- Email: [email protected]
Predictors of Survival Following Open and Endovascular Repair of Abdominal Aortic Aneurysms
Clinical Research Predictors of Survival Following Open and Endovascular Repair of Abdominal Aortic Aneurysms Jon S. Matsumura,1 Barry T. Katzen,2 Timothy M. Sullivan,3 Michael D. Dake,4 and David C. Naftel,5 for the Excluder Bifurcated Endoprosthesis Investigators
Clinical decision making for asymptomatic abdominal aortic aneurysms (AAAs) weighs risk of aneurysm rupture, treatment hazards, and overall survival expectations. AAA diameter is the primary parameter in assessing rupture risk. Perioperative risk assessment has been extensively studied, and in-hospital mortality has been reduced to less than 8% with higher-risk open repair and less than 3% with endovascular repair. The purpose of this report is to determine risk factors that predict 2-year survival following open and endovascular AAA repair. We studied 334 patients enrolled in a multicenter clinical trial evaluating an endovascular graft in comparison to standard open repair of infrarenal AAA. Demographic, medical history, physical examination, laboratory, anatomic, procedural, and standardized risk score system variables were analyzed in a multivariable Cox proportional hazard model. Overall survival was 89% at 2 years. Heart disease, cancer, and stroke were the most common causes of death, and no deaths were due to AAA rupture. Cox modeling demonstrated that there were several independent predictors for death after AAA repair: smaller body mass index ( p ¼ 0.005), Society for Vascular Surgery pulmonary risk score 1 ( p ¼ 0.005), history of erectile dysfunction ( p ¼ 0.008), history of heart valve replacement ( p ¼ 0.008), lower preoperative platelet count ( p ¼ 0.012), larger ratio of AAA diameter/proximal neck diameter ( p ¼ 0.020), and lower ankle-brachial index ( p ¼ 0.031). Age, gender, and open or endovascular treatment group are not significant independent risk factors for 2-year mortality in this study. Clinical, laboratory, and anatomic factors predict survival after open and endovascular repair of AAAs. With progressive reduction of in-hospital mortality, assessment of patient longevity after AAA repair has become a more important factor in clinical decision making. Use of valid predictors of patient survival will optimize resource utilization and improve overall patient outcomes. Better selection of patients for any method of repair may improve overall utility more than choice of open or endovascular techniques.
1 Department of Surgery, Northwestern University, Feinberg School of Medicine, Chicago, Illinois. 2
Chief Medical Officer Baptist Cardiac and Vascular Institute, Miami, Florida. 3 Chairman, Vascular Endovascular Surgery, Manneapolis Heart Institute, Minneapolis, Minnesota, South Dakota. 4 Department of Cardiothoracic Surgery, Stanfod University Stanford, CA University of Virginia, Charlottesville, Virginia. 5 Department of Surgery University of Alabama at Birmingham, Birmingham, Alabama.
Correspondence to: Jon S. Matsumura, MD, Department of Surgery, Northwestern University, Feinberg School of Medicine, 201 E. Huron, Suite 10-105, Chicago, IL 60611, E-mail: [email protected] Ann Vasc Surg 2009; 23: 153-158 DOI: 10.1016/j.avsg.2008.07.006 Ó Annals of Vascular Surgery Inc. Published online: September 6, 2008
INTRODUCTION Abdominal aortic aneurysms (AAAs) are commonly diagnosed in older patients. There is no proven medical treatment, and decision making is limited to three options: watchful waiting, standard open surgical repair and endovascular repair. The latter two options are associated with significant morbidity and mortality, occasional prolonged recovery, and late complications. Because of these limitations, many patients and their physicians choose to defer any intervention. In weighing these options, much is known about the risk of aneurysm rupture; specifically diameter of the AAA and gender have been correlated with rupture.1,2 Similarly, the 30-day 153
154 Matsumura et al.
periprocedural mortality hazards of both open and endovascular repair have been well defined. While data exist on predictors of long-term complications and survival after open surgical repair, less information is available on risk factors for late mortality after endovascular repair.3-14 For asymptomatic patients with borderline risk of AAA rupture or high risk of interventional treatment, predicting survival becomes a critical element in medical decision making. Patients must expect to live long enough to benefit from a prophylactic repair of the AAA by any technique. For example, EVAR Trial 2 showed that prophylactic repair was not beneficial for survival in a group of asymptomatic ‘‘unfit’’ patients with relatively high perioperative and late mortality.15 The goal of this study is to identify clinical, laboratory, anatomic, or procedural predictors of survival following AAA repair. Application of accurate risk factors could help stratify patients who gain more or less benefit from repair.
METHODS Analysis was performed on data from a U.S. Food and Drug Administration-approved, multicenter, prospective trial comparing standard open repair of infrarenal AAA to treatment with the Excluder (W. L. Gore & Associates, Inc., Sunnyvale, CA). The study was conducted at 19 U.S. centers, and participating investigators have been previously published.16 Local institutional review boards approved the study, and all subjects provided informed consent for participation. Patients in the open repair group had standard aneurysmorrhaphy with placement of a sutured fabric graft, and endovascular patients had repair with a modular bifurcated endoprosthesis system that is performed by femoral cutdown or percutaneous techniques.17 Over 100 clinical characteristics, laboratory values, standardized risk scoring systems, and detailed anatomic measurements (>40) were systematically recorded on case report forms that were verified against source documents by a contract clinical research organization. Earlier reporting standards were used when applicable, as later reporting standards criteria were not available during the design of this trial.18-22 Two-year survival was selected for primary analysis because perioperative mortality may approach 8% in high-risk patients, and the annualized risk of rupture is about 5% to 10% for AAAs of 5.5 to 6 cm. For an individual patient with a large aneurysm, surviving at least 3 years after repair is likely to result in an overall benefit after suffering the
Annals of Vascular Surgery
initial risk of repair. Therefore, more accurate prediction of 2-year survival is a practical point at which better discrimination would be expected to be useful. Further, clinicians frequently use this time horizon in discussions with patients, and entry into these trials is often conditional upon a predicted survival of 2 years. For this publication, the authors were involved in data acquisition and analysis, had full access to case report forms and reviewed selected source documents, wrote the article, and approved the final manuscript version. The sponsor scanned the paper only for proprietary information, while the authors retained final review authority. Statistics Cox proportional hazard multivariable regression was used for analysis of independent risk factors for survival. Kaplan-Meier curves were used to depict event frequencies over time, and a dotted line indicates the standard error is over 10%. Stratified actuarials were constructed in similar fashion based on univariable analysis. Comparisons of KaplanMeier curves were done with the log-rank test.
