Outcome of Abdominal Aortic Endografting in High-Risk Patients: A 4-Year Single-Center Study

736 J ENDOVASC THER CLINICAL INVESTIGATION Outcome of Abdominal Aortic Endografting in High-Risk Patients: A 4-Year Single-Center Study Fabio Verzini, MD; Piergiorgio Cao, MD; Simona Zannetti, MD; Gianbattista Parlani, MD; Paola De Rango, MD; Agostino Maselli, MD*; Luciano Lupattelli, MD*; and Basso Parente, MD Unit of Vascular Surgery and *Division of Interventional Radiology, Policlinico Monteluce, Perugia, Italy Purpose: To evaluate feasibility, safety, and effectiveness of endovascular abdominal aortic aneurysm (AAA) repair in patients whose fitness for surgery is questionable. Methods: Between April 1997 and December 2001, 389 consecutive patients underwent endovascular AAA repair. Of these, 51 (13.1%) were ASA grade IV. The perioperative and late outcomes of this group were compared to the remaining 338 patients with ASA grades IV. Failure of AAA exclusion was defined as late conversion to open repair, AAA rupture, increased aneurysm diameter, or persisting graft-related endoleak. Gender, age, ASA grade IV, EUROSTAR class E, and AAA diameter were examined by logistic regression analysis for their influence on perioperative death, survival, and failure of AAA exclusion. Results: Four (7.8%) perioperative deaths occurred in the ASA IV group compared to 1 (0.3%) in the ASA IV group (p0.001). Median follow-up was 22 months (range 1 56). Failure of AAA exclusion occurred in 3 (5.9%) patients in ASA IV group and in 25 (7.4%) in ASA IV group (p0.05). Actuarial survival at 30 months was 62.9% in ASA IV group and 88.0% in ASA IV group (p0.001, log-rank test). There were no independent predictors for failure of AAA exclusion; ASA IV was independently associated with perioperative mortality (HR 17.8; 95% CI 1.6 to 188; p0.016). Conclusions: Endovascular AAA repair in ASA IV patients is feasible and effective in preventing AAA rupture in the mid term. High-risk patients experience a worse prognosis than their good-risk counterparts. An individualized approach in selecting high-risk patients for endoluminal repair is mandatory. J Endovasc Ther Key words: abdominal aortic aneurysm, endovascular repair, treatment outcome, severe comorbidities, stent-graft Elective open repair of abdominal aortic aneurysm (AAA) is often denied to patients whose comorbidities significantly increase the operative risk. Endovascular aneurysm repair (EVAR) has been proposed as a valid alternative in these patients because of the reduced operative time, blood loss, and stress to the patient. 1 4 However, high-risk patients often die from complications associated with their coexistent diseases. The role of prophylactic AAA surgery is still unclear. 5,6 We reviewed our experience with high-risk patients (ASA [American Society of Anesthesiology] class IV) undergoing EVAR to deter- Presented at International Congress XV of Endovascular Interventions, Scottsdale, Arizona, USA, February 10 14, 2002. Address for correspondence and reprints: Piergiorgio Cao, MD, Unità Operativa di Chirurgia Vascolare, Policlinico Monteluce, 06122 Perugia, Italy. Fax: 39-075-578-3194; E-mail: pcao@unipg.it 2002 by the INTERNATIONAL SOCIETY OF ENDOVASCULAR SPECIALISTS Available at www.jevt.org

