Descending Thoracic Aortic Aneurysm: Surgical Approach and Treatment Using the Adjuncts Cerebrospinal Fluid Drainage and Distal Aortic Perfusion Anthony L. Estrera, MD, Forrest S. Rubenstein, MD, Charles C. Miller III, PhD, Tam T. T. Huynh, MD, George V. Letsou, MD, and Hazim J. Safi, MD Department of Cardiothoracic and Vascular Surgery, The University of Texas at Houston Medical School, Memorial Hermann Hospital, Houston, Texas Background. Neurologic deficit (paraplegia or paraparesis) remains a significant morbidity in the repair of descending thoracic aortic aneurysm. Methods. Between February 1991 and February 2000, we operated on 182 patients for descending thoracic aortic aneurysm. For the purpose of this study to identify the impact of the combined adjuncts distal aortic perfusion and cerebrospinal fluid (CSF) drainage on neurologic outcome we selected the 148 of 182 nonemergent patients who had received conventional treatment (simple cross-clamping with or without adjuncts). The mean patient age was 61 years, and 49 of the 148 (33%) patients were women. Nine of the 148 patients (6%) had acute type B dissections. We compared the results of 105 of the 148 patients (71%) who received the combined adjuncts of CSF drainage and distal aortic perfusion with the remaining 43 (29%) patients who underwent repair using the simple cross-clamp with or without the addition of a single adjunct. Results. Overall 30-day mortality was 13 of 148 patients (8.8%). Overall early neurologic deficit was 4 of 148 (2.7%): 1 of 105 (0.9%) patients who had received distal aortic perfusion and CSF drainage, versus 3 of 43 (7%) in all other patients (p < 0.04). Conclusions. In our practice the use of the combined adjuncts of CSF drainage and distal aortic perfusion has all but eliminated the incidence of immediate postoperative neurologic deficit in nonemergent patients with aneurysms of the descending thoracic aorta. (Ann Thorac Surg 2001;72:481 6) 2001 by The Society of Thoracic Surgeons Repair of descending thoracic aortic aneurysms remains a surgical challenge. Since inception of repair techniques, the need for spinal cord protection has been of paramount importance. Although surgical adjuncts, such as the Gott shunt and distal aortic perfusion, were used in the earlier days of descending thoracic aortic aneurysm repair, these adjuncts were often reported to have either little effect on spinal cord protection [1 3] or adverse effects on survival [4]. Emphasis was placed on the speed with which operations were completed. This emphasis created a dilemma in carrying out an operation that often requires more than 30 minutes, because that is the time generally accepted as the margin for spinal cord safety. The incidence of neurologic deficit from the 1970s into the 1980s was reported at 3% to 10% [3]. Today the incidence of neurologic deficit has dropped significantly. Surgeons currently disagree over which particular adjunct provides superior spinal cord protection, but the general consensus is that adjuncts are Presented at the Forty-seventh Annual Meeting of the Southern Thoracic Surgical Association, Marco Island, FL, Nov 9 11, 2000. Address reprint requests to Dr Safi, Department of Cardiothoracic and Vascular Surgery, UTH Medical Center, 6410 Fannin St, Suite 450, Houston, TX 77030; e-mail: hazim.i.safi@uth.tmc.edu. necessary and may be the reason for the decline in neurologic complications [5 9]. In our practice we have noted the greatest degree of success with the combination of cerebrospinal fluid (CSF) drainage and distal aortic perfusion. This was first demonstrated in thoracoabdominal aortic aneurysm repair [10] and later in descending thoracic aortic aneurysm operations [11]. Intercostal artery reimplantation and moderate hypothermia have also played significant roles [12]. The purpose of this study was to examine the significant factors in the prevention of neurologic deficit during nonemergent repair of descending thoracic aortic aneurysms with the combination of the adjuncts distal aortic perfusion and CSF drainage. Material and Methods Patients Between February 1991 and February 2000, 182 patients underwent graft repair of descending thoracic aortic aneurysms. We excluded from this study 26 (12.6%) of 182 patients who had involvement of the transverse aortic arch that precluded placement of a proximal clamp and who were treated using profound hypothermic circula- 2001 by The Society of Thoracic Surgeons 0003-4975/01/$20.00 Published by Elsevier Science Inc PII S0003-4975(01)02679-0