RESULTS A total of 334 patients were enrolled in the pivotal trial; 99 subjects had open repair and 235 patients had endovascular repair. Survival curves for the two groups are depicted in Figure 1. As previously reported, there is no significant difference by logrank ( p ¼ 0.13). Further, when forced into the Cox model, treatment group was not a statistically significant independent predictor of mortality (hazard ratio 1.448, 95% CL 0.625-3.354, p ¼ 0.388).16 Most deaths were due to cardiac disease (Table 1), and none were related to device or graft failure or aneurysm rupture. The multivariable analysis demonstrated several independent risk factors for death.(Table 2). They are smaller body mass index ( p ¼ 0.005), Society for Vascular Surgery (SVS) pulmonary risk score 1 ( p ¼ 0.005), history of erectile dysfunction ( p ¼ 0.008), history of heart valve replacement ( p ¼ 0.008), lower preoperative platelet count ( p ¼ 0.012), larger ratio of AAA diameter/proximal neck diameter ( p ¼ 0.020), and lower ankle-brachial index ( p ¼ 0.031). Figure 2 is a Kaplan-Meier stratified actuarial that demonstrates the predictive value of pretreatment erectile dysfunction in this model. There is a 15% survival difference at 16 months between male patients with and without preoperative erectile
Vol. 23, No. 2, March 2009
Survival after repair of AAAs 155
Table 2. Cox multivariable model of late mortality
Fig. 1. Kaplan-Meier curves depicting subject survival by group, with control (squares) and test (circles) curves that are similar. (Modified from Matsumura et al.16)
Risk factor
Hazard ratio [95% CL]
p-Value
Smaller body mass index SVS pulmonary risk score 1 History of erectile dysfunction History of heart valve replacement Platelet count (103/mm3) (lower) AAA diameter/proximal neck diameter (mm) Ankle brachial index (lower) Endovascular group
0.288 [0.121-0.689]
0.005
2.573 [1.327-4.988]
0.005
2.842 [1.314-6.148]
0.008
4.260 [1.451-12.509] 0.008 0.992 [0.986-0.998]
0.012
1.031 [1.005-1.059]
0.020
0.092 [0.010-0.804]
0.031
1.448 [0.625-3.354]
0.388
Table 1. Causes of death Cause
Cardiovascular Myocardial infarction Stroke Congestive heart failure Other cardiac Aneurysm rupture Cancer Lung Gastrointestinal Prostate Miscellaneous Pneumonia Liver failure Urosepsis Motor vehicle accident Pulmonary fibrosis Ischemic bowel Gunshot Anaphylactic reaction
First year (N ¼ 22)
Second year (N ¼ 18)
9 2 3 2 2 0 4 1 3
7 3 2 1 1 0 8 5 2 1 3 1 1
9 2 1 2 1 1 1 1
1
dysfunction. Women’s survival is equivalent to men’s without erectile dysfunction. Figure 3 shows a similarly strong predictive value of body mass index. Figure 4 illustrates the predictive value of SVS pulmonary risk score.
DISCUSSION Prophylactic treatment of asymptomatic diseases should be performed only when the benefits to an individual outweigh the risks.23 While the hazards of acute complications and mortality have been well defined for interventional therapy of AAAs,
Fig. 2. Stratified actuarials of predicted Kaplan-Meier survival curves for male patients with and without erectile dysfunction and female patients.
the risks of rupture are cumulative depending on the annual risk of rupture and the expected survival of the patient.24-32 This study attempts to identify predictors of survival so that the cumulative annual risk of rupture can be more accurately estimated for an individual. Several factors that predict poor survival were identified in this trial. When comparing these to the literature, some factors are previously known to be associated with late death in studies after open surgical repair. The Association for Academic Research in Vascular Surgery found overall survival of 67% at 5 years in 794 patients who survived open repair in 1989, and Cox modeling showed that death was associated with larger diameter of aneurysm, left ventricular failure, older age, carotid artery occlusion, choice of surgical incision, cardiac
156 Matsumura et al.
Fig. 3. Stratified actuarials of predicted Kaplan-Meier survival curves for patients by body mass index.
arrhythmia, duration of aortic clamp, myocardial ischemia, abnormality at the upper limit of the aneurysm, and advanced renal insufficiency.10 In a study of 1135 asymptomatic elective infrarenal open repair cases from 1989 to 1998 from the Cleveland Clinic, survival was 75% at 5 years and 49% at 10 years. Cardiac events and cancer were the primary causes of death in 43% of the patients. Age over 75 years, congestive heart failure, chronic pulmonary disease, renal insufficiency, and configuration of the replacement graft were associated with worse late survival in their Cox model.7 The presence of erectile dysfunction was found to be a strong independent predictor of poor survival in men in this study. This risk factor may have been useful because it includes subjective self-assessment of the patient’s health, unlike other variables that are objectively measured or rated. The value of this risk factor has been identified by others. Impotence has been strongly linked to future risk of angina, myocardial infarction, and stroke in a recent study of 8,063 men without pre-existing heart disease.33 The present study has several features that are collectively different from previous publications: 1) it included patients undergoing endovascular repair, 2) detailed anatomic measurements of aortoiliac morphology were analyzed, 3) multiple risk scoring systems were prospectively rated, 4) functional characteristics such as erectile dysfunction were analyzed in addition to commonly collected clinical attributes, and 5) there was third party auditing with cross-examination to source medical documentation, which is expected to be more complete and accurate than registry data. It is important to note that data from a clinical trial that evaluated the safety and effectiveness of a medical device have some limitations in
Annals of Vascular Surgery
Fig. 4. Stratified actuarials of predicted Kaplan-Meier survival curves for patients according to SVS pulmonary risk score.
