III, and IV TAAAs also received cold crystalloid perfusion containing mannitol into the renal arteries after aortic occlusion. Fig 1 shows the timeline of observed/expected paralysis in the patient population (O/E ratios for paralysis)19,21 and changes in anesthetic and surgical techniques during the time of the study. Cardiac anesthesiologists placed all spinal drains in the operating room immediately before surgery. Abnormal preoperative coagulation parameters were corrected before drain placement, and after 2003, anesthesiologists followed the American Association of Regional Anesthesia Guidelines for Anticoagulation and Neuraxial Blocks22 in patients anticoagulated

XPERIMENTAL STUDIES in animal models of spinal cord ischemia have shown that spinal fluid drainage (SFD) reduces the risk of paraplegia in thoracic (TAA) and thoracoabdominal aortic aneurysm (TAAA) repair.1,2 Although only a small number of randomized clinical trials support the role of SFD in reducing paralysis in TAAA surgery,3–5 most centers reporting results of TAAA surgery use this adjunct.6–11 The risks associated with spinal fluid drainage in TAAA surgery are significant. Reported complications include drain failure,12 catheter fracture,13 headache,14,15 spinal fluid leak,14 infection,14 spinal and spinal epidural hematoma,15,16 intracranial bleeding (subdural,12,14,15 epidural,17 and intraparenchymal12), neurologic deficit,14,15 and death.12,14,15 These risks must be balanced against the benefit of SFD in reducing paralysis in TAA and TAAA repair. This retrospective study reports the incidence of complications associated with SFD in patients undergoing open TAA and TAAA surgery and thoracic endovascular aortic aneurysm repair (TEVAR) at a single institution from 1987 to 2013 to help define the morbidity and mortality associated with SFD. The authors hypothesized that the benefit of spinal fluid drainage in decreasing paralysis after TAAA repair justified the risks associated with this intervention. METHODS All patients treated for TAA and TAAA from 1987 to 2013 were analyzed retrospectively using a concurrently maintained, institutionally approved database to study the incidence of complications associated with SFD. Open TAA and TAAA surgeries were performed by 4 vascular surgeons using simple cross-clamp technique without assisted circulation or systemic heparinization. TEVAR patients were heparinized and activated clotting times of 200 to 250 seconds were maintained. Aneurysms involving the distal aortic arch were repaired using deep hypothermic circulatory arrest. Indications for SFD in patients having open surgery were TAA; Crawford types-I, II, and III TAAA; and Crawford type-IV TAAA, for which repair required clamping in the distal third of the descending

KEY WORDS: spinal fluid drainage, thoracic aortic aneurysm, thoracoabdominal aortic aneurysm, paraplegia, TEVAR

From the Departments of *Anesthesiology and †Surgery, University of Wisconsin School of Medicine and Public Health, Madison WI. Address reprint requests to Martha M. Wynn, MD, B6/319 UW CSC, 600 Highland Avenue, Madison, WI 53792-3272. E-mail: mmwynn@ facstaff.wisc.edu © 2014 Elsevier Inc. All rights reserved. 1053-0770/2601-0001$36.00/0 http://dx.doi.org/10.1053/j.jvca.2014.06.024

Journal of Cardiothoracic and Vascular Anesthesia, Vol ], No ] (Month), 2014: pp ]]]–]]]

1

2

WYNN ET AL

1 0.9 0.8 0.7 0.6 0.5 O/E 0.4 0.3 0.2 0.1 0

Intercostal artery ligation, mannitol, high MAP No morphine SFD

Hypothermia

Naloxone No nitroprusside Barbiturate burst suppression

Intercostal artery reimplantation

Fig 1. Timeline of O/E paralysis ratios19,21,62 (y axis) and changes in anesthetic management and surgical technique by year (x axis). E = [0.1
CI þ 0.2
CII þ 0.05
CIII þ 0.02
CIV þ 0.01
TAA] þ [0.3(acute þ dissection)], where CI, CII, CIII, and CIV are Crawford TAAA Types I to IV, TAA is thoracic aneurysm, and O/E ratio is the observed/expected paralysis in a patient population.19 Abbreviations: O, observed paralysis; E, expected paralysis; MAP, mean arterial pressure; SFD, spinal fluid drainage. (Color version of figure is available online.)

preoperatively. Thrombin inhibitors that became available during the years of the study were held for a minimum of 2 to 3 half-lives before spinal drain placement in elective patients, depending on drug metabolism and excretion, renal function, and age. Dabigatran was discontinued 3 to 6 days before surgery, depending on age and renal function.23 Before 2000, 19-gauge ARROW Epidural Catheterization Kit (Reading, PA) (through a 17-gauge Tuohy needle), placed using anatomic landmarks, were used for SFD. Since 2000, 16-gauge Medtronic EMD Lumbar Drains (Minneapolis, MN) (through a 14-gauge Tuohy needle), positioned using fluoroscopy with the catheter tip at T9-10, have been used for SFD. To minimize trauma from needle placement, a 22-gauge finder needle and fluoroscopy were used to position the 14-gauge needle. Because of time urgency, acute patients had drain placement using anatomic landmarks. Although the catheter and method of placement changed during the time of the study, drain protocols and management were the same. Protocol required sterile drain placement (in the operating room with gown and gloves) and manually draining spinal fluid to gravity in increments of 5 to 10 mL to achieve an SFP r6 mmHg during thoracic aortic occlusion and reperfusion in open procedures and r8 mmHg during and after endograft deployment in TEVAR. Guidelines did not specify a limit on total volume drained or volume drained/hour. Low SFP goals were chosen to ensure a greater difference between spinal cord collateral network pressure and SFP, because it was observed that distal mean arterial pressure during aortic occlusion was o30 mmHg. Nonheparinized, nonpressurized transducers were zeroed at the level of the right atrium and SFP was monitored continuously, except when manually draining. Following surgery, SFP was kept at 6 to 8 mmHg until normal leg strength was observed, usually within 6 hours of surgery. After normal leg strength was observed, spinal fluid pressure was monitored but fluid was not drained unless leg weakness occurred. Thus, very little drainage occurred postoperatively in patients with intact leg strength. If any trace of blood appeared in the fluid during SFD (aside from the initial blood tinge that sometimes occurs at placement), SFD was stopped