J ENDOVASC THER ENDOGRAFTING IN HIGH-RISK PATIENTS 737 TABLE 1 Risk Factors in ASA Grade IV and ASA IV Patients Mean age, y Men Smoking Hypertension Diabetes Cardiac disease Respiratory disease Renal insufficiency ASA IV n51 (13.1%) 75.1 45 (88.2%) 29 (56.9%) 32 (62.7%) 7 (13.7%) 37 (72.5%) 42 (82.3%) 8 (15.7%) ASA IV n338 (86.9%) p 70.9 319 (94.4%) 203 (60.1%) 223 (66.0%) 28 (8.3%) 152 (45.0%) 163 (48.2%) 34 (10.1%) 0.001 ASA American Society of Anesthesiologists, not significant. mine their prognosis and whether the endovascular approach in patients unfit for open repair could be justified. METHODS Between April 1997 and December 2001, 389 consecutive patients (364 men; mean age 71.4 years, range 51 94) underwent EVAR at our unit. Preoperative, operative, and follow-up data were collected prospectively in a database. An anesthesiologist analyzed patient comorbidities and risk factors (defined according to the SVS/ISCVS guidelines 7 ) and assigned a preoperative risk grade according to the ASA classification. 8,9 Fifty-one (13.1%) patients were classified as ASA grade IV (severe, life-threatening systemic disorder that may not be correctable by operation, i.e., congestive heart failure, unstable angina 8,9 ) and 338 (86.9%) had an ASA grade between I and III TABLE 2 Comorbidities of 51 ASA Grade IV Patients Severe cardiac disease: unstable angina, congestive heart failure, myocardial infarction within 6 months, ejection fraction 30%, or symptomatic ventricular arrhythmia 26 (51.0%) Severe respiratory disease: forced expiratory volume 800 ml in 1 second, vital capacity 1800 ml, O 2 therapy, or pulmonary hypertension 24 (47.1%) Chronic renal failure: creatinine 3 mg/dl, creatinine clearance 30 ml/min, or need for dialysis 8 (15.7%) Liver failure, cirrhosis History of cancer Previous major stroke Lupus ASA American Society of Anesthesiologists. 1 (2.0%) 10 (19.6%) 8 (15.7%) 1 (2.0%) (ASA IV group). The ASA grade IV patients were older (75.1 versus 70.9 years) and had more cardiac and respiratory diseases (Tables 1 and 2). Anatomical characteristics were evaluated preprocedurally using color duplex sonography, contrast-enhanced conventional or spiral computed tomography (CT) with 5-mm slices, and either angiography or magnetic resonance angiography with gadolinium. Maximal AAA diameter was measured at the widest portion of the aneurysm on the CT scans. When the aneurysm had an elliptical shape, the smaller diagonal was used to avoid errors in size evaluation due to aortic tortuosity. 10 Median AAA diameter (Table 3) was 50.8 mm (range 40 86) in the ASA IV group and 58.2 mm (range 45 74) in the ASA grade IV group. Six ASA grade IV patients had TABLE 3 Anatomical Features of the ASA Grade IV and ASA IV Cohorts ASA IV n51 ASA IV n338 p Median AAA diameter, mm EUROSTAR class E Severe iliac calcification Proximal neck thrombus Short (1.5 cm) proximal neck 58.2 12 (23.5%) 6 (11.8%) 4 (7.8%) 6 (11.8%) 50.8 22 (6.5%) 50 (14.8%) 29 (8.6%) 28 (8.3%) ASA American Society of Anesthesiologists, AAA abdominal aortic aneurysms, EUROSTAR European Collaborators on Stent-graft Techniques for Abdominal Aortic Aneurysm Repair, not significant.