application to all patients with AAA. The most important limitation is that there were eligibility criteria for this trial that excluded patients with serum creatinine greater than 2.5 mg/dL, ruptured aneurysms, aneurysms that were anatomically unsuitable by prespecified criteria, and known limited life expectancy of less than 2 years. Another limitation is there was no standardized protocol for medical management of chronic comorbidities. This is important given the long-term survival advantage of early repair of small aneurysms versus watchful waiting in the United Kingdom Small Aneurysm Trial.34 Given only 6% of patient mortality was due to unrepaired ruptured AAA, the United Kingdom Trialists hypothesized that smoking cessation and other lifestyle changes may have been responsible for long-term survival advantages. Finally, most patients in this study were treated in 1999, and even in the short interval since then, there have been significant advances in medical treatment of diseases attending this patient group. A fatalistic approach of triaging patients with limited life expectancy, based only on the risk factors identified in this study, could become obsolete. While the academic exercise of predicting survival is possible, translating this to individual patient care is not simple. Clearly, none of these predictive factors is 100% accurate and patient preferences and clinical judgement of other factors must also be factored into clinical decision making. An important hypothesis to test is whether treatment of the identified comorbidities and risk factors will improve long-term survival, and hence improve the utility of prophylactic treatment of an asymptomatic aneurysm. Debate and controversy have accompanied newer, expensive medical treatments, and endovascular
Vol. 23, No. 2, March 2009
repair of AAAs is no exception.23,35,36 Randomized trials have helped elucidate the relative value of endovascular and open repair, and their interpretation remains mixed.37,38 These large trials compared heterogeneous cohorts of low- and moderate-risk patients, and the current study has identified predictors of absolute survival differences considerably larger than mortality differences between treatment groups. Improved selection of patients for any type of repair may improve overall utility more than selection of open or endovascular techniques.
CONCLUSION Several demographic, laboratory, and anatomic factors have been identified that predict mortality following open and endovascular repair of AAAs. In this study, smaller body mass, pulmonary disease, erectile dysfunction, previous valve replacement, thrombocytopenia, relatively larger aneurysms, and lower ankle-brachial index were found to predict poorer survival with up to 20% absolute risk differences at 2 years. Treatment group was not an independent predictor. Inclusion of predictors of survival along with perioperative risk assessment and rupture risk assessment will contribute to better clinical decision making and optimization of overall utility. REFERENCES 1. The UK Small Aneurysm Trial Participants. Mortality results for randomized controlled trial of early elective surgery or ultrasonographic surveillance for small abdominal aortic aneurysms. Lancet 1998;352:1649-1655. 2. Lederle FA, Johnson GR, Wilson SE, et al. Rupture rate of large abdominal aortic aneurysms in patients refusing or unfit for elective repair. JAMA 2002;287:2968-2972. 3. Kalman PG, Rappaport DC, Merchant N, et al. The value of late computed tomographic scanning in identification of vascular abnormalities after abdominal aortic aneurysm repair. J Vasc Surg 1999;29:442-450. 4. Hallett J, Marshall D, Petterson T, Gray D, Bower T, Cherry K, et al. Graft-related complications after abdominal aortic aneurysm repair: Reassurance from a 36-year population-based experience. J Vasc Surg 1997;25:277-286. 5. Edwards J, Teefey S, Zierler R, Kohler T. Intraabdominal paraanastomotic aneurysms after aortic bypass grafting. J Vasc Surg 1992;15:344-353. 6. Kazmers A, Kohler TR. Very late survival after vascular surgery. J Surg Res 2002;105:109-114. 7. Hertzer NR, Mascha EJ, Karafa MT, O’Hara PJ, Krajewski LP, Beven EG. Open infrarenal abdominal aortic aneurysm repair: the Cleveland Clinic experience from 1989 to 1998. J Vasc Surg 2002;35:1145-1154. 8. Crawford ES, Saleh SA, Babb JW, 3rd, Glaeser D, Vaccaro PS, Silvers A. Infrarenal abdominal aortic aneurysm: factors influencing survival after operation performed over a 25-year period. Ann Surg 1981;193:699-709.