immediately. Head computed tomography (CT) and, if indicated, spine CT or magnetic resonance imaging (MRI) were performed in patients with bloody spinal fluid or neurologic deficit. If head CT was negative, SFD sometimes was continued, but to a higher goal pressure in the highest-risk patients, until normal leg strength was demonstrated. Drainage was stopped in low-risk patients even if head CT was negative. Spinal fluid drainage was stopped if head CT showed intracranial blood. Spinal drains were removed 48 hours after surgery if leg strength was normal. Patients with headache were treated with epidural blood patch. Persistent fluid leak was treated with epidural blood patch or skin suture at the puncture site. If delayed paresis/paralysis occurred, the spinal drain was reinserted and SFD resumed until neurologic function stabilized. Patient demographic characteristics, intraoperative hemodynamic variables, aortic occlusion time, spinal fluid pressures at baseline and during aortic occlusion, reduction in spinal fluid pressure from baseline, volume of spinal fluid drained, surgical blood loss, and complications associated with spinal fluid drainage were analyzed. The incidence of being unable to place spinal drains, spinal drain failure (not being able to drain enough fluid to achieve target SFP), catheter fracture with retained intrathecal fragment, headache and epidural blood patch, spinal fluid leak, infection, bloody spinal fluid (any blood tinge occurring during SFD, other than the transient blood tinge that sometimes occurs with placement), intracranial and spinal bleeding found by CT or MRI, neurologic deficit, and death were examined. Statistical analysis used SAS-JMP software for univariate analysis and multivariate modeling. Fisher’s exact test, Pearson chi-square test, and one-way analysis of variance were used to evaluate for significance. Variables significant on univariate analysis were chosen for multivariate modeling. RESULTS

One thousand patients had open TAA/TAAA repair and TEVAR between 1987 and 2013. Seven hundred twenty-four

3

COMPLICATIONS OF SPINAL FLUID DRAINAGE

Table 1. Volume of Spinal Fluid Drained and Spinal Fluid Blood, Intracranial Blood on CT, and Neurologic Deficit in Open TAAA and Thoracic Endovascular Aortic Surgery SFD, SFB, ICB, neurologic deficit

Mean SFD, mL SFB Mean SFD (mL) in patients with SFB/mean SFD (mL) in patients with no blood ICB on CT Mean SFD (mL) in patients with ICB/mean SFD (mL) in patients with no blood Patients with neurologic complications Mean SFD (mL) in patients with neurologic complications/ mean SFD (mL) in patients with no neurologic complications

Open

133 (95% CI 129-137) 62/622 (9.9%) 152 (CI 140-164)/131(CI 126-135) (p ¼ 0.0009) 34/622 (5.5%) 170 (CI 154-187)/131(CI 127-134) (p o 0.0001) 5/622 (0.80%) 186 (CI 147-225)/132 (CI 128-136) (p ¼ 0.0070)

TEVAR

82 (95% CI 73-92) 11/102 (10.8%) 101( CI 81-122)/80(CI 73-87) (p ¼ 0.0517) 4/102(3.9%) 132 (CI 99-165)/80(CI 74- 87) (p ¼ 0.0027) 1/101 (0.98%) 30*/83 (CI NA) (p ¼ 0.1263)

p Value

o0.0001 0.7996

0.5167

0.8554

Abbreviations: CI, confidence interval; CT, computed tomography; ICB, intracranial blood; NA, not applicable; SFB, spinal fluid blood; SFD, spinal fluid drained; TEVAR, thoracic endovascular aortic aneurysm repair. * The neurologic complication in the TEVAR patient was leg weakness from lumbar spine hematoma and not from intracranial hemorrhage.

patients had spinal drains. Spinal drains were placed in 76% of open surgery patients and 57% of TEVAR patients. Of the 724 patients who had spinal fluid drainage, 150 (21%) had TAA and 574 (79%) had TAAA repair. Six hundred twenty-two patients (86%) had open repair, and 102 (14%) had TEVAR. Fifty-nine percent (206 of 347) of acute patients had SFD. Mean volume of spinal fluid drained was 127 mL (median 122 mL, range 25-424 mL, 95% confidence interval [CI] 123.11/ 130.94). Thirty-two of the 724 patients (4.4%) died before all postoperative complications could be determined, but bloody spinal fluid was assessed in these patients. Spinal drains could not be placed in 4 patients (0.55%). One hundred forty-nine complications occurred in 108 patients (14.9%). There were 29 (4%) drain failures. Drain failure occurred in 25 of 322 patients (7.76%) with small drains placed using anatomic landmarks and in 4 of 402 patients (0.99%) with large drains placed using fluoroscopy; this difference was significant (odds ratio [OR] 6.633; 95% CI 2.510-17.531, p o 0.0001). Spinal fluid leak occurred in 32 patients (4.42%) and was treated with puncture site suture in 10 patients (31%) and epidural blood patch in 22 patients (71%). Twenty-six patients (3.6%) had headache, and 22 (85%) of these were treated with epidural blood patch. Patients with headache were younger (56 v 68 years; p o 0.0001) and more likely to be male (20 of 385 [5.2%] male v 6 of 307 [1.95%] female; p ¼ 0.0270). Spinal fluid pressure and volume drained did not correlate with headache. One EMD drain fractured, and a small silastic fragment was retained in the intrathecal space without sequelae. There were no infections. Blood appeared in the spinal fluid of 73 of 724 (10.1%) patients who had SFD. The initial appearance of blood was almost always intraoperative and occurred after visceral reperfusion. Thirty-two of the 73 (43.8%) patients with bloody spinal fluid had intracranial blood on CT and 41(56.2%) did not. An additional 6 patients who did not have bloody spinal fluid had intracranial blood on CTs done because of neurologic deficit or persistent headache after spinal drain removal. Thus, 38 of 724 (5.2%) patients had intracranial blood on CT after SFD. Four of the patients with intracranial blood after SDF had TEVAR.