482 ESTRERA ET AL Ann Thorac Surg TREATMENT OF DESCENDING THORACIC AORTIC ANEURYSM 2001;72:481 6 Table 1. Patient Characteristics Variable No. Patients (%) No. Neurologic Deficit (%) Odds Ratio a 95% CI b p c All patients 148 (100) 4 (2.7) Age (y) 8 53 36 (24.3) 0 (0.0) 1.04 0.95 1.13 0.42 54 64 35 (23.7) 1 (2.9) (0.63) 65 71 36 (24.3) 1 (2.8) 72 85 41 (27.7) 2 (4.9) Sex Female 49 (33.1) 1 (2.0) 0.67 0.07 6.58 0.73 Male 99 (66.9) 3 (3.0) 1 Hypertensive 97 (65.5) 3 (3.1) 1.57 0.16 15.74 0.69 Normotensive 51 (34.5) 1 (2.0) 1 COPD 32 (21.6) 2 (6.3) 3.80 0.51 28.10 0.17 Otherwise 116 (78.4) 2 (1.7) 1 Extent C 82 (55.4) 4 (4.9) 7.63 0.40 144.2 0.07 Extent A or B 66 (44.6) 0 (0.0) 1 Acute dissection 9 (6.1) 1 (11.1) 5.67 0.53 60.79 0.11 Otherwise 139 (93.9) 3 (2.2) 1 Chronic dissection 62 (41.9) 1 (1.6) 0.45 0.05 4.46 0.49 Otherwise 86 (58.1) 3 (3.5) 1 Intercostal reattachment 59 (40.1) 1 (1.7) 0.49 0.05 4.81 0.53 Otherwise 88 (59.9) 3 (3.4) 1 Adjunct otherwise 105 (70.9) 1 (0.9) 0.13 0.01 1.27 0.04 43 (29.1) 3 (7.0) 1 Cross-clamp time (min) Not available 4 7 22 33 (22.9) 1 (3.0) 1.00 0.93 1.08 0.94 23 30 34 (23.6) 1 (2.9) (0.62) 31 39 37 (25.7) 0 (0.0) 40 85 40 (27.8) 2 (5.0) a Logit. For dichotomous variables, the odds ratio represents a test against a reference category whose referent odds ratio is equal to 1. For continuous data, the odds ratio refers to the increase in odds associated with a one-unit increase in the variable value. This value can be converted to an odds ratio for an N unit change in the variable by taking the natural logarithm of the odds ratio, multiplying this value by the number of variable units desired, and taking the exponent of the product. Although continuous data are presented in quartiles, the odds ratios are against the continuous variable. b 95% CI 95% confidence interval, reflecting the units against which its companion odds ratio is computed. Confidence intervals are test-based. c p probability of type I statistical error (common p value). Values without parentheses are Pearson chi-square probabilities. Probability values in parentheses are univariate logistic regression likelihood ratio p values. COPD chronic obstructive pulmonary disease. tory arrest. Emergent patients had either free or contained rupture and were hemodynamically unstable, and all underwent simple cross-clamp repair. Consequently, 8 emergent patients were also excluded from analysis, as their inclusion in the comparison group would have created an immediate bias. The 148 patients who underwent nonemergent repair of descending thoracic aortic aneurysms were analyzed with respect to the impact of adjuncts. Patient characteristics at the time of repair are listed in Table 1. There were 99 men (67%) and 49 women (33%). Patient age ranged from 8 to 85 years (mean 61 years). One hundred five patients received the spinal cord adjuncts of distal aortic perfusion and CSF drainage described below. The remaining 43 patients were operated on with simple cross-clamp alone (11 patients) or received the single adjunct of either distal aortic perfusion (28 patients) or CSF drainage (4 patients). Surgical Technique Details of the technique have been described previously [10]. Briefly, the patient was anesthetized and intubated using a double-lumen endotracheal tube. An arterial line and a pulmonary artery catheter were placed to monitor patient hemodynamics. A CSF catheter was placed in the third or fourth lumbar space to allow CSF drainage and monitoring of CSF pressure (Fig 1). The CSF pressure was maintained at less than 10 mm Hg throughout the procedure. The patient was positioned in the right lateral decubitus position and was prepared and draped. We