Survival after repair of AAAs 157
9. Johnston KW the Canadian Society for Vascular Surgery Aneurysm Study Group. Nonruptured abdominal aortic aneurysm: six-year follow-up results from the multicenter prospective Canadian aneurysm study. J Vasc Surg 1994;20:163-169. 10. Koskas F, Kieffer E. Long-term survival after elective repair of infrarenal abdominal aortic aneurysm: results of a prospective multicentric study. Ann Vasc Surg 1997;11:473-481. 11. Berry AJ, Smith RB, III, Weintraub WS, Chaikof EL, Dodson TF, Lumsden AB, et al. Age versus comorbidities as risk factors for complications after elective abdominal aortic reconstructive surgery. J Vasc Surg 2001;33:345-352. 12. Brady AR, Fowkes FGR, Greenhalgh RM, Powell TC, Ruckley CV. Thompson SG, on behalf of the UK Small Aneurysm Trial participants. Risk factors for postoperative death following elective surgical repair of abdominal aortic aneurysm: Results from the UK Small Aneurysm Trial Br J Surg 2000;87:742-749. 13. Reigel MM, Holllier LH, Kazmier FJ, O’Brien PC, Paiolero PC, Cherry KJ Jr, et al. Late survival in abdominal aortic aneurysm patients: The role of selective myocardial revascularization on the basis of clinical symptoms. J Vasc Surg 1987;5:222-227. 14. Aune S, Amundsen SR, Evjensvold J, Trippestad A. Operative mortality and long-term relative survival of patients operated on for asymptomatic abdominal aortic aneurysm. Eur J Vasc Endovasc Surg 1995;9:293-298. 15. EVAR Trial Participants. Endovascular aneurysm repair and outcome in patients unfit for open repair of abdominal aortic aneurysm (EVAR trial 2): randomised controlled trial. Lancet 2005;365:2187-2192. 16. Matsumura JS, Brewster DC, Makaroun MS, Naftel DC. A multicenter controlled clinical trial of open versus endovascular treatment of abdominal aortic aneurysms. J Vasc Surg 2003;37:262-271. 17. Morasch MD, Kibbe MR, Evans ME, Meadows WS, Eskandari MK, Matsumura JS, Pearce WH. Percutaneous repair of abdominal aortic aneurysm. J Vasc Surg 2004;40: 12-16. 18. Sacks D, Marinelli D, Martin L, Spies J and members of Technology Assessment Committee SCVIR. Reporting standards for clinical evaluation of new peripheral arterial revascularization devices. JVIR 1997;8:137-149. 19. Ahn SS, Rutherford RB, Johnston KW, et al. Reporting standards for infrarenal endovascular abdominal aortic aneurysm repair. J Vasc Surg 1997;25:405-410. 20. 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. 21. Chaikof EL, Fillinger MF, Matsumura JS, Rutherford RB, White GH, Blankensteijn JD, et al. Identifying and grading factors that modify the outcome of endovascular aortic aneurysm repair. J Vasc Surg 2002;35:1061-1066. 22. Chaikof EL, Blankensteijn JD, Harris PL,White GH, Zarins CK, Bernhard VM, et al. Reporting standards for endovascular aortic aneurysm repair. J Vasc Surg 2002;35: 1048-1060. 23. Brewster DC, Cronenwett JL, Hallett JW, Jr, et al. for the Joint Council of the American Association for Vascular Surgery and Society for Vascular Surgery. Guidelines for the treatment of abdominal aortic aneurysms. Report of a subcommittee of the Joint Council. J Vasc Surg 2003;37:1106-1117. 24. Williamson W, Nicoloff A, Taylor L, Moneta G, et al. Functional outcome after open repair of abdominal aortic aneurysm. J Vasc Surg 2001;33:913-920.
158 Matsumura et al.
25. May J, White GH, Yu W, et al. Concurrent comparison of endoluminal versus open repair in treatment of abdominal aortic aneurysms: analysis of 303 patients by life-table method. J Vasc Surg 1998;27:213-221. 26. Zarins CK, White RA, Schwarten D, et al. for the Investigators of the Medtronic AneuRx Multicenter Clinical Trial. AneuRx stent graft vs. open surgical repair of abdominal aortic aneurysms: Multicenter prospective clinical trial. J Vasc Surg 1999;29:292-308. 27. Brewster DC, Geller SC, Kaufman JA, et al. Initial experience with endovascular aneurysm repair: comparison of early results with outcome of conventional open repair. J Vasc Surg 1998;27:992-1003. 28. Moore W, Kashyap V, Vescera C, Quinones-Baldrich W. Abdominal aortic aneurysm: A 6-year comparison of endovascular versus transabdominal repair. Ann Surg 1999;230:298-306. 29. Teufelsbauer H, Prusa AM, Wolff K, Polerauer P, Nanobashvili J, Prager M, et al. Endovascular stent grafting versus open surgical operation in patients with infrarenal aortic aneurysms. A propensity score-adjusted analysis. Circulation 2002;106:782-787. 30. Anderson PL, Arons RR, Moskowitz AJ, et al. A statewide experience with endovascular abdominal aortic aneurysm repair: rapid diffusion with excellent early results. J Vasc Surg 2004;(1):10-19. 31. Lee WA, Carter JW, Upchurch G, Seeger JM, Huber TS. Perioperative outcomes after open and endovascular repair of
Annals of Vascular Surgery
32.
33.
34.
35.
36.
37.
38.
intact abdominal aortic aneurysms in the United States during 2001. J Vasc Surg 2004;39(3):491-496. Prinssen M, Verhoeven ELG, Buth J, et al. A randomized trial comparing conventional and endovascular repair of abdominal aortic aneurysms. N Engl J Med 2004;351: 1607-1618. Thompson IM, Tangen CM, Goodman PJ, Probstfield JL, Moinpour CM, Coltman CA. Erectile dysfunction and subsequent cardiovascular disease. JAMA 2005;294:2996-3002. United Kingdom Small Aneurysm Trial Participants. Longterm outcomes of immediate repair compared with surveillance of small abdominal aortic aneurysms. N Engl J Med 2002;346:1445-1452. Sapirstein W, Chandeysson P, Wentz C. The Food and Drug Administration approval of endovascular grafts for abdominal aortic aneurysm: an 18-month retrospective. J Vasc Surg 2001;34:180-183. Collin J, Murie JA. Endovascular treatment of abdominal aortic aneurysm: a failed experiment. BJS 2001;88: 1281-1282. EVAR Trial Participants. Endovascular aneurysm repair versus open repair in patients with abdominal aortic aneurysm (EVAR trial 1): randomised controlled trial. Lancet 2005;365:2179. Blankensteijn JD, de Jong SE, Prinssen M, et al. Two-year outcomes after conventional or endovascular repair of abdominal aortic aneurysms. N Engl J Med 2005;352: 2398-2405.