On univariate analysis, higher volume of spinal fluid drained (144 v 127 mL, p ¼ 0.0013) in all patients (TEVAR and open repair) and higher mean arterial pressure at baseline after induction of anesthesia (92 v 87 mmHg, p ¼ 0.0123) and after renal reperfusion (89 v 83 mmHg, p ¼ 0.0009) in patients having open repair were significant for bloody spinal fluid. Only volume of spinal fluid drained (144 mL v 124 mL, OR 1.0074/mL drained, 21.35 over range, p ¼ 0.0029) and reperfusion MAP (OR 1.0319/mmHg, 14.975 over range, p ¼ 0.0009) were significant for bloody spinal fluid on multivariate analysis (Table 1 and Fig 2). In the patients with intracranial blood on CT, baseline MAP, MAP during cross-clamp, and central venous pressure (CVP) during cross-clamp in open patients, and volume of spinal fluid drained in both open and TEVAR patients were significant factors on univariate analysis. In multivariate modeling, only CVP during cross-clamp in open patients (20.3 v 16.9 mmHg, OR 1.076/mmHg/14.20 over range, p ¼ 0.0149) and volume of spinal fluid drained in all patients (166 mL v 127 mL, OR 1.012/mL removed, 119.72 over range, p ¼ 0.0006) remained significant for intracranial blood (Table 1 and Fig 2).

Fig 2. Volume (mL) of spinal fluid by no blood, bloody spinal fluid, intracranial blood, and neurologic complication with deficit. Green horizontal lines indicate mean and confidence intervals. Abbreviations: SFD, spinal fluid drainage; NB, no blood, SFB, bloody spinal fluid, ICB, intracranial blood; NC, neurologic complication with deficit. (Color version of figure is available online.)

4

WYNN ET AL

Table 2. Complications by Drain Type Complications

Drain failure Spinal fluid drained (mL) Headache SF leak SF blood Intracranial blood on CT Neurologic complications

19-Gauge Catheter

25/322 126 1/322 2/321 15/322 11/322 2/322

(7.76%) (95% CI 120-132) (0.31%) (0.62%) (4.66%) (3.42%) (0.62%)

16-Gauge Catheter

4/402 126 25/399 30/399 58/402 27/402 4/402

(0.99%) (95% CI 120-131) (6.26%) (7.52%) (14.43%) (6.72%) (0.99%)

OR (95% CI)

p Value

6.633 (2.510-17.531) N/A 21.457 (2.891-150.237) 12.967 (3.075-54.687) 3.481 (1.933-6.269) 2.052 (1.002-4.203) 1.621 (0.295-8.903)

o0.0001* 0.9233 o0.0001* o0.0001* o0.0001* 0.0453* 0.5752

Abbreviations: CI, confidence interval; CT, computed tomography; N/A, not applicable; OR, odds ratio; SF, spinal fluid.

None of the patients with intracranial hemorrhage on CT had neurosurgical intervention. Most of these patients had small asymptomatic bleeds that neurosurgeons evaluated with serial examination and head CT. In the few patients with neurologic deficit from large intracranial hemorrhage, neurosurgeons decided there would be no benefit from surgical intervention. One TEVAR patient who did not have bloody spinal fluid or intracranial blood on head CT developed a delayed epidural hematoma at the site of needle placement in the lumbar spine and had surgical evacuation. Six of the 724 SFD patients (0.8%) had neurologic deficits associated with SFD: 5 had large subdural hematomas, 1 had a moderate-size intraparenchymal bleed, and 1 had an epidural hematoma at the site of needle insertion in the lumbar spine. Patients with neurologic deficits from intracranial hemorrhage after SFD had a larger volume of spinal fluid drained (186 v 133 mL, 95% CI 147-225/133 128-136, p ¼ 0.0070) (Table 1 and Fig 2). The incidence of death or permanent injury in the patients with neurologic deficits was 67%. Death associated with SFD occurred in 3 patients (0.4%), and 2 of the deaths occurred 48 hours or more after spinal drain removal following the administration of low-molecular-weight heparin (LMWH). Of the remaining 3 patients with neurologic deficits associated with SFD, 1 had permanent hemiparesis (0.1%) and 2 recovered (one recovered after emergent laminectomy to evacuate a delayed lumbar spine hematoma, also associated with LMWH, and the patient with intraparenchymal cerebellar bleed and ataxia also recovered). The 3 patients with neurologic deficit who received LMWH 24 hours after spinal drain removal and suffered intracranial or neuraxial bleeds in the 24 to 36 hours following its administration all had uncomplicated SFD and drain removal 48 hours after surgery and accounted for half the neurologic complications associated with SFD. These were the only patients who received postoperative LMWH. There was no statistical difference between TEVAR and open surgery in occurrence of bloody spinal fluid, blood on head CT, or neurologic complications, even though TEVAR patients had significantly less spinal fluid drained (Table 1). Comparing complications associated with spinal fluid drainage with the small epidural catheters and the large EMD catheters, there was a higher incidence of headache (OR 21.457), spinal fluid leak (OR 12.967), bloody spinal fluid (OR 3.481), and intracranial blood (OR 2.052) with the large EMD catheters. There was no statistical difference between catheters in amount of spinal fluid drained or neurologic deficit (OR 1.621) (Table 2).

DISCUSSION

Rationale for Spinal Fluid Drainage Draining spinal fluid to reduce spinal fluid pressure has been shown to reduce paralysis in experimental models of spinal cord ischemia1,2 and in clinical studies of patients having TAAA surgery.4,5 Two randomized controlled trials of SFD in high-risk TAAA patients showed that SFD reduced paralysis. In 1998, Svensson et al reported 12.5% paralysis in patients treated with the combination of SFD and papaverine compared with 43.75% paralysis in control patients treated without SFD and papaverine.4 In 2002, Coselli et al reported 2.9% paralysis in patients treated with SFD compared with 11.8% paralysis in patients treated without SFD.5 Until this time, most centers doing TAAA surgery did not use SFD and were reporting 8% to 22% paralysis.19,24–27 Despite the small number of patients in both studies and the confounding addition of papaverine in the Svensson et al trial, after 1995 most high-volume centers began using SFD in TAAA surgery. Over the next 20 years, paralysis risk improved significantly in all centers.6,10,21,28–32 Aortic occlusion and replacement by graft during TAAA surgery and extensive aortic endograft coverage during TEVAR interrupt segmental arteries, reduce direct arterial blood flow to the spinal cord, and cause spinal cord ischemia.33 Proximal aortic clamping increases CVP and SFP.33 Because spinal cord perfusion pressure is the difference between the mean arterial pressure in the collateral network circulation to the spinal cord34,35 and the spinal fluid pressure, draining spinal fluid reduces spinal fluid pressure, improves spinal cord perfusion pressure,33 and may increase oxygen delivery to the spinal cord. Although SFD is beneficial in reducing paraplegia in TAAA surgery, the significant risks of this protective adjunct must be balanced with its benefit. Volume or Pressure Endpoint for Spinal Fluid Drainage? Some groups treating TAAAs drain to a volume endpoint and specify a volume limit and maximum volume drained per hour for intraoperative SFD.14,36,37 Protocol at the authors’ institution is to drain to a pressure endpoint that is a lower pressure than the pressure used by many groups.15,36,38 A goal SFP of r6 mmHg during aortic occlusion and reperfusion usually requires draining 110 mL to 120 mL of spinal fluid, an observation that suggests the 120-mL volume limit used by some groups for intraoperative drainage reliably reduces SFP to o10 mmHg in most patients. The authors advocate a somewhat flexible approach to volume and pressure limits for intraoperative and postoperative