Ann Thorac Surg ESTRERA ET AL 2001;72:481 6 TREATMENT OF DESCENDING THORACIC AORTIC ANEURYSM 483 Fig 1. Cerebrospinal fluid catheter insertion. Cerebrospinal pressure was maintained at less than 10 mm Hg. used a modified thoracoabdominal incision, beginning in the abdomen 3 cm below the costal margin and continuing over the sixth rib before curving cephalad just posterior to the tip of the scapula. The lung was deflated and the sixth rib excised. We completed this incision by dividing the costal cartilage with heavy scissors. A selfretaining retractor was then inserted and the aneurysm was inspected. The pericardium was opened posterior to the phrenic nerve and the patient was given intravenous heparin (1 mg/kg). The left atrium was cannulated through the left pulmonary vein or the left atrial appendage. A BioMedicus (Minneapolis, MN) pump with an in-line heat exchanger was attached to this cannula and the arterial inflow was established through the left femoral artery or the descending thoracic aorta (Fig 2). Circumferential dissection of the aorta proximal to the aneurysm and at the diaphragm permitted safe crossclamping of the aorta. Distal aortic perfusion began as the proximal and distal cross-clamps were applied. The sequence of clamp placement depended on the extent of the descending thoracic aneurysm. The anesthesiologist carefully maintained a normal proximal aortic pressure during this time. The aorta was opened longitudinally and separated from the esophagus. Stay sutures were applied to the aneurysm wall and hemostasis obtained by oversewing any bleeding intercostal arteries that were not to be reimplanted. Blood salvage was accomplished Fig 2. Distal aortic perfusion through the left common femoral artery and left superior pulmonary vein. Fig 3. Descending thoracic aortic aneurysms are classified as type A, left subclavian artery to T6 (A); type B, T6 to the diaphragm (B); or type C, left subclavian artery to the diaphragm (C). with a cell saver device, and blood was reinfused using a rapid infuser system. Once adequate hemostasis was obtained, an appropriately sized, woven Dacron tube graft was anastomosed to the proximal aorta with a running polypropylene suture. If patent intercostal arteries were to be reattached, the graft was cut in a beveled fashion and the distal anastomosis completed. Reimplantation of patent, lower intercostal arteries (T8 to T12) was performed routinely except in cases of acute dissection or when technically impossible. The distal anastomosis was then performed with the graft flushed just before its completion. The aortic clamps were slowly removed and suture lines checked for hemostasis. The patient was weaned from bypass once the rectal temperature reached 36 C. Intravenous protamine was administered to reverse the effect of the heparin and the atrial and femoral cannulas were removed. Postoperatively, the mean arterial pressure was maintained between 80 and 100 mm Hg. Cerebrospinal fluid was drained no more than 20 ml/hour to maintain a CSF pressure of less than 10 mm Hg for 3 days. If a delayed neurologic deficit appeared after removal of the drain, a new CSF drain was reinserted immediately to decrease the CSF pressure, a practice that may lead to prompt resolution of the neurologic deficit [13]. Outcome Variables and Statistical Analysis Descending thoracic aortic aneurysms were classified according to Figure 3. Type A descending thoracic aortic aneurysms (left subclavian artery to T6) were repaired in 42 patients (28.4%), type B (T6 to the diaphragm) in 24 patients (16.2%), and type C (left subclavian artery to the diaphragm) in 82 patients (55.4%). Aneurysms with dissection were considered acute if an operation was performed in less than 14 days from the onset of pain, and chronic if after 14 days. Postoperative neurologic deficit was defined as paraplegia or paraparesis observed upon the patient awakening from anesthesia, regardless of severity. Those patients who developed paraplegia or paraparesis after a period of normal neurologic function