Clinical decision making for asymptomatic abdominal aortic aneurysms (AAAs) weighs risk of aneurysm rupture, treatment hazards, and overall survival expectations. AAA diameter is the primary parameter in assessing rupture risk. Perioperative risk assessment has been extensively studied, and in-hospital mortality has been reduced to less than 8% with higher-risk open repair and less than 3% with endovascular repair. The purpose of this report is to determine risk factors that predict 2-year survival following open and endovascular AAA repair. We studied 334 patients enrolled in a multicenter clinical trial evaluating an endovascular graft in comparison to standard open repair of infrarenal AAA. Demographic, medical history, physical examination, laboratory, anatomic, procedural, and standardized risk score system variables were analyzed in a multivariable Cox proportional hazard model. Overall survival was 89% at 2 years. Heart disease, cancer, and stroke were the most common causes of death, and no deaths were due to AAA rupture. Cox modeling demonstrated that there were several independent predictors for death after AAA repair: smaller body mass index ( p ¼ 0.005), Society for Vascular Surgery pulmonary risk score 1 ( p ¼ 0.005), history of erectile dysfunction ( p ¼ 0.008), history of heart valve replacement ( p ¼ 0.008), lower preoperative platelet count ( p ¼ 0.012), larger ratio of AAA diameter/proximal neck diameter ( p ¼ 0.020), and lower ankle-brachial index ( p ¼ 0.031). Age, gender, and open or endovascular treatment group are not significant independent risk factors for 2-year mortality in this study. Clinical, laboratory, and anatomic factors predict survival after open and endovascular repair of AAAs. With progressive reduction of in-hospital mortality, assessment of patient longevity after AAA repair has become a more important factor in clinical decision making. Use of valid predictors of patient survival will optimize resource utilization and improve overall patient outcomes. Better selection of patients for any method of repair may improve overall utility more than choice of open or endovascular techniques.
1 Department of Surgery, Northwestern University, Feinberg School of Medicine, Chicago, Illinois. 2
Chief Medical Officer Baptist Cardiac and Vascular Institute, Miami, Florida. 3 Chairman, Vascular Endovascular Surgery, Manneapolis Heart Institute, Minneapolis, Minnesota, South Dakota. 4 Department of Cardiothoracic Surgery, Stanfod University Stanford, CA University of Virginia, Charlottesville, Virginia. 5 Department of Surgery University of Alabama at Birmingham, Birmingham, Alabama.
Correspondence to: Jon S. Matsumura, MD, Department of Surgery, Northwestern University, Feinberg School of Medicine, 201 E. Huron, Suite 10-105, Chicago, IL 60611, E-mail: [email protected] Ann Vasc Surg 2009; 23: 153-158 DOI: 10.1016/j.avsg.2008.07.006 Ó Annals of Vascular Surgery Inc. Published online: September 6, 2008
INTRODUCTION Abdominal aortic aneurysms (AAAs) are commonly diagnosed in older patients. There is no proven medical treatment, and decision making is limited to three options: watchful waiting, standard open surgical repair and endovascular repair. The latter two options are associated with significant morbidity and mortality, occasional prolonged recovery, and late complications. Because of these limitations, many patients and their physicians choose to defer any intervention. In weighing these options, much is known about the risk of aneurysm rupture; specifically diameter of the AAA and gender have been correlated with rupture.1,2 Similarly, the 30-day 153
154 Matsumura et al.
periprocedural mortality hazards of both open and endovascular repair have been well defined. While data exist on predictors of long-term complications and survival after open surgical repair, less information is available on risk factors for late mortality after endovascular repair.3-14 For asymptomatic patients with borderline risk of AAA rupture or high risk of interventional treatment, predicting survival becomes a critical element in medical decision making. Patients must expect to live long enough to benefit from a prophylactic repair of the AAA by any technique. For example, EVAR Trial 2 showed that prophylactic repair was not beneficial for survival in a group of asymptomatic ‘‘unfit’’ patients with relatively high perioperative and late mortality.15 The goal of this study is to identify clinical, laboratory, anatomic, or procedural predictors of survival following AAA repair. Application of accurate risk factors could help stratify patients who gain more or less benefit from repair.
METHODS Analysis was performed on data from a U.S. Food and Drug Administration-approved, multicenter, prospective trial comparing standard open repair of infrarenal AAA to treatment with the Excluder (W. L. Gore & Associates, Inc., Sunnyvale, CA). The study was conducted at 19 U.S. centers, and participating investigators have been previously published.16 Local institutional review boards approved the study, and all subjects provided informed consent for participation. Patients in the open repair group had standard aneurysmorrhaphy with placement of a sutured fabric graft, and endovascular patients had repair with a modular bifurcated endoprosthesis system that is performed by femoral cutdown or percutaneous techniques.17 Over 100 clinical characteristics, laboratory values, standardized risk scoring systems, and detailed anatomic measurements (>40) were systematically recorded on case report forms that were verified against source documents by a contract clinical research organization. Earlier reporting standards were used when applicable, as later reporting standards criteria were not available during the design of this trial.18-22 Two-year survival was selected for primary analysis because perioperative mortality may approach 8% in high-risk patients, and the annualized risk of rupture is about 5% to 10% for AAAs of 5.5 to 6 cm. For an individual patient with a large aneurysm, surviving at least 3 years after repair is likely to result in an overall benefit after suffering the
Annals of Vascular Surgery
initial risk of repair. Therefore, more accurate prediction of 2-year survival is a practical point at which better discrimination would be expected to be useful. Further, clinicians frequently use this time horizon in discussions with patients, and entry into these trials is often conditional upon a predicted survival of 2 years. For this publication, the authors were involved in data acquisition and analysis, had full access to case report forms and reviewed selected source documents, wrote the article, and approved the final manuscript version. The sponsor scanned the paper only for proprietary information, while the authors retained final review authority. Statistics Cox proportional hazard multivariable regression was used for analysis of independent risk factors for survival. Kaplan-Meier curves were used to depict event frequencies over time, and a dotted line indicates the standard error is over 10%. Stratified actuarials were constructed in similar fashion based on univariable analysis. Comparisons of KaplanMeier curves were done with the log-rank test.