5

COMPLICATIONS OF SPINAL FLUID DRAINAGE

SFD. That said, because the total volume of spinal fluid produced in 24 hours is 400 mL to 600 mL and the average circulating volume is only 140 mL to 165 mL,39 SFD must occur within these constraints. Spinal fluid is produced even during SFD and replenishes circulating volume at any time. The risk of intracranial bleeding increases when volume drained approaches the upper range of circulating volume. The authors believe that the appearance of bloody spinal fluid during SFD signals that the volume drained is in the lower range of circulating spinal fluid volume and is a sensitive indicator of increased risk of intracranial bleeding. Volume drained depends on rate of cerebrospinal fluid production, length of surgery, management of increased preload during proximal clamping, and reperfusion edema after clamp release, whether achieving a target pressure is a goal and whether reaching the target pressure can be timed to coincide with aortic occlusion. There is a physiologic reason to drain to a target SFP, knowing that the target pressure may vary with paralysis risk determined by extent of aneurysm and acuity, with systemic arterial pressure, with clinical or neurophysiologic evidence of spinal cord ischemia, and with clinical setting. Studies show that at baseline the mean arterial pressure in the spinal cord collateral network circulation is about 75% of systemic mean arterial pressure. Collateral network pressure decreases with aortic occlusion, falling to about 40% of systemic pressure during assisted circulation (left-heart bypass), and begins to recover in the 24 hours after repair.34 Increasing MAP and decreasing SFP during this vulnerable period increases spinal cord perfusion pressure.34 Many groups drain intraoperatively to a pressure o10 mmHg.3,15,36 In the present patients, the average baseline SFP was 16.5 mmHg (median 16, SD 6.47), so draining to a pressure of 10 mmHg represented only a 39% reduction in SFP. If systemic MAP is 70 mmHg and collateral network MAP is 40% of systemic MAP (in this case o30 mmHg) during assisted circulation, an SFP of 10 mmHg may not ensure adequate spinal cord perfusion pressure. Paralysis outcomes in TAAA surgery are determined by the interaction of multiple risk factors (extent of aneurysm, dissection, acuity25,40–42) and neuroprotective interventions (hypothermia, increasing arterial pressure, reimplanting segmental arteries in the highest risk patients, SFD43). Spinal fluid drainage is only 1 intervention in a multimodal approach to reducing spinal cord ischemia. Specific pressure or volume limits serve as safety guidelines to reduce the risk of complications associated with SFD. However, the pathophysiology of spinal cord injury in TAAA surgery is complex and multifactorial and applying physiologic neuroprotective adjuncts to changing clinical situations in an adaptable way is critical to paralysis risk reduction. COMPLICATIONS OF SPINAL FLUID DRAINAGE

Drain Failure Drains could be placed in all but 4 (0.55%) patients who were to be treated with SFD. In 2000, the authors switched to 16-gauge EMD drains because of the unacceptable intraoperative failure rate with 19-gauge epidural catheters (7.76%) and since then have used fluoroscopy for drain placement in elective patients. Fluoroscopy is helpful in patients with anatomic abnormalities of the spine, severe osteoarthritis,

lumbar spine hardware, or posterior lumbar spine fusion. Knowing the exact location of the catheter is also useful when troubleshooting drain failure. Three of the 4 failed EMD drains had inadequate intraoperative drainage but drained adequately after surgery, and no EMD drains were replaced because of failure. Drain Infection and Fracture The risk of infection is extremely low when spinal drains are placed in the operating room using sterile technique.12,14,15 The authors have seen no infections from SFD in 724 patients treated over 25 years. However, high mortality is reported from the rare infection associated with SFD.14 There is 1 report in the neurosurgical literature indicating meningitis risk associated with lumbar catheters increases if catheters remain in place for a longer period of time.44 Intrathecal drain fracture is also rare, but can sometimes require surgery to remove the retained portion of the catheter.13,14 Headache and Persistent Fluid Leak The most common complications associated with SFD come from intracranial hypotension produced when cerebrospinal fluid is removed.45 Intracranial hypotension causes headache from tension on the sensory receptors in the dural sinuses.46 The incidence of headache in the present study (3.6%) was higher than that reported by other groups using a similar size needle and catheter;14,15,37 this could have been because of draining to a lower SFP. Younger patients and male patients were more likely to have headache. Headache was more frequent with the 16-gauge catheter. Headache was treated with epidural blood patch because early on the authors observed that conservative therapy for headache usually was unsuccessful and prolonged length of stay, whereas epidural blood patch always was successful. Persistent spinal fluid leak also was more frequent with the larger catheters and was treated with epidural blood patch or suture at the leak site. Although bloody spinal fluid and intracranial blood on head CT were more likely with the larger drains, there was no difference between the large and small drains in neurologic deficit or death associated with SFD. Because the large catheters have a much lower failure rate, they are preferred in most patients. Neuraxial Hemorrhage Neuraxial bleeding, a far more serious complication associated with SFD, can cause neurologic deficit. Lumbar spine hematoma is a known complication of neuraxial anesthetic techniques associated with difficult or traumatic needle placement, needle size, and anticoagulation.47 Neuraxial bleeding presents with back pain and lower extremity weakness. However, lumbar spine hematoma associated with SFD in TAAA surgery, although reported, is very rare.15,16,36 Because of this, lower extremity weakness after TAAA surgery or TEVAR initially must be assumed to be from spinal cord ischemia and immediate measures taken to treat ischemia because the window for rescue is narrow. However, if SFD is used during surgery, the very rare risk of epidural hematoma at the site of needle insertion as a cause of lower extremity weakness always must be considered and ruled out with spinal

