- Email: [email protected]
Abdominal Aortic Aneurysm Repair in Nonagenarians
Abdominal Aortic Aneurysm Repair in Nonagenarians Kakra Hughes,1 Hamdi Abdulrahman,1 Tahira Prendergast,1 David A. Rose,1 Sharon Ongu’ti,2 Daniel Tran,1 Edward E. Cornwell III,1 Thomas Obisesan,3 and Kwame S. Amankwah,4 Washington, DC; Galveston, Texas; and Syracuse, New York
Background: The feasibility of abdominal aortic aneurysm (AAA) repair in nonagenarians on a national level is largely unknown. We undertook this study to determine the outcomes of open and endovascular AAA repair in this population on a national level. Methods: A retrospective review of the Nationwide Inpatient Sample Database was conducted to determine all patients 90 years and older who underwent either an open or endovascular repair of a nonruptured AAA from 1997 to 2008. Preoperative comorbidities and postoperative complications in the inpatient setting were recorded. The primary end point was mortality. Secondary end points were postoperative neurologic, cardiac, and respiratory complications. This group was then compared with all adult patients less than 90 years old (age, 18e89) who had undergone repair of a nonruptured AAA during this same period. Results: Four hundred twenty-three patients 90 years and older underwent repair of a nonruptured AAA (compared with 52,370 < 90). Of these, 132 patients underwent open repair (31%) and 291 (69%) underwent endovascular repair. Inpatient mortality was 18.3% for the 90 open, 4.6% for the <90 open, 3.1% for the 90 endovascular, and 1.2% for <90 endovascular group. Conclusions: Open repair of AAA’s in nonagenarians is associated with significantly high perioperative mortality, whereas endovascular repair is feasible with acceptable perioperative mortality. This mortality, although significantly higher than that obtained for endovascular repair in patients <90, is nonetheless not significantly different for the mortality noted for patients <90 undergoing open AAA repair.
INTRODUCTION
1 Department of Surgery, Howard University College of Medicine and Hospital, Washington, DC. 2 Department of Internal Medicine, Howard University College of Medicine and Hospital, Washington, DC. 3 Department of Internal Medicine, University of Texas Medical Branch, Galveston, TX. 4 Department of Surgery, Upstate Medical University, Syracuse, NY. Correspondence to: Kakra Hughes, MD, FACS, Department of Surgery, Howard University College of Medicine and Hospital, 2041 Georgia Avenue, NW, # 4B.04, Washington, DC 20060, USA; E-mail: [email protected]
Ann Vasc Surg 2015; 29: 183–188 http://dx.doi.org/10.1016/j.avsg.2014.07.037 Ó 2015 Elsevier Inc. All rights reserved. Manuscript received: March 17, 2014; manuscript accepted: July 30, 2014; published online: October 15, 2014.
Abdominal aortic aneurysms (AAAs) are largely a disease of the elderly. As the nation’s population continues to agedthe US Census Bureau data from 2008 predict that by the year 2050, 20 million individuals, representing approximately 5% of the population, will be older than 85 years1,2dit will be increasingly necessary to treat AAA’s in an elderly population. Kazmers et al. have shown that open aneurysm repair may be beneficial in octogenarians despite the increased mortality associated with their age.2,3 Furthermore, more recent studies have reported the feasibility of endovascular AAA repair in the elderly.4e8 Most of these reports, however, have been single institution studies, with results not necessarily generalizable to the entire 183
184 Hughes et al.
Annals of Vascular Surgery
Table I. Bivariate analysis of patient demographics, comorbidities, and timing of surgery Variable
Gender, % (n) Male Female Ethnicity, % (n) White Black Hispanic Asian Comorbidities, % (n) Cardiac disease COPD Kidney disease Diabetes Elective, % (n) Emergent, % (n)
<90 Open, 59% (N ¼ 30,924)
<90 Endo, 40% (N ¼ 21,033)
90 Open, 0.3% (N ¼ 132)
90 Endo, 0.6% (N ¼ 291)
77.2 (23,874) 22.8 (7,050)
83.2 (17,502) 16.8 (3,531)
67.4 (89) 32.6 (43)
74.9 (218) 25.1 (73)
92.8 3.4 2.8 1.1
(28,687) (1,050) (849) (338)
92.6 3.5 2.7 1.3
(19,475) (730) (560) (268)
87.9 4.6 2.3 5.3
(116) (6) (3) (7)
92.1 3.4 2.4 2.1
(268) (10) (7) (6)
71.8 35.4 1.6 10.4 78.4 21.6
(22,197) (10,945) (503) (3,209) (21,595) (5,958)
78.3 31.4 3.6 15.2 85.7 14.3
(16,467) (6,611) (758) (3,202) (16,035) (2,677)
68.2 21.2 1.5 6.1 69.5 30.5
(90) (28) (2) (8) (82) (36)
69.8 20.3 8.3 6.5 77.8 22.2
(203) (59) (24) (19) (196) (56)
COPD, chronic obstructive pulmonary disease.