484 ESTRERA ET AL Ann Thorac Surg TREATMENT OF DESCENDING THORACIC AORTIC ANEURYSM 2001;72:481 6 were classified as having had a delayed neurologic deficit. Patients who sustained cerebral infarction identified by a thorough neurologic examination and computed tomographic scan of the head, were excluded from the neurologic deficit group. Operative mortality was defined as death occurring within 30 days of an operation. Data were collected from chart reviews by a trained nurse abstractor, and were entered into a dedicated Microsoft Access (Microsoft, Redmond, WA) database. Analysis was retrospective. Data were exported to SAS for data analysis, and all computations were performed using SAS version 6.12 running under Windows NT (SAS Institute, Cary, NC). Univariate categorical data were analyzed using contingency table analyses. For 2 2 tables, common odds ratios with test-based confidence intervals were computed, and 2 statistics are reported for hypothesis tests. For tables greater than 2 2, 2 statistics were computed for hypothesis tests, and univariate logistic regression estimates were also computed keeping data in their native continuous distribution. The p values and confidence intervals for continuous data were based on maximum-likelihood estimates. Results The 30-day mortality for elective repair of the descending thoracic aortic aneurysm was 8.8% (13 of 148 patients). The 30-day mortality for emergent repair was 25%. Age, aortic dissection, and aortic clamp time were not significant with regard to neurologic deficit (Table 1). The incidence of cerebral vascular accident for repair of descending thoracic aneurysms was 2.7%, and renal failure was 7.2%. The overall rate of neurologic deficit was 2.7% (4 of 148 patients). There were no cases of neurologic deficit in aneurysm types A or B. All 4 patients with neurologic deficit had aneurysms of the entire descending thoracic aorta classified as type C (p 0.7). The rate of immediate neurologic deficit was 0.9% (1 of 105 patients) in those patients whose aneurysms were repaired using the combination of distal aortic perfusion and CSF drainage versus 7.0% (3 of 43 patients) in the comparison group (p 0.04). Within the comparison group, two cases of neurologic deficit were observed in patients who received distal aortic perfusion (2 of 28 patients), and one case in a patient who underwent simple clamp and sew technique alone (1 of 11 patients). No cases of neurologic deficit were observed in the 4 patients who received CSF drainage alone. Two cases of delayed neurologic deficit were observed, one in the combined adjunct group and one in the comparison group (p NS). Intercostal artery reattachment was performed in 59 cases (40.1%), but did not demonstrate a benefit for protection against neurologic deficit (p 0.55). The aortic cross-clamp periods for each aneurysm type (A, B, C) were 33, 27, and 33 minutes, respectively (p NS). Comment Neither distal aortic perfusion nor CSF drainage used individually proved to be exceptionally effective in the Fig 4. (A) Dynamics of aortic cross-clamp: cerebrospinal fluid (CSF) pressure increases and distal aortic (DAo) pressure decreases. (B) Dynamics of aortic cross-clamp with adjuncts: CSF pressure decreases and DAo pressure increases, thus increasing perfusion pressure of the spinal cord. previous era of descending thoracic aortic aneurysm repair [14, 15]. More contemporary series, however, have demonstrated the benefit of these adjuncts on the incidence of neurologic deficit [16 18]. We believe that the combination of these two adjuncts may provide significant spinal cord protection. This defense against spinal cord ischemia is achieved during aortic cross-clamping by creating a balance between decreased distal aortic pressure and increased CSF pressure [11, 19]. Specifically, the decrease in distal aortic pressure causes a decrease in the spinal artery pressure. A concomitant rise in CSF pressure can lead to spinal cord compartment syndrome [20], resulting in further spinal cord ischemia. By draining the excess CSF, pressure is reduced, relieving the compartment syndrome and augmenting perfusion to the spinal cord (Fig 4). At the same time, distal aortic perfusion increases the distal aortic pressure and increases perfusion pressure of the spinal cord [21]. The correlation between aneurysm extent and patient outcome in thoracoabdominal aortic aneurysms was recognized in 1986 [22]. A similar correlation in descending thoracic aortic aneurysms was demonstrated in an analysis of Crawford s experience of 832 descending thoracic aneurysm cases [21]. In the report of Crawford and colleagues [22], descending thoracic aneurysms were classified as A, proximal third; B, middle third; or C, distal third of the descending thoracic aorta. Moreover, Safi and colleagues [21] found that resection of the entire descending thoracic aorta when compared with resection of just the proximal extent (A) was a risk factor for neurologic deficit. Based on this analysis, we modified this classification scheme of descending thoracic aortic