RESULTS A total of 334 patients were enrolled in the pivotal trial; 99 subjects had open repair and 235 patients had endovascular repair. Survival curves for the two groups are depicted in Figure 1. As previously reported, there is no significant difference by logrank ( p ¼ 0.13). Further, when forced into the Cox model, treatment group was not a statistically significant independent predictor of mortality (hazard ratio 1.448, 95% CL 0.625-3.354, p ¼ 0.388).16 Most deaths were due to cardiac disease (Table 1), and none were related to device or graft failure or aneurysm rupture. The multivariable analysis demonstrated several independent risk factors for death.(Table 2). They are smaller body mass index ( p ¼ 0.005), Society for Vascular Surgery (SVS) pulmonary risk score 1 ( p ¼ 0.005), history of erectile dysfunction ( p ¼ 0.008), history of heart valve replacement ( p ¼ 0.008), lower preoperative platelet count ( p ¼ 0.012), larger ratio of AAA diameter/proximal neck diameter ( p ¼ 0.020), and lower ankle-brachial index ( p ¼ 0.031). Figure 2 is a Kaplan-Meier stratified actuarial that demonstrates the predictive value of pretreatment erectile dysfunction in this model. There is a 15% survival difference at 16 months between male patients with and without preoperative erectile
Vol. 23, No. 2, March 2009
Survival after repair of AAAs 155
Table 2. Cox multivariable model of late mortality
Fig. 1. Kaplan-Meier curves depicting subject survival by group, with control (squares) and test (circles) curves that are similar. (Modified from Matsumura et al.16)
Risk factor
Hazard ratio [95% CL]
p-Value
Smaller body mass index SVS pulmonary risk score 1 History of erectile dysfunction History of heart valve replacement Platelet count (103/mm3) (lower) AAA diameter/proximal neck diameter (mm) Ankle brachial index (lower) Endovascular group
0.288 [0.121-0.689]
0.005
2.573 [1.327-4.988]
0.005
2.842 [1.314-6.148]
0.008
4.260 [1.451-12.509] 0.008 0.992 [0.986-0.998]
0.012
1.031 [1.005-1.059]
0.020
0.092 [0.010-0.804]
0.031
1.448 [0.625-3.354]
0.388
Table 1. Causes of death Cause
Cardiovascular Myocardial infarction Stroke Congestive heart failure Other cardiac Aneurysm rupture Cancer Lung Gastrointestinal Prostate Miscellaneous Pneumonia Liver failure Urosepsis Motor vehicle accident Pulmonary fibrosis Ischemic bowel Gunshot Anaphylactic reaction
First year (N ¼ 22)
Second year (N ¼ 18)
9 2 3 2 2 0 4 1 3
7 3 2 1 1 0 8 5 2 1 3 1 1
9 2 1 2 1 1 1 1
1
dysfunction. Women’s survival is equivalent to men’s without erectile dysfunction. Figure 3 shows a similarly strong predictive value of body mass index. Figure 4 illustrates the predictive value of SVS pulmonary risk score.
DISCUSSION Prophylactic treatment of asymptomatic diseases should be performed only when the benefits to an individual outweigh the risks.23 While the hazards of acute complications and mortality have been well defined for interventional therapy of AAAs,
Fig. 2. Stratified actuarials of predicted Kaplan-Meier survival curves for male patients with and without erectile dysfunction and female patients.
the risks of rupture are cumulative depending on the annual risk of rupture and the expected survival of the patient.24-32 This study attempts to identify predictors of survival so that the cumulative annual risk of rupture can be more accurately estimated for an individual. Several factors that predict poor survival were identified in this trial. When comparing these to the literature, some factors are previously known to be associated with late death in studies after open surgical repair. The Association for Academic Research in Vascular Surgery found overall survival of 67% at 5 years in 794 patients who survived open repair in 1989, and Cox modeling showed that death was associated with larger diameter of aneurysm, left ventricular failure, older age, carotid artery occlusion, choice of surgical incision, cardiac
156 Matsumura et al.
Fig. 3. Stratified actuarials of predicted Kaplan-Meier survival curves for patients by body mass index.
arrhythmia, duration of aortic clamp, myocardial ischemia, abnormality at the upper limit of the aneurysm, and advanced renal insufficiency.10 In a study of 1135 asymptomatic elective infrarenal open repair cases from 1989 to 1998 from the Cleveland Clinic, survival was 75% at 5 years and 49% at 10 years. Cardiac events and cancer were the primary causes of death in 43% of the patients. Age over 75 years, congestive heart failure, chronic pulmonary disease, renal insufficiency, and configuration of the replacement graft were associated with worse late survival in their Cox model.7 The presence of erectile dysfunction was found to be a strong independent predictor of poor survival in men in this study. This risk factor may have been useful because it includes subjective self-assessment of the patient’s health, unlike other variables that are objectively measured or rated. The value of this risk factor has been identified by others. Impotence has been strongly linked to future risk of angina, myocardial infarction, and stroke in a recent study of 8,063 men without pre-existing heart disease.33 The present study has several features that are collectively different from previous publications: 1) it included patients undergoing endovascular repair, 2) detailed anatomic measurements of aortoiliac morphology were analyzed, 3) multiple risk scoring systems were prospectively rated, 4) functional characteristics such as erectile dysfunction were analyzed in addition to commonly collected clinical attributes, and 5) there was third party auditing with cross-examination to source medical documentation, which is expected to be more complete and accurate than registry data. It is important to note that data from a clinical trial that evaluated the safety and effectiveness of a medical device have some limitations in
Annals of Vascular Surgery
Fig. 4. Stratified actuarials of predicted Kaplan-Meier survival curves for patients according to SVS pulmonary risk score.