738 ENDOGRAFTING IN HIGH-RISK PATIENTS J ENDOVASC THER AAAs 50 mm, but they were associated with large iliac aneurysms. The extent of the aneurysm was recorded according to the EU- ROSTAR (European Collaborators on Stentgraft Techniques for Abdominal Aortic Aneurysm Repair) classification. 11 Aneurysms in the ASA IV group were larger (58.2 versus 50.8 mm) and more extensive (23% class E versus 6.5%) than in the ASA IV group. Anatomical criteria for endografting were an aortic neck of adequate diameter (30 mm) and length (10 mm) and iliac arteries of suitable diameter (6 mm) without extensive calcification or tortuosity. Although the presence of neck thrombus, severe neck angulations (60), and aneurysmal extension to both hypogastric arteries have been considered relative contraindications for EVAR, high-risk patients were individually evaluated and, in some cases, treated despite these anatomical characteristics. The AneuRx stent-graft (Medtronic, Santa Rosa, CA, USA) was implanted in 230 procedures, the Excluder (W.L. Gore and Associates, Flagstaff, AZ, USA) in 51, the Talent graft (Medtronic World Medical, Sunrise, FL, USA) in 47, the Zenith graft (William Cook Europe, Bjaeverskov, Denmark) in 47, the Anaconda (Sulzer Vascutek, Inchinnan, Scotland, UK) in 11, and the Endologix in 3 (Endologix, Irvine, CA, USA). Endograft configuration included 3 tubes, 380 bifurcated grafts, and 6 aortomonoiliac grafts combined with contralateral iliac occlusion and femorofemoral bypass. EVAR was performed in an operating room by a team of vascular surgeons and interventional radiologists. The decision to use epidural anesthesia was left to the discretion of anesthesiologist and/or patient choice (43% of ASA IV patients and 47% of ASA IV patients, p0.05). Intraoperative imaging was performed with a portable C-arm fluoroscopic device (9000 9800 OEC, GE Medical Systems, Salt Lake City, UT, USA) with digital imaging and road mapping capabilities. Initial assessment of endograft function and position and verification of satisfactory AAA exclusion were evaluated by intraoperative angiography, predischarge color duplex scanning, and plain abdominal radiographs. Physical examination, color duplex scan, and plain abdominal radiography were repeated at 1, 6, and 12 months after the procedure and annually thereafter. Contrast-enhanced CT scan was repeated 1 month after surgery and then annually. Definitions and Statistical Analysis Perioperative and late outcomes in the ASA IV patients were compared to patients with ASA grades I, II or III (ASA IV group) based on perioperative mortality, survival, and failure of AAA exclusion (i.e., rupture of the AAA, late conversion to open repair, aneurysm diameter increase 3 mm, or persisting graftrelated endoleak). Other outcome measures included blood loss, operative time, major morbidity, and length of hospital stay. Endoleak detected either on CT or duplex scan was defined as the presence of contrast enhancement or color and Doppler signal within the aneurysm sac and outside the endograft, respectively. 12 Intergroup comparisons were performed using the 2-sample t test for continuous variables and chi-square or Fisher exact tests for discrete variables. The Wilcoxon rank sum test was used for non-normally distributed variables. Survival was examined by life-table analysis with log-rank test for significance. The influence of 5 confounding variables (gender, age, ASA IV grade, EUROSTAR class E, AAA diameter) on outcome measures was tested by multivariate regression analysis to determine the odds ratio (OR) and in a Cox proportional hazards model to define the hazard ratio (HR), both presented with the 95% confidence intervals (CI). Statistical analysis was conducted with SPSS software (SPSS, Chicago, IL, USA); p0.05 was considered significant. RESULTS Outcome measures for the study groups are compared in Table 4. There were no early conversions to open repair in the ASA IV patients compared to 6 (1.8%) in the ASA IV group (p0.05). Overall perioperative major morbidity (24 events, 6.2%) included 3 acute myocardial infarctions, 1 stroke, 1 case of pancreatitis, 2 cases of congestive heart failure, 3 renal infarctions, 1 intraparenchymal