Ann Thorac Surg ESTRERA ET AL 2001;72:481 6 TREATMENT OF DESCENDING THORACIC AORTIC ANEURYSM 485 aneurysms by renaming type A as the proximal half (subclavian to T6), type B as the distal half (T6 to T12), and type C as the entire descending thoracic aorta. Although not statistically significant, (p 0.07), in the current series all 4 cases of neurologic deficit occurred in patients with aneurysms involving the entire descending thoracic aorta (type C). Continued adherence to this classification system may allow the determination of its prognostic significance. Although neurologic deficit has been directly linked to the aortic ischemic period in the simple cross-clamp technique, we found no correlation in the current series between aortic cross-clamp time and neurologic deficit; two neurologic deficits were noted in patients with crossclamp times longer than 30 minutes and two in patients with aortic cross-clamp times of less than 30 minutes (p 0.62). In the analysis by Safi and coworkers [21], distal aortic perfusion was shown to negate the effect of more than 40 minutes of aortic cross-clamp time on neurologic deficit. Similar to this previous series, the use of the combined adjuncts appears to negate the effect of prolonged ischemic time. Reattachment of the lower (T8 to T12) intercostal arteries was previously shown to reduce the risk of neurologic deficits during thoracoabdominal aortic aneurysm repair [12]. Although we emphasize the importance of intercostal artery reattachment in descending thoracic aortic repair, its significance to spinal cord protection was inconclusive (p 0.55). This finding may have been due to the low number of patients who had intercostal artery reattachment (40.1%). Limitations of this study included the retrospective nature of the analysis. In addition, the method selection was nonrandomized. Although the overall number of neurologic deficits was small, the advantage of distal aortic perfusion and CSF drainage may prove to be more evident in a future larger series. The use of the combined adjuncts of distal aortic perfusion and CSF drainage was performed safely and significantly reduced the rate of neurologic deficit during nonemergent repair of descending thoracic aortic aneurysms. Because classification of descending thoracic aortic aneurysms may have prognostic significance, future studies reporting outcomes of repair should include this classification scheme for risk analysis and accurate reporting. We thank Amy Wirtz Newland for her editorial assistance. References 1. Crawford ES, Rubio PA. Reappraisal of adjuncts to avoid ischemia in the treatment of aneurysms in descending thoracic aorta. J Thorac Cardiovasc Surg 1973;66:693 704. 2. Crawford ES, Walker HS, Saleh SA, Normann NA. Graft replacement of aneurysm in descending thoracic aorta: results without bypass or shunting. Surgery 1981;89:73 85. 3. Livesay JL, Cooley DA, Ventemiglia RA, et al. Surgical experience in descending thoracic aneurysmectomy with and without adjuncts to avoid ischemia. Ann Thorac Surg 1985;145:37 46. 4. Stavens B, Hashim SW, Hammond GL, et al. Optimal methods of repair of descending thoracic aortic transections and aneurysms. Am J Surg 1983;145:508 13. 5. Verdant A, Page A, Cossette R, Dontigny L, Page P. Development of circulatory support during 420 resections of the descending thoracic aorta. Ann Chir 1996;50:619 25. 6. Svensson LG. An approach to spinal cord protection during descending or thoracoabdominal aortic repairs. Ann Thorac Surg 1999;67:1935 8. 7. Rokkas CK, Kouchoukos NT. Profound hypothermia for spinal cord protection in operations on the descending thoracic and thoracoabdominal aorta. Semin Thorac Cardiovasc Surg 1998;10:57 60. 