6

MR imaging as soon as rescue measures to treat spinal cord ischemia have been initiated. Intracranial Hemorrhage By far the most frequent serious complication associated with SFD in TAAA surgery is intracranial bleeding14,15,36,48 from intracranial hypotension. Intracranial hypotension produces engorgement in the dural venous sinuses49 and caudad displacement of the brain,50 creating tension on the sinuses46 and stretching fragile bridging veins inside the cranium.51 Excessive stretch from high volume or very rapid removal of fluid causes bleeding. The association between CT-identified intracranial blood and higher CVP during aortic occlusion in the present patients may reflect the increased likelihood of tearing intracranial bridging veins when higher CVP during aortic occlusion increases venous engorgement. This association was first reported in 2009,12 was confirmed in the current analysis, and probably is clinically significant. Retrospective studies from centers doing TAAA surgery reported 0.3% to 5.5% incidence of intracranial hemorrhage with neurologic complications and 0.2% to 0.45% mortality from SFD.14,15,36,48 The present results of 0.8% neurologic complications and 0.4% mortality associated with SFD were consistent with high-volume centers performing TAAA surgery. In retrospective reports, higher volume of fluid drained intraoperatively (178 mL v124 mL) and higher total volume of fluid drained (690 mL v 359 mL) were associated with greater risk of bloody spinal fluid and CT-identified intracranial bleeding,12,14,36 and the present analysis was consistent with other reports. Ten percent of the present patients had bloody spinal fluid. The initial appearance of blood intraoperatively after visceral reperfusion suggested that reperfusion hyperemia and edema may play a role. The explanation for the association of bloody spinal fluid with higher arterial blood pressure after reperfusion in the present analysis was not clear, because venous intracranial bleeding was hypothesized as the source of bloody spinal fluid and intracranial bleeding. The difference in MAP in the present analysis, although statistically significant, was small and judged to be clinically insignificant. Because intracranial hemorrhage was suspected in patients with bloody spinal fluid intraoperatively, coagulopathy proactively was prevented or corrected. The first appearance of bloody fluid in the present group of patients occurred after a considerable volume of fluid already had been drained, and draining fluid was stopped if blood appeared. Over time, it was observed that almost half the patients with bloody spinal fluid had intracranial blood on head CT. Because the risk of producing neurologic deficit from continued SFD in the presence of intracranial hemorrhage exceeds the benefit of SFD in reducing paralysis, spinal fluid drainage always was stopped if head CT showed intracranial blood. It is possible that the number of patients with intracranial blood and neurologic deficit associated with SFD would have been higher if patients were allowed to become coagulopathic or SFD was continued after bloody spinal fluid appeared. In the present analysis, patients with bloody spinal fluid and intracranial hemorrhage had a higher volume of spinal fluid removed, and in those with neurologic deficits, mean volume drained exceeded the average circulating cerebrospinal fluid volume.

WYNN ET AL

More than 80% of the patients who had CT evidence of intracranial hemorrhage were asymptomatic. These bleeds go unrecognized when head CT is performed only in symptomatic patients. Continued drainage in patients with bloody spinal fluid with unrecognized intracranial blood may increase the size of hemorrhage and produce neurologic deficit. Although neurologic deficit-associated SFD was uncommon (0.8%) in the present group of patients, the associated mortality and morbidity were very high. Because half the neurologic complications occurred only in the patients who received postoperative LMWH, the authors were very attentive to the risks of postoperative anticoagulation even after drains were removed and did not administer LMWH after SFD. SPINAL FLUID DRAINAGE IN TEVAR

The effectiveness of SFD in preventing paralysis in TEVAR is less clear than with open surgery. Extensive endograft coverage of the descending thoracic aorta permanently interrupts segmental arteries that supply direct arterial flow to the spinal cord circulation, and spinal cord ischemia results when the collateral network circulation cannot compensate for this reduced flow. Thus, paralysis can, and does, occur without aortic occlusion; high-volume centers report 2% to 8% paralysis after TEVAR.52–55 Some retrospective reports suggested paralysis risk is increased in patients who have longer aortic coverage with endograft and those with previous aortic repair.54,56 The use of SFD in TEVAR varies among centers. Centers reporting TEVAR results use SFD in 1 of 3 ways:57 always when most of the descending thoracic aortic is covered,58 selectively in patients judged high risk for paralysis,54,57,59 and as a rescue measure when paresis or paralysis occur after TEVAR.60,61 One retrospective study concluded that SFD in TEVAR had acceptable risk but that restricting its use in high-risk patients did not increase paralysis compared with patients who did not have SFD.54 However, the groups with and without SFD were not comparable; half the patients in the SFD group had hybrid repair of Crawford type-I, II, and III TAAAs, more extensive aneurysms with much higher paralysis risk than simple descending thoracic aortic aneurysms. In the present group of patients, the risk of headache, bloody spinal fluid, and neurologic complications associated with SFD were similar in TEVAR and open TAAA repair, even though TEVAR patients had lower volume drainage. This could have been related to the routine use of heparin in TEVAR but not in open surgery, or the generally faster rate of drainage in TEVAR, and 1 of 102 (0.98%) of the TEVAR patients had a neurologic complication associated with SFD. Randomized controlled trials of SFD, done in patients with comparable aneurysm extent and comorbidities, are needed to study the benefit and define the indications for SFD in TEVAR. Such trials may best be performed by centers currently using SFD for rescue after TEVAR. RISK/BENEFIT EVALUATION OF SPINAL FLUID DRAINAGE IN TAAA SURGERY

Spinal fluid drainage, although effective in reducing paralysis in TAAA surgery, has serious risks. However, most groups judged the benefit of SFD in reducing paralysis in

7

COMPLICATIONS OF SPINAL FLUID DRAINAGE

Table 3. Paralysis Without and With Spinal Fluid Drainage Series

Hamilton & Cox 199263 Svensson et al 199325 Acher et al 199018 Acher et al 199419 Bavaria et al 199526 Griepp et al 199627 Acher et al 199419 Jacobs et al 200328 Safi et al 200319 Etz et al 200629 Jacobs et al 200630 Coselli et al 200731 Acher et al 201032 Acher et al 201032 Lima et al 201210