J ENDOVASC THER ENDOGRAFTING IN HIGH-RISK PATIENTS 739 TABLE 4 Outcome of Endovascular Repair in ASA IV and ASA IV Groups Median hospital stay, d Mean blood loss, ml Mean surgical time, min Major complications Perioperative death Failure of AAA exclusion Late reintervention Late death ASA IV n51 2 446 146 7 (13.7%) 4 (7.8%) 3 (5.9%) 4 (7.8%) 8 (15.7%) ASA IV n338 2 326 125 17 (5.0%) 1 (0.3%) 25 (7.4%) 25 (7.4%) 28 (8.3%) p 0.001 0.02 0.001 ASA American Society of Anesthesiologists, AAA abdominal aortic aneurysm, not significant. renal artery hemorrhage, 1 bowel occlusion, 1 bleeding duodenal ulcer, 2 cases of mucosal colon ischemia, 6 graft limb occlusions, 1 limb amputation, 1 subdural hematoma from accidental trauma, and 1 graft infection (femorofemoral bypass in a patient with an aortomonoiliac device). There were proportionally more major complications in the ASA IV group (13% versus 5%; OR3.0; 95% CI 1.2 to 7.6; p0.02). Five (1.3%) perioperative deaths (within 30 days) were reported: 4 (7.8%) in the ASA IV group and 1 (0.3%) in the ASA IV group (OR28.6; 95% CI 3.1 to 262; p0.001). In the ASA IV group deaths, the causes were (1) congestive heart failure occurring 2 days after EVAR in a patient with respiratory and severe cardiac disease, (2) congestive heart failure occurring 5 days after EVAR in a patient with severe cardiac disease, (3) pulmonary edema 28 days after successful EVAR in a patient with severe cardiac and respiratory disease, and (4) sepsis in a patient with chronic leukemia and a symptomatic AAA. The single perioperative death in the ASA IV patient was due to massive hemorrhage from an intraprocedural aortic rupture. During a median 22-month follow-up (range 1 56), failure of AAA exclusion was noted in 3 (5.9%) patients in the ASA IV group and in 25 (7.4%) from the ASA IV group (p0.05). Thirty-six (9.3%) patients died: 8 (15.7%) in the ASA IV group and 28 (8.3%) in the ASA IV group (p0.05). Causes of late mortality included 8 cases of cardiac failures, 1 bowel occlusion, 1 pulmonary embolism, 8 myocardial infarctions, 8 cancers, 2 traumas, 1 renal failure, 1 suicide, 2 strokes, 3 cases of respiratory insufficiency, and 1 bleeding duodenal ulcer. Life-table estimates of cumulative survival (Figure) were 62.9% in the ASA IV group and 88.0% in the ASA IV group (p0.001) at 30 months; after this interval, the standard error was 9.4%, and the 8 patients at risk did not allow any definitive conclusion. Cox regression analysis demonstrated that ASA IV classification was an independent predictor of late death (HR2.5; 95% CI 1.5 to 5.4; FigureActuarial survival of ASA IV and ASA IV patients (ASA American Society of Anesthesiologists).

740 ENDOGRAFTING IN HIGH-RISK PATIENTS J ENDOVASC THER p0.01) while age was not (HR1.08; 95% CI 0.98 to 1.09; p0.12), thus excluding the age difference between the groups as a possible confounding factor. AAA diameter was a statistically significant negative predictor of survival, with a hazard ratio of 1.04 per millimeter (95% CI 1.01 to 1.08; p0.001). None of the 5 variables (gender, age, ASA IV score, EU- ROSTAR class E, AAA diameter) had an influence on failure of AAA exclusion according to regression analysis; however, ASA IV was independently associated with perioperative mortality (HR17.8; 95% CI 1.6 to 188; p0.016). DISCUSSION According to our experience, EVAR appears to be safe and effective in the prevention of AAA rupture in ASA IV patients in the mid term. This supports other studies 1 3 in which EVAR was successfully performed in patients unfit for open surgery; however, indications for EVAR in high-risk patients are as yet unclear. Moreover, one must question whether this represents appropriate use of health-care resources or whether it can improve patient survival. Ongoing trials, such as the UK EVAR 2 trial, 13 are designed to focus on this question, comparing survival of patients unfit