8. Hamilton IN Jr, Hollier LH. Adjunctive therapy for spinal cord protection during thoracoabdominal aortic aneurysm repair. Semin Thorac Cardiovasc Surg 1998;10:35 9. 9. Cambria RP, Davison JK, Carter C, et al. Epidural cooling for spinal cord protection during thoracoabdominal aneurysm repair: a five-year experience. J Vasc Surg 2000;31:1093 102. 10. Safi HJ, Hess KR, Randel M, et al. Cerebrospinal fluid drainage and distal aortic perfusion: reducing neurologic complications in repair of thoracoabdominal aortic aneurysm types I and II. J Vasc Surg 1996;23:223 9. 11. Safi HJ, Campbell MP, Ferreira ML, Azizzadeh A, Miller CC. Spinal cord protection in descending thoracic and thoracoabdominal aortic aneurysm repair. Semin Thorac Cardiovasc Surg 1998;10:41 4. 12. Safi HJ, Miller CC 3rd, Carr C, Iliopoulos DC, Dorsay DA, Baldwin JC. Importance of intercostal artery reattachment during thoracoabdominal aortic aneurysm repair. J Vasc Surg 1998;27:58 68. 13. Azizzadeh A, Huynh TT, Miller CC 3rd, Safi HJ. Reversal of twice-delayed neurologic deficits with cerebrospinal fluid drainage after thoracoabdominal aneurysm repair: a case report and plea for a national database collection. J Vasc Surg 2000;31:592 8. 14. Crawford ES, Svensson LG, Hess KR, et al. A prospective randomized study of cerebrospinal fluid drainage to prevent paraplegia after high-risk surgery on the thoracoabdominal aorta. J Vasc Surg 1991;13:36 46. 15. Crawford ES, Mizrahi EM, Hess KR, Coselli JS, Safi HJ, Patel VM. The impact of distal aortic perfusion and somatosensory evoked potential monitoring on prevention of paraplegia after aortic aneurysm operation [published erratum appears in J Thorac Cardiovasc Surg 1989 May;97:665]. J Thorac Cardiovasc Surg 1988;95:357 67. 16. Svensson LG, Crawford ES, Hess KR, Coselli JS, Safi HJ. Variables predictive of outcome in 832 patients undergoing repairs of the descending thoracic aorta. Chest 1993;104: 1248 53. 17. Borst HG, Jurmann M, Buhner B, Laas J. Risk of replacement of descending aorta with a standardized left heart bypass technique. J Thorac Cardiovasc Surg 1994;107:126 33. 18. Coselli JS, LeMaire SA. Left heart bypass reduces paraplegia rates after thoracoabdominal aortic aneurysm repair. Ann Thorac Surg 1999;67:1931 4; discussion 1953 8. 19. Safi HJ, Miller CC 3rd. Spinal cord protection in descending thoracic and thoracoabdominal aortic repair. Ann Thorac Surg 1999;67:1937 9; discussion 1953 8. 20. Mauney MC, Tribble CG, Cope JT, et al. Is clamp and sew still viable for thoracic aortic resection? Ann Surg 1996;223: 534 43. 21. Safi HJ, Campbell MP, Miller CC 3rd, et al. Cerebral spinal fluid drainage and distal aortic perfusion decrease the incidence of neurological deficit: the results of 343 descending and thoracoabdominal aortic aneurysm repairs. Eur J Vasc Endovasc Surg 1997;14:118 24. 22. 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 1986;3:389 404.
486 ESTRERA ET AL Ann Thorac Surg TREATMENT OF DESCENDING THORACIC AORTIC ANEURYSM 2001;72:481 6 DISCUSSION DR DARRYL S. WEIMAN (Memphis, TN): I would like to ask, when you use left atrial to femoral bypass, how much heparin do you give in the circuit? DR ESTRERA: We give 1 mg/kg DR WEIMAN: Some people are not using any heparin. DR ESTRERA: That is correct. Since 1995, Dr Safi has been using moderate heparinization, since he has had some cases in which the circuit thrombosed. In general, circuit flows between 800 and 1,800 cc/minute are utilized during left atrial to femoral bypass. In some situations, however, circuit flows may decrease, and thus flows of less than 500 cc/minute are more of a concern for thrombosis. DR CHARLES WILLEKES (Muskegon, MI): A very nice paper. I have two questions. First, when you set up partial left heart bypass, when do you make your decision to go on total bypass if you cannot place a proximal clamp? Second, you say that you use only two adjuncts, yet you reattach intercostal arteries T8 to T12; do you not consider intercostal reattachment as another adjunct? DR ESTRERA: We do consider intercostal artery reattachment an adjunct. But in this study, we evaluated 148 patients and we compared only the combined adjuncts of distal aortic perfusion and cerebrospinal fluid drainage with no combined adjuncts. Intercostal artery