Fig. 1. Age distribution of the study population.
nation. We undertook the present study to determine, on a national level, the outcomes of AAA repair in patients 90 years and older.
METHODS The Nationwide Inpatient Sample (NIS) Database was queried to identify all patients who underwent an open or endovascular nonruptured AAA repair between 1998 and 2007 (International Classification of Disease 9 Clinical Modification, ICD 9 CM codes: 38.44 and 39.71, respectively). Patients younger than 18 years were excluded. Patient demographics, comorbidities, and postoperative complications during the inpatient period were compared between patients 90 years and older and patients younger than 90 years. Preoperative comorbidities were defined as cardiac (a history of previous coronary artery bypass grafting or percutaneous coronary intervention, coronary artery disease, and/or congestive heart failure); respiratory (a diagnosis of chronic obstructive pulmonary
disease); renal (chronic kidney disease); and diabetes mellitus. Postoperative complications assessed included: mortality, neurologic (stroke), cardiac, respiratory, and renal complications. Bivariate analysis was conducted using the Pearson chi-squared test for categorical variables and Student’s t-test for continuous variables with a level of significance set at less than 0.05%. Multivariate regression was performed, controlling for patient age, sex, ethnicity, insurance status, and preoperative comorbidities.
RESULTS There were 52,380 patients who were identified in the NIS as having undergone an AAA repair from 1998 to 2007. The population was primarily men (79%, n ¼ 41,683) and white (93%). Open AAA repair was performed in 59% of patients (Table I). Patients aged 90 years or older represented 0.8% of the population (n ¼ 423). Most patients in the study group were in their 70s (46% of the population), followed by patients in their 60s (27% of the population; Fig. 1). Four groups were identified patients <90 years old undergoing open repair represented 59% of the population (n ¼ 30,924); patients < 90 years undergoing endovascular repair represented 40% of the population (n ¼ 21,033); patients 90 years undergoing open repair represented 0.3% of the population (n ¼ 132); and patients 90 years undergoing endovascular repair comprised 0.6% of the population (n ¼ 291; Table I). In all the 4
Vol. 29, No. 2, February 2015
AAA repair in nonagenarians
185
Table II. Bivariate analysis of postoperative complications Complication
<90 Open, 59% (N ¼ 30,924)
<90 Endo, 40% (N ¼ 21,033)
90 Open, 0.3% (N ¼ 132)
90 Endo, 0.6% (N ¼ 291)
Death, % (n) Neurologic, % (n) Cardiac, % (n) Respiratory, % (n) Renal, % (n)
4.6 0.5 6.7 6.4 3.5
1.2 0.2 2.1 1.2 1.4
18.3 0 15.2 7.6 4.5
3.1 0.3 3.1 2.1 3.1
(1,414) (165) (2,076) (1,977) (1,086)
(243) (45) (439) (253) (294)
(24) (0) (20) (10) (6)
(9) (1) (9) (6) (9)
Table III. Adjusted multivariate analysis of postoperative complications <90 Endo
90 Open
90 Endo
Complication
<90 Open
OR
95% CI
OR
95% CI
OR
95% CI
Mortality Neurologica Cardiacb Respiratoryc Renald
Referent Referent Referent Referent Referent
0.29 0.44 0.3 0.19 0.39
0.25e0.34 0.32e0.62 0.27e0.33 0.17e0.22 0.35e0.45
4.32 0 2.44 1.26 1.27
2.73e6.85 0 1.51e3.94 0.66e2.41 0.56e2.88
0.66 0.66 0.43 0.34 0.77
0.33e1.30 0.92e4.73 0.22e0.85 0.15e0.76 0.39e1.50
a
Post operative stroke, cerebral hypoxia, and anoxic brain damage. Heart failure, cardiac arrest, and cardiac insufficiency. c Ventilatory-associated pneumonia, pneumonia, and aspiration pneumonitis. d Acute renal failure, acute renal insufficiency, acute tubular necrosis, oliguria, and anuria. b
groups, the most common comorbidity was cardiac disease, representing 71% of the open <90, 78% of the endovascular <90, 68% of the open 90, and 69% of the endovascular 90 patients. Additional preoperative comorbidities are listed in Table I. Open AAA repair in nonagenarians had the highest rate of inpatient postoperative complications. The mortality rate for this group was 18%. Mortality for <90 open group was 4.6%; the <90 endovascular 1.2%; and the 90 endovascular 3.1% (Table II). Multivariate analysis adjusting for sex, ethnicity, and comorbidities was done using the <90 open group as a reference (Table III). There was a 4-fold increased mortality identified in patients 90 years undergoing open repair compared with patients <90 years (odds ratio [OR], 4.32; 95% confidence interval [CI], 2.73e6.85). Endovascular repair in patients 90 years was associated with no increased risk of inpatient mortality as compared with open repair in patients <90 years (OR, 0.66; 95% CI, 0.33e1.3; Fig. 2). Table IV depicts the association of various demographic variables with postoperative outcomes and the likelihood of presenting comorbidities to result in mortality. Of note, blacks undergoing open repair were more likely to die as compared with whites, and dramatically so in the >90 years age group (OR, 8 for blacks). A postoperative neurologic or renal complication was noted to greatly increase the chances of mortality.