application to all patients with AAA. The most important limitation is that there were eligibility criteria for this trial that excluded patients with serum creatinine greater than 2.5 mg/dL, ruptured aneurysms, aneurysms that were anatomically unsuitable by prespecified criteria, and known limited life expectancy of less than 2 years. Another limitation is there was no standardized protocol for medical management of chronic comorbidities. This is important given the long-term survival advantage of early repair of small aneurysms versus watchful waiting in the United Kingdom Small Aneurysm Trial.34 Given only 6% of patient mortality was due to unrepaired ruptured AAA, the United Kingdom Trialists hypothesized that smoking cessation and other lifestyle changes may have been responsible for long-term survival advantages. Finally, most patients in this study were treated in 1999, and even in the short interval since then, there have been significant advances in medical treatment of diseases attending this patient group. A fatalistic approach of triaging patients with limited life expectancy, based only on the risk factors identified in this study, could become obsolete. While the academic exercise of predicting survival is possible, translating this to individual patient care is not simple. Clearly, none of these predictive factors is 100% accurate and patient preferences and clinical judgement of other factors must also be factored into clinical decision making. An important hypothesis to test is whether treatment of the identified comorbidities and risk factors will improve long-term survival, and hence improve the utility of prophylactic treatment of an asymptomatic aneurysm. Debate and controversy have accompanied newer, expensive medical treatments, and endovascular
Vol. 23, No. 2, March 2009
repair of AAAs is no exception.23,35,36 Randomized trials have helped elucidate the relative value of endovascular and open repair, and their interpretation remains mixed.37,38 These large trials compared heterogeneous cohorts of low- and moderate-risk patients, and the current study has identified predictors of absolute survival differences considerably larger than mortality differences between treatment groups. Improved selection of patients for any type of repair may improve overall utility more than selection of open or endovascular techniques.
CONCLUSION Several demographic, laboratory, and anatomic factors have been identified that predict mortality following open and endovascular repair of AAAs. In this study, smaller body mass, pulmonary disease, erectile dysfunction, previous valve replacement, thrombocytopenia, relatively larger aneurysms, and lower ankle-brachial index were found to predict poorer survival with up to 20% absolute risk differences at 2 years. Treatment group was not an independent predictor. Inclusion of predictors of survival along with perioperative risk assessment and rupture risk assessment will contribute to better clinical decision making and optimization of overall utility. REFERENCES 1. The UK Small Aneurysm Trial Participants. Mortality results for randomized controlled trial of early elective surgery or ultrasonographic surveillance for small abdominal aortic aneurysms. Lancet 1998;352:1649-1655. 2. Lederle FA, Johnson GR, Wilson SE, et al. Rupture rate of large abdominal aortic aneurysms in patients refusing or unfit for elective repair. JAMA 2002;287:2968-2972. 3. Kalman PG, Rappaport DC, Merchant N, et al. The value of late computed tomographic scanning in identification of vascular abnormalities after abdominal aortic aneurysm repair. J Vasc Surg 1999;29:442-450. 4. Hallett J, Marshall D, Petterson T, Gray D, Bower T, Cherry K, et al. Graft-related complications after abdominal aortic aneurysm repair: Reassurance from a 36-year population-based experience. J Vasc Surg 1997;25:277-286. 5. Edwards J, Teefey S, Zierler R, Kohler T. Intraabdominal paraanastomotic aneurysms after aortic bypass grafting. J Vasc Surg 1992;15:344-353. 6. Kazmers A, Kohler TR. Very late survival after vascular surgery. J Surg Res 2002;105:109-114. 7. Hertzer NR, Mascha EJ, Karafa MT, O’Hara PJ, Krajewski LP, Beven EG. Open infrarenal abdominal aortic aneurysm repair: the Cleveland Clinic experience from 1989 to 1998. J Vasc Surg 2002;35:1145-1154. 8. Crawford ES, Saleh SA, Babb JW, 3rd, Glaeser D, Vaccaro PS, Silvers A. Infrarenal abdominal aortic aneurysm: factors influencing survival after operation performed over a 25-year period. Ann Surg 1981;193:699-709.