for open surgery randomized to EVAR or medical treatment. Undoubtedly, the role of aortic endografting will be definitively clarified in the near future. In the meantime, there is little doubt that a portion of patients considered unfit for open surgery will survive long enough to experience AAA rupture if left untreated. In our study, over 60% of ASA IV patients were alive at 30 months. As shown in other studies, 5,6 AAA rupture remains a common cause of death despite coexisting medical conditions. Conway et al. 5 analyzed outcomes of 106 patients with AAA 5.5 cm in diameter who were rejected for elective open repair; all deaths were certified. They found that the 3- year survival rate was only 17%: ruptured aneurysm was the cause of death in 49% of these patients, and the median time of certified rupture was 9 months. Similarly, Jones et al. 6 reported their experience with 57 patients rejected for elective repair of AAAs 5 cm. Overall, 19 (38%) of 50 deaths were attributed to aneurysm rupture. Both populations had much shorter survival times compared to ours, but their mean ages were 78 and 81 years, respectively, while ours was 75 years. Our results correlate well with those reported by Riambau et al., 1 who analyzed data included in the EUROSTAR study. The early/ late mortality rates for 272 patients defined as unfit for open surgery (5.1%/11.4%) and for 109 patients unfit for anesthesia (3.7%/ 11.0%) were significantly worse than those of 2531 low-risk patients (2.7%/5.2%). Actuarial survival at 3 years for EUROSTAR patients unfit for surgery was close to 60%, 1 very similar to our 62.3%. A slightly different figure has been published by Chuter et al., 3 who reported a 30- month survival rate of 75% for a population of high-risk patients after EVAR. 3 Their designation of high risk was based on clinical and laboratory data without applying rigorous risk quantification; only 31% of the patients were classified ASA grade IV, which may explain the better prognosis of this group compared to ours. Considering studies on high-risk patients, the main concern is to evaluate the efficacy of patient selection. In fact, there is no consensus on the criteria to be employed. Riambau et al. 1 found that mortality was best predicted by the surgeon s definition of unfit for surgery rather than patient classification according to the SVS/ISCVS risk score. In our study, we chose the ASA score, which was objectively assessed by an independent auditor (anesthesiologist), to identify a subgroup of high-risk AAA patients. Our study showed that ASA IV patients have a poorer prognosis compared to their lower-risk counterparts. Perioperative and late mortality rates were both significantly higher in ASA IV patients. Similarly, general morbidity was higher, and hospital stays consequently longer. How do we consider our 7.8% perioperative mortality rate in the ASA IV group? It is satisfactory if compared to the 1% and 10% reported from recent multicenter and singlecenter experiences involving good-risk patients undergoing elective open and endovascular AAA repair. 14 22 On the other hand, it was 15 times higher than the rate for low-risk

J ENDOVASC THER ENDOGRAFTING IN HIGH-RISK PATIENTS 741 patients in our center. These figures may imply that the endovascular criteria were too broad. In some cases, the indication was pushed because we were confident of the technical success of the procedure. However, our data show that perioperative mortality can be completely independent from technical aspects and that a meticulous individualized approach with careful evaluation of comorbidities is necessary. In conclusion, EVAR is a valid therapeutic option in high-risk patients with appropriate anatomical characteristics. Early and late mortality in our patients was not procedure related. However, high-risk patients experienced a substantially worse prognosis compared to their good-risk counterparts. Until data from randomized trials are available, a careful and individualized approach is mandatory in the selection of high-risk patients for endoluminal repair. REFERENCES 1. Riambau V, Laheij RJ, García-Madrid C, et al. The association between co-morbidity and mortality after abdominal aortic aneurysm endografting in patients ineligible for elective open surgery. Eur J Vasc Endovasc Surg. 2001; 22:265 270. 