DISCUSSION The concept behind elective treatment of AAA’s is to prevent potential rupture with an attendant high mortality. This benefit must be carefully weighed against the inherent risks associated with the proposed operation. Nonagenarians and their associated comorbidities make them regularly considered unfit for repair and a decision not to treat has frequently been the sole management for this group of high-risk patients in the past. Nonagenarians, however, represent a growing demographic within the United States. From 1990 to 2010, individuals aged 90 to 94 years increased by 1.4% according to the US Census Bureau.1 The calculated life expectancy at 90 years of age is 4.8 years and 3.6 years for individuals who reach the age of 95 years.2 This reality has influenced attitudes in recent years with several studies demonstrating reasonable results after AAA repair in this age group.3e8 Our study demonstrates that in patients 90 years there are differences in inpatient morbidity and mortality compared with patients <90. Nonagenarians have an overall mortality risk of 18.3% for open repair and 3.1% for endovascular repair compared with a mortality of 4.6% for open and 1.2% for endovascular in patients <90 years. As expected, mortality was significantly lower in nonagenarians undergoing endovascular repair of AAA (EVAR) validating the attractiveness of EVAR as a
186 Hughes et al.
Fig. 2. Odds ratios of mortality compared by age and type of repair.
Fig. 3. (A) Postoperative morbidity and mortality by age groupdendovascular. (B) Postoperative morbidity and mortality by age groupdopen.
preferred modality for this age group. Tsilimparis et al.4 demonstrated similar mortality rates as found in our study when comparing open with endovascular repair in nonagenarians. The finding that open repair is associated with worse outcomes when compared with endovascular repair in the elderly has also been previously reported by others.9,10 Not surprisingly, we noted that inpatient postoperative complications increased with increasing age, although at a lower rate for EVAR than for open repair (Fig. 3). The key finding in this study is that nonagenarians undergoing EVAR had an inpatient mortality that was at least as good as that obtained for patients <90 undergoing open repair (OR, 0.66; 95% CI, 0.33e1.3; Table III). Beyond the immediate postoperative period, several studies have attempted to determine if EVAR offers a survival benefit in high-risk groups.
Annals of Vascular Surgery
The EVAR 2 trial, conducted in the United Kingdom, randomized patients to endovascular repair or no intervention and concluded that there was no survival benefit for endovascular repair in patients deemed too high risk to undergo an open repair.11,12 In a similar study, Iannelli et al.13 randomized highrisk patients to either endovascular repair or open repair and found no significant difference in late mortality. In contrast, in a study conducted by the Society for Vascular Surgery Outcomes Committee, using data from 5 multicenter trials to evaluate high-risk patients as defined by the EVAR 2 trial, the authors concluded that endovascular repair was safe and effective with procedural and clinical outcomes that were superior to that reported by the EVAR 2 trial.12 Enthusiastic as we may be, the benefits of EVAR should not compromise proper surgical judgment regarding appropriate selection of nonagenarians for AAA repair. Our report establishes clear risks of morbidity and mortality associated with EVAR, stressing the continued need for careful patient selection. Currently, there is no universally accepted standard to define ‘‘high-risk’’ patients. The assumption of ‘‘fit’’ is based on a clinical judgment of a patient’s cardiovascular, pulmonary, and renal comorbidities in relation with the operative risk. Several models such as the Glasgow aneurysm score14 and the modified Leiden score15 have been developed to predict postoperative mortality for open AAA repair and have also been used to evaluate patients being considered for EVAR. Other investigators have developed similar scoring models to predict outcomes. Unfortunately, all these models use age as a prominent predictor, so that age older than 80 years alone often places an individual at high risk. A potential downside of EVAR is the need for reintervention or secondary procedures reported as high as 27e35% over a 2-year period in some studies.16,17 Clearly, the ability of an elderly patient to tolerate a potentially necessary secondary procedure should factor in the clinical decision-making process.3 The finding that blacks are more likely to die after AAA repair has been previously reported by many.18e20 Osborne et al. demonstrated worse outcomes for blacks after AAA repair even after adjusting for type of repair. Postulated reasons have included socioeconomic factors and access to care including the fact that black patients tend to receive care in lower quality and low volume hospitals.18 Vogel et al. also reported increased risk-adjusted mortality rates in blacks compared with whites for elective AAA repair. In their study, Hispanics and uninsured subjects were also less likely to receive EVAR as compared with whites and insured
Vol. 29, No. 2, February 2015
AAA repair in nonagenarians
187
Table IV. Multivariate analysis of associated variables affecting mortality Odds ratio for mortality
Variable
<90 Open
<90 Endo
OR
OR
95% CI
Ethnicity (compared with white) Black 1.76 1.37e2.26 Hispanic 1.69 1.21e2.36 Asian 0.3 0