Survival after repair of AAAs 157
9. Johnston KW the Canadian Society for Vascular Surgery Aneurysm Study Group. Nonruptured abdominal aortic aneurysm: six-year follow-up results from the multicenter prospective Canadian aneurysm study. J Vasc Surg 1994;20:163-169. 10. Koskas F, Kieffer E. Long-term survival after elective repair of infrarenal abdominal aortic aneurysm: results of a prospective multicentric study. Ann Vasc Surg 1997;11:473-481. 11. Berry AJ, Smith RB, III, Weintraub WS, Chaikof EL, Dodson TF, Lumsden AB, et al. Age versus comorbidities as risk factors for complications after elective abdominal aortic reconstructive surgery. J Vasc Surg 2001;33:345-352. 12. Brady AR, Fowkes FGR, Greenhalgh RM, Powell TC, Ruckley CV. Thompson SG, on behalf of the UK Small Aneurysm Trial participants. Risk factors for postoperative death following elective surgical repair of abdominal aortic aneurysm: Results from the UK Small Aneurysm Trial Br J Surg 2000;87:742-749. 13. Reigel MM, Holllier LH, Kazmier FJ, O’Brien PC, Paiolero PC, Cherry KJ Jr, et al. Late survival in abdominal aortic aneurysm patients: The role of selective myocardial revascularization on the basis of clinical symptoms. J Vasc Surg 1987;5:222-227. 14. Aune S, Amundsen SR, Evjensvold J, Trippestad A. Operative mortality and long-term relative survival of patients operated on for asymptomatic abdominal aortic aneurysm. Eur J Vasc Endovasc Surg 1995;9:293-298. 15. EVAR Trial Participants. Endovascular aneurysm repair and outcome in patients unfit for open repair of abdominal aortic aneurysm (EVAR trial 2): randomised controlled trial. Lancet 2005;365:2187-2192. 16. Matsumura JS, Brewster DC, Makaroun MS, Naftel DC. A multicenter controlled clinical trial of open versus endovascular treatment of abdominal aortic aneurysms. J Vasc Surg 2003;37:262-271. 17. Morasch MD, Kibbe MR, Evans ME, Meadows WS, Eskandari MK, Matsumura JS, Pearce WH. Percutaneous repair of abdominal aortic aneurysm. J Vasc Surg 2004;40: 12-16. 18. Sacks D, Marinelli D, Martin L, Spies J and members of Technology Assessment Committee SCVIR. Reporting standards for clinical evaluation of new peripheral arterial revascularization devices. JVIR 1997;8:137-149. 19. Ahn SS, Rutherford RB, Johnston KW, et al. Reporting standards for infrarenal endovascular abdominal aortic aneurysm repair. J Vasc Surg 1997;25:405-410. 20. 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. 21. Chaikof EL, Fillinger MF, Matsumura JS, Rutherford RB, White GH, Blankensteijn JD, et al. Identifying and grading factors that modify the outcome of endovascular aortic aneurysm repair. J Vasc Surg 2002;35:1061-1066. 22. Chaikof EL, Blankensteijn JD, Harris PL,White GH, Zarins CK, Bernhard VM, et al. Reporting standards for endovascular aortic aneurysm repair. J Vasc Surg 2002;35: 1048-1060. 23. Brewster DC, Cronenwett JL, Hallett JW, Jr, et al. for the Joint Council of the American Association for Vascular Surgery and Society for Vascular Surgery. Guidelines for the treatment of abdominal aortic aneurysms. Report of a subcommittee of the Joint Council. J Vasc Surg 2003;37:1106-1117. 24. Williamson W, Nicoloff A, Taylor L, Moneta G, et al. Functional outcome after open repair of abdominal aortic aneurysm. J Vasc Surg 2001;33:913-920.
158 Matsumura et al.
25. May J, White GH, Yu W, et al. Concurrent comparison of endoluminal versus open repair in treatment of abdominal aortic aneurysms: analysis of 303 patients by life-table method. J Vasc Surg 1998;27:213-221. 26. Zarins CK, White RA, Schwarten D, et al. for the Investigators of the Medtronic AneuRx Multicenter Clinical Trial. AneuRx stent graft vs. open surgical repair of abdominal aortic aneurysms: Multicenter prospective clinical trial. J Vasc Surg 1999;29:292-308. 27. Brewster DC, Geller SC, Kaufman JA, et al. Initial experience with endovascular aneurysm repair: comparison of early results with outcome of conventional open repair. J Vasc Surg 1998;27:992-1003. 28. Moore W, Kashyap V, Vescera C, Quinones-Baldrich W. Abdominal aortic aneurysm: A 6-year comparison of endovascular versus transabdominal repair. Ann Surg 1999;230:298-306. 29. Teufelsbauer H, Prusa AM, Wolff K, Polerauer P, Nanobashvili J, Prager M, et al. Endovascular stent grafting versus open surgical operation in patients with infrarenal aortic aneurysms. A propensity score-adjusted analysis. Circulation 2002;106:782-787. 30. Anderson PL, Arons RR, Moskowitz AJ, et al. A statewide experience with endovascular abdominal aortic aneurysm repair: rapid diffusion with excellent early results. J Vasc Surg 2004;(1):10-19. 31. Lee WA, Carter JW, Upchurch G, Seeger JM, Huber TS. Perioperative outcomes after open and endovascular repair of
Annals of Vascular Surgery
32.
33.
34.
35.
36.
37.
38.
intact abdominal aortic aneurysms in the United States during 2001. J Vasc Surg 2004;39(3):491-496. Prinssen M, Verhoeven ELG, Buth J, et al. A randomized trial comparing conventional and endovascular repair of abdominal aortic aneurysms. N Engl J Med 2004;351: 1607-1618. Thompson IM, Tangen CM, Goodman PJ, Probstfield JL, Moinpour CM, Coltman CA. Erectile dysfunction and subsequent cardiovascular disease. JAMA 2005;294:2996-3002. United Kingdom Small Aneurysm Trial Participants. Longterm outcomes of immediate repair compared with surveillance of small abdominal aortic aneurysms. N Engl J Med 2002;346:1445-1452. Sapirstein W, Chandeysson P, Wentz C. The Food and Drug Administration approval of endovascular grafts for abdominal aortic aneurysm: an 18-month retrospective. J Vasc Surg 2001;34:180-183. Collin J, Murie JA. Endovascular treatment of abdominal aortic aneurysm: a failed experiment. BJS 2001;88: 1281-1282. EVAR Trial Participants. Endovascular aneurysm repair versus open repair in patients with abdominal aortic aneurysm (EVAR trial 1): randomised controlled trial. Lancet 2005;365:2179. Blankensteijn JD, de Jong SE, Prinssen M, et al. Two-year outcomes after conventional or endovascular repair of abdominal aortic aneurysms. N Engl J Med 2005;352: 2398-2405.