reattachment was used in both groups, but was performed only in 40% of the cases. Remember that not all cases required reattachment, therefore the numbers were relatively small. Thus, possibly because of the relatively low number of intercostal reattachments, this was not significant. Dr Safi, however, has previously shown with thoracoabdominal aortic aneurysms that intercostal artery reattachment was significant. The decision to utilize total cardiopulmonary bypass when we cannot place a proximal clamp is made at the time of opening the chest. For us, it is simple to add a venous cannula in the left groin, if need be. INVITED COMMENTARY This paper documents the results of 148 patients operated on for thoracic (ie, not thoracoabdominal aneurysms) aneurysms between February 1991 and February 2000. This study excluded all emergency procedures requiring an open proximal aortic arch, or hypothermic technique, or the inability to place a proximal cross-clamp. This is a fairly straightforward and simple retrospective study. The fundamental findings corroborate the relentless findings by most groups this decade performing thoracic aortic surgery, namely that the addition of distal aortic perfusion with cerebrospinal fluid (CSF) drainage has more spinal cord protection than a straight cross-clamp or any single adjunct alone. However, the series presented by Borst and colleagues 1 showed similar results with LA-FA bypass alone (mortality 3%, paraplegia 2.3%). The only statistically significant finding in this series is that the combination of distal aortic perfusion with CSF drainage provided protection against paraplegia compared to a control group. The control group actually consisted of three subsets of patients: straight cross-clamping, CSF drainage alone, and distal aortic perfusion alone. The authors present a fairly large series of elective descending thoracic aortic procedures. There are not too many thoracic aortic surgery groups capable of supplying these numbers from a single institution. A proposed classification scheme nicely delineates anatomically distinct higher risk thoracic aneurysms versus lower risk thoracic aneurysms and this classification should be considered in all future papers and reviews on descending thoracic aneurysms. The data identified that it was Type C aneurysms (ie, aneurysms from the subclavian artery to the diaphragm) that produced paraplegia in this series. Type A and Type B descending aneurysms had no cases of neurological deficit. Importantly, the statistical analysis also revealed that age, gender, dissection, intercostal reattachment, and chronic obstructive pulmonary disease did not effect the incidence of paraplegia. The information regarding paraplegia differences between Types A, B, and C descending thoracic aneurysms are important, although this does not quite reach statistical significance at p equal to 0.07. This will become more important in the future, as many descending thoracic aortic aneurysms will probably be repaired using endovascular stent grafting technology. During endograft procedures, the following question always arises; How far from the aneurysm can we land the device at distal and proximal landing zones. There is a dilemma as the stent graft seal is better if the landing zone is further away from the true aneurysm, however, as this study implies, the risk of paraplegia is also greater as more of the descending thoracic aorta is covered. We may want to think twice before we pave via endovascular stents, the full descending aorta if it s not absolutely necessary. This series shows, that in expert hands, a relatively low incidence of paraplegia (approximately 1% to 2%) can be attained in elective descending thoracic aortic aneurysm surgery using LA-FA bypass and CSF drainage. Joseph E. Bavaria, MD Department of Thoracic Surgery Hospital of the University of Pennsylvania 6 Silverstein 3400 Spruce St Philadelphia, PA 19104 Reference 1. Borst HG, Jurmann M, Buhner B, Laas R. Risk of replacement of descending aorta with a standardized left heart bypass technique. J Thorac Cardiovasc Surg 1994;107:126 33. 2001 by The Society of Thoracic Surgeons 0003-4975/01/$20.00 Published by Elsevier Science Inc PII S0003-4975(01)02928-9