N

Spinal Fluid Drainage

% Paralysis

129 1509 24 49 81 138 61 188 1004 100 112 2286 750 261 251

N N N N N N Y Y Y Y Y Y Y Y Y

19 16 29 22.4 13.6 7.97 1.6 2.66 3.98 2.0 4.46 3.81 3.87 2.30 5.18

TAAA surgery to more than outweighed the risks of serious neurologic complications and death from this invasive intervention. The recent review of adjuncts used to reduce paralysis after TAAA surgery by Augoustides et al concluded that the morbidity associated with paralysis more than balanced the risks of spinal fluid drainage and other interventions used in open repair of these complex aneurysms.43 The authors’ experience supported this conclusion, and in open TAAA surgery and TEVAR, SFD was used whenever the risk of paralysis without SFD was judged to exceed the risk of serious complications associated with SFD. Paralysis in patients treated from 1983 to 1986, without any spinal cord protective interventions (no hypothermia, hemodynamic optimization, or SFD), was 28% in all patients.18,19 Since 1987, paralysis in patients who had hypothermia, hemodynamic optimization, and SFD was 3.25% in elective patients, 8.2% in acute patients, and 4.65% overall.21,62 There was no difference between these 2 groups in age, extent of aneurysm, presence of dissection, or acuity. Table 3 shows the percentage of paralysis with and without SFD, reported by centers doing TAAA surgery. Table 4 shows the percentage of serious neurologic complications and death associated with SFD. It is clear that, in open TAAA surgery, the benefit of SFD in reducing paralysis more than justifies the risk of serious neurologic complications and death associated with SFD. Because the effect of SFD on paralysis in TEVAR has not been determined, the risk/benefit judgment about SFD is less clear in TEVAR. It is important to compare equivalent risk

Table 4. Neurologic Complications and Death Associated With Spinal Fluid Drainage

Series

Weaver et al 200116 Dardik et al 200236 Estrera et al 200914 Youngblood et al15 Wynn et al 2014 (present study)

% Neurologic

%

N

Complications*

Death

62 230 1107 504 724

3.2 3.5 0.6 2.8 0.8

0 1.7 0.3 0.2 0.4

*Neurologic complications refers to deficit from intracranial bleeding, deficit from neuraxial hematoma, and meningitis.

groups in evaluating paralysis after TEVAR. The risk/benefit ratio is lower in TEVAR performed to treat simple descending thoracic aneurysms for which paralysis risk is low. When TEVAR is used to treat Crawford type-I TAAAs and as hybrid TEVAR procedures are used to treat Crawford type-II and III TAAAs, which have a much higher risk of paralysis than simple thoracic aneurysms, the risk/benefit ratio for SFD likely will be similar to open TAAA surgery. CONCLUSIONS

The present analysis supported the following findings. Spinal fluid drainage to a pressure o6 mmHg in open TAAA surgery and o8 mmHg in TEVAR can be performed with acceptably low risk. If leg strength is normal postoperatively, draining more fluid is not necessary in the 48 hours after surgery and may increase the risk of intracranial hemorrhage without additional reduction in paralysis risk. The present analysis identified areas of concern. The appearance of blood-tinged spinal fluid during SFD is not uncommon, and almost half the patients with bloody spinal fluid have intracranial hemorrhage on head CT. Most of these patients are asymptomatic. It is important to identify patients with a small amount of intracranial hemorrhage so that continued drainage does not increase hemorrhage and the risk of serious neurologic injury. With attentive management, neurologic deficit associated with SFD is uncommon. However, although only 0.8% patients with SFD suffered a neurologic complication associated with SFD, patients with neurologic deficit had very high morbidity and mortality. If blood appears in the spinal fluid, immediately stopping draining fluid, determining if intracranial bleeding is present, and preventing or correcting coagulopathy may reduce the incidence of permanent neurologic injury and death associated with SFD.

REFERENCES 1. Blaisdell FW, Cooley DA: The mechanism of paraplegia after temporary thoracic aortic occlusion and its relationship to spinal fluid pressure. Surgery 51:351-355, 1962 2. Miyamoto K, Ueno A, Wada T, et al: A new and simple method of preventing spinal cord damage following temporary occlusion of the thoracic aorta by draining cerebral spinal fluid. J Cardiovasc Surg (Torino) 1:188-197, 1960 3. Khan SN, Stansby G: Cerebrospinal fluid drainage for thoracic and thoracoabdominal aortic aneurysm surgery. Cochrane Database Syst Rev 10: CD003635, 2012

4. Svensson LG, Hess KR, D'Agostino RS, et al: Reduction of neurologic injury after high-risk thoracoabdominal aortic operation. Ann Thorac Surg 66:132-138, 1998 5. Coselli JS, LeMaire SA, Köksoy C, et al: Cerebrospinal fluid drainage reduces paraplegia after thoracoabdominal aortic aneurysm repair: Results of a randomized clinical trial. J Vasc Surg 35:631-639, 2002 6. Safi HJ, Miller CC 3rd, Huynh TT, et al: Distal aortic perfusion and cerebrospinal fluid drainage for thoracoabdominal and descending thoracic aortic repair: Ten years of organ protection. Ann Surg 238: 372-380; discussion 380-381, 2003