2. Sicard GA, Rubin BG, Sanchez LA, et al. Endoluminal graft repair for abdominal aortic aneurysms in high-risk patients and octogenarians: is it better than open repair? Ann Surg. 2001;234:427 437. 3. Chuter TAM, Reilly LM, Faruqi RM, et al. Endovascular aneurysm repair in high-risk patients. J Vasc Surg. 2000;31:122 133. 4. Zannetti S, De Rango P, Parlani G, et al. Endovascular abdominal aortic aneurysm repair in high-risk patients: a single center experience. Eur J Vasc Endovasc Surg. 2001;21:334 338. 5. Conway KP, Byrne J, Townsend M, et al. Prognosis of patients turned down for conventional abdominal aortic aneurysm repair in the endovascular and sonographic era: Szilagyi revisited? J Vasc Surg. 2001;33:752 757. 6. Jones A, Cahill D, Gardham R. Outcome in patients with a large abdominal aortic aneurysm considered unfit for surgery. Br J Surg. 1998; 85:1382 1384. 7. Rutherford RB, Baker JD, Ernst C, et al. Recommended standards for reports dealing with lower extremity ischemia: revised version. J Vasc Surg. 1997;26:517 538. 8. Dentz ME, Grichnik KP, Sibert KS, et al. Anesthesia and postoperative analgesia. In: Sabiston DC, Lyerly HK, eds. Textbook of Surgery. 15th ed. Philadelphia: WB Saunders; 1997:186 206. 9. Keats AS. The ASA classification of physical status a recapitulation. Anesthesiology. 1978; 49:233 236. 10. Malina M, Ivancev K, Chuter TAM, et al. Changing aneurysmal morphology after endovascular grafting: relation to leakage or persistent perfusion. J Endovasc Surg. 1997;4:23 30. 11. Harris PL, Buth J, Mialhe C, et al. The need for clinical trials of endovascular stent-graft techniques for abdominal aortic aneurysm repair: the EUROSTAR project. J Endovasc Surg. 1997; 4:72 77. 12. White GH, Yu W, May J, et al. Endoleak as a complication of endoluminal grafting of abdominal aortic aneurysms: classification, incidence, diagnosis, and management. J Endovasc Surg. 1997;4:152 168. 13. Greenhalgh RM, Brown LC, Powell JT for the UK EVAR Trial Participants. The UK endovascular aneurysm repair (EVAR) trials: background and methodology. In: Greenhalgh RM, Powell JT, Mitchell AW, eds. Vascular and Endovascular Opportunities. London: WB Saunders; 2000:215 227. 14. Zarins CK, White RA, Schwarten D, et al. AneuRx stent graft versus open surgical repair of abdominal aortic aneurysms: multicenter prospective clinical trial. J Vasc Surg. 1999;29: 292 308. 15. Blum U, Voshage G, Lammer J, et al. Endoluminal stent-grafts for infrarenal abdominal aortic aneurysms. N Engl J Med. 1997;336:13 20. 16. May J, White GH, Waugh R, et al. Adverse events after endoluminal repair of abdominal aortic aneurysms: a comparison during two successive periods of time. J Vasc Surg. 1999; 29:32 39. 17. Ohki T, Veith FJ, Shaw P, et al. Increasing incidence of midterm and long-term complications after endovascular graft repair of abdominal aortic aneurysms: a note of caution based on a 9-year experience. Ann Surg. 2001;234:323 335. 18. Ernst CB. Abdominal aortic aneurysm. N Engl J Med. 1993;328:1167 1172. 19. Katz DJ, Stanley JC, Zelenock GB. Operative mortality rates for intact and ruptured abdominal aortic aneurysms in Michigan: an eleven-

742 ENDOGRAFTING IN HIGH-RISK PATIENTS J ENDOVASC THER year statewide experience. J Vasc Surg. 1994; 19:804 817. 20. Kazmiers A, Jacobs L, Perkins A, et al. Abdominal aortic aneurysm repair in Veterans Affairs medical center. J Vasc Surg. 1996;23:191 200. 21. Mortality results from randomised controlled trial of early elective surgery or ultrasonographic surveillance for small abdominal aortic aneurysms. The UK Small Aneurysm Trial Participants. Lancet. 1998;352:1649 1655. 22. Lawrence PF, Gazak C, Bhirangi L, et al. The epidemiology of surgically repaired aneurysms in United States. J Vasc Surg. 1999;30: 632 40.