8

7. Wong DR, Parenti JL, Green SY, et al: Open repair of thoracoabdominal aortic aneurysm in the modern surgical era: contemporary outcomes in 509 patients. J Am Coll Surg 212: 569-579; discussion 579-581, 2011 8. Greiner A, Mess WH, Schmidli J, et al: Cyber medicine enables remote neuromonitoring during aortic surgery. J Vasc Surg 55: 1227-1232; discussion 1232-1233, 2012 9. Lancaster RT, Conrad MF, Patel VI, et al: Further experience with distal aortic perfusion and motor-evoked potential monitoring in the management of extent I-III thoracoabdominal aortic aneurysms. J Vasc Surg 58:283-290, 2013 10. Lima B, Nowicki ER, Blackstone EH, et al: Spinal cord protective strategies during descending and thoracoabdominal aortic aneurysm repair in the modern era: The role of intrathecal papaverine. J Thorac Cardiovasc Surg 143:945-952, 2012 11. Di Luozzo G, Geisbüsch S, Lin HM, et al: Open repair of descending and thoracoabdominal aortic aneurysms and dissections in patients aged younger than 60 years: Superior to endovascular repair? Ann Thorac Surg 95:12-19, 2013 12. Wynn MM, Mell MW, Tefera G, et al: Complications of spinal fluid drainage in thoracoabdominal aortic aneurysm repair: A report of 486 patients treated from 1987 to 2008. J Vasc Surg 49:29-34; discussion 34-35, 2009 13. Cheung AT, Pochettino A, Guvakov DV, et al: Safety of lumbar drains in thoracic aortic operations performed with extracorporeal circulation. Ann Thorac Surg 76:1190-1196; discussion 1196-1197, 2003 14. Estrera AL, Sheinbaum R, Miller CC, et al: Cerebrospinal fluid drainage during thoracic aortic repair: safety and current management. Ann Thorac Surg 88:9-15; discussion 15, 2009 15. Youngblood SC, Tolpin DA, LeMaire SA, et al: Complications of cerebrospinal fluid drainage after thoracic aortic surgery: A review of 504 patients over 5 years. J Thorac Cardiovasc Surg 146:166-171, 2013 16. Weaver KD, Wiseman DB, Farber M, et al: Complications of lumbar drainage after thoracoabdominal aortic aneurysm repair. J Vasc Surg 34:623-627, 2001 17. Jeong YB, Choi WJ, Han SH, et al: Cerebral epidural hematoma following cerebrospinal fluid drainage during thoracoabdominal aortic repair. Acta Anaesthesiol Scand 53:1221-1222, 2009 18. Acher CW, Wynn MM, Archibald J: Naloxone and spinal fluid drainage as adjuncts in the surgical treatment of thoracoabdominal and thoracic aneurysms. Surgery 108:755-761; discussion 761-752, 1990 19. Acher CW, Wynn MM, Hoch JR, et al: Combined use of cerebral spinal fluid drainage and naloxone reduces the risk of paraplegia in thoracoabdominal aneurysm repair. J Vasc Surg 19: 236-246; discussion 247-248, 1994 20. Acher CW, Wynn MM, Hoch JR, et al: Cardiac function is a risk factor for paralysis in thoracoabdominal aortic replacement. J Vasc Surg 27:821-828; discussion 829-830, 1998 21. Acher CW, Wynn MM, Mell MW, et al: A quantitative assessment of the impact of intercostal artery reimplantation on paralysis risk in thoracoabdominal aortic aneurysm repair. Ann Surg 248:529-540, 2008 22. Horlocker TT, Wedel DJ, Benzon H, et al: Regional anesthesia in the anticoagulated patient: Defining the risks (the second ASRA Consensus Conference on Neuraxial Anesthesia and Anticoagulation). Reg Anesth Pain Med 28:172-197, 2003 23. van Ryn J, Stangier J, Haertter S, et al: Dabigatran etexilate—A novel, reversible, oral direct thrombin inhibitor: Interpretation of coagulation assays and reversal of anticoagulant activity. Thromb Haemost 103:116-127, 2010 24. Hamilton IN Jr, Hollier LH: Adjunctive therapy for spinal cord protection during thoracoabdominal aortic aneurysm repair. Semin Thorac Cardiovasc Surg 10:35-39, 1998

WYNN ET AL

25. Svensson LG, Crawford ES, Hess KR, et al: Experience with 1509 patients undergoing thoracoabdominal aortic operations. J Vasc Surg 17:357-368; discussion 368-370, 1993 26. Bavaria JE, Woo YJ, Hall RA, et al: Retrograde cerebral and distal aortic perfusion during ascending and thoracoabdominal aortic operations. Ann Thorac Surg 60:345-352; discussion 352-343, 1995 27. Griepp RB, Ergin MA, Galla JD, et al: Looking for the artery of Adamkiewicz: A quest to minimize paraplegia after operations for aneurysms of the descending thoracic and thoracoabdominal aorta. J Thorac Cardiovasc Surg 112:1202-1213; discussion 1213-1215, 1996 28. Jacobs MJ, Mess WH: The role of evoked potential monitoring in operative management of type I and type II thoracoabdominal aortic aneurysms. Semin Thorac Cardiovasc Surg 15:353-364, 2003 29. Etz CD, Halstead JC, Spielvogel D, et al: Thoracic and thoracoabdominal aneurysm repair: Is reimplantation of spinal cord arteries a waste of time? Ann Thorac Surg 82:1670-1677, 2006 30. Jacobs MJ, Mess W, Mochtar B, et al: The value of motor evoked potentials in reducing paraplegia during thoracoabdominal aneurysm repair. J Vasc Surg 43:239-246, 2006 31. Coselli JS, Bozinovski J, LeMaire SA: Open surgical repair of 2286 thoracoabdominal aortic aneurysms. Ann Thorac Surg 83: S862-S864, 2007 32. Acher C, Wynn M: Outcomes in open repair of the thoracic and thoracoabdominal aorta. J Vasc Surg 52(suppl):3S-9S, 2010 33. Marini CP, Levison J, Caliendo F, et al: Control of proximal hypertension during aortic cross-clamping: Its effect on cerebrospinal fluid dynamics and spinal cord perfusion. Semin Thorac Cardiovasc Surg 10:51-56, 1998 34. Etz CD, Zoli S, Bischoff MS, et al: Measuring the collateral network pressure to minimize paraplegia risk in thoracoabdominal aneurysm resection. J Thorac Cardiovasc Surg 140(suppl): S125-S130; discussion S142-S146, 2010 35. Etz CD, Kari FA, Mueller CS, et al: The collateral network concept: Remodeling of the arterial collateral network after experimental segmental artery sacrifice. J Thorac Cardiovasc Surg 141: 1029-1036, 2011 36. Dardik A, Perler BA, Roseborough GS, et al: Subdural hematoma after thoracoabdominal aortic aneurysm repair: An underreported complication of spinal fluid drainage? J Vasc Surg 36:47-50, 2002 37. Fedorow CA, Moon MC, Mutch WA, et al: Lumbar cerebrospinal fluid drainage for thoracoabdominal aortic surgery: Rationale and practical considerations for management. Anesth Analg 111:46-58, 2011 38. Hiratzka LF, Bakris GL, Beckman JA, et al: 2010 ACCF/AHA/ AATS/ACR/ASA/SCA/SCAI/SIR/STS/SVM guidelines for the diagnosis and management of patients with thoracic aortic disease. A report of the American College of Cardiology Foundation/American Heart Association Task Force on Practice Guidelines, American Association for Thoracic Surgery, American College of Radiology, American Stroke Association, Society of Cardiovascular Anesthesiologists, Society for Cardiovascular Angiography and Interventions, Society of Interventional Radiology, Society of Thoracic Surgeons, and Society for Vascular Medicine. J Am Coll Cardiol 55:e27-e129, 2010 39. Huang TY, Chung HW, Chen MY, et al: Supratentorial cerebrospinal fluid production rate in healthy adults: quantification with two-dimensional cine phase-contrast MR imaging with high temporal and spatial resolution. Radiology 233:603-608, 2004 40. Zoli S, Roder F, Etz CD, et al: Predicting the risk of paraplegia after thoracic and thoracoabdominal aneurysm repair. Ann Thorac Surg 90:1237-1244; discussion 1245, 2010 41. Crawford ES, Crawford JL, Safi HJ, et al: Thoracoabdominal aortic aneurysms: Preoperative and intraoperative factors determining immediate and long-term results of operations in 605 patients. J Vasc Surg 3:389-404, 1986

COMPLICATIONS OF SPINAL FLUID DRAINAGE

42. LeMaire SA, Miller CC, Conklin LD, et al: Estimating group mortality and paraplegia rates after thoracoabdominal aortic aneurysm repair. Ann Thorac Surg 75:508-513, 2003 43. Augoustides JG, Stone ME, Drenger B: Novel approaches to spinal cord protection during thoracoabdominal aortic interventions. Curr Opin Anaesthesiol 27:98-105, 2014 44. Governale LS, Fein N, Logsdon J, et al: Techniques and complications of external lumbar drainage for normal pressure hydrocephalus. Neurosurgery 63(suppl):379-384; discussion 384, 2008 45. Francia A, Parisi P, Vitale AM, et al: Life-threatening intracranial hypotension after diagnostic lumbar puncture. Neurol Sci 22: 385-389, 2001 46. Gaucher DJ Jr, Perez JA Jr: Subdural hematoma following lumbar puncture. Arch Intern Med 162:1904-1905, 2002 47. Horlocker TT, Wedel DJ, Rowlingson JC, et al: Regional anesthesia in the patient receiving antithrombotic or thrombolytic therapy: American Society of Regional Anesthesia and Pain Medicine Evidence-Based Guidelines (Third Edition). Reg Anesth Pain Med 35: 64-101, 2010 48. Leyvi G, Ramachandran S, Wasnick JD, et al: Case 3—2005 risk and benefits of cerebrospinal fluid drainage during thoracoabdominal aortic aneurysm surgery. J Cardiothorac Vasc Anesth 19:392-399, 2005 49. Roll JD, Larson TC 3rd, Soriano MM: Cerebral angiographic findings of spontaneous intracranial hypotension. AJNR Am J Neuroradiol 24:707-708, 2003 50. Reina MA, Alvarez-Linera J, López A, et al: Magnetic resonance in dural post-puncture headache in patient with cerebrospinal fluid hypotension [article in Spanish]. Rev Esp Anestesiol Reanim 49: 89-100, 2002 51. de Noronha RJ, Sharrack B, Hadjivassiliou M, et al: Subdural haematoma: A potentially serious consequence of spontaneous intracranial hypotension. J Neurol Neurosurg Psychiatry 74:752-755, 2003 52. Bavaria JE, Appoo JJ, Makaroun MS, et al: Endovascular stent grafting versus open surgical repair of descending thoracic aortic aneurysms in low-risk patients: A multicenter comparative trial. J Thorac Cardiovasc Surg 133:369-377, 2007 53. Cheung AT, Pochettino A, McGarvey ML, et al: Strategies to manage paraplegia risk after endovascular stent repair of descending thoracic aortic aneurysms. Ann Thorac Surg 80:1280-1288; discussion 1288-1289, 2005

9

54. Hanna JM, Andersen ND, Aziz H, et al: Results with selective preoperative lumbar drain placement for thoracic endovascular aortic repair. Ann Thorac Surg 95:1968-1974; discussion 1974-1965, 2013 55. Hnath JC, Mehta M, Taggert JB, et al: Strategies to improve spinal cord ischemia in endovascular thoracic aortic repair: Outcomes of a prospective cerebrospinal fluid drainage protocol. J Vasc Surg 48: 836-840, 2008 56. Hiratzka LF, Bakris GL, Beckman JA, et al: 2010 ACCF/AHA/ AATS/ACR/ASA/SCAI/SIR/STS/SVM guidelines for the diagnosis and management of patients with thoracic aortic disease: A report of the American College of Cardiology Foundation/American Heart Association Task Force on Practice Guidelines, American Association for Thoracic Surgery, American College of Radiology, American Stroke Association, Society of Cardiovascular Anesthesiologists, Society for Cardiovascular Angiography and Interventions, Society of Interventional Radiology, Society of Thoracic Surgeons, and Society for Vascular Medicine. Circulation 121:266-369, 2010 57. Wong CS, Healy D, Canning C, et al: A systematic review of spinal cord injury and cerebrospinal fluid drainage after thoracic aortic endografting. J Vasc Surg 56:1438-1447, 2012 58. Bobadilla JL, Wynn M, Tefera G, et al: Low incidence of paraplegia after thoracic endovascular aneurysm repair with proactive spinal cord protective protocols. J Vasc Surg 57:1537-1542, 2013 59. Keith CJ Jr, Passman MA, Carignan MJ, et al: Protocol implementation of selective postoperative lumbar spinal drainage after thoracic aortic endograft. J Vasc Surg 55:1-8; discussion 8, 2012 60. Reilly LM, Rapp JH, Grenon SM, et al: Efficacy and durability of endovascular thoracoabdominal aortic aneurysm repair using the caudally directed cuff technique. J Vasc Surg 56:53-63; discussion 63-54, 2012 61. Guillou M, Bianchini A, Sobocinski J, et al: Endovascular treatment of thoracoabdominal aortic aneurysms. J Vasc Surg 56: 65-73, 2012 62. Acher CW, Wynn M: A modern theory of paraplegia in the treatment of aneurysms of the thoracoabdominal aorta: An analysis of technique specific observed/expected ratios for paralysis. J Vasc Surg 49:1117-1124; discussion 1124, 2009 63. Cox GS, O’Hara PJ, Hertzer NR, et al: Thoracoabdominal aneurysm repair: A representative experience. J Vasc Surg 15: 780-787: discussion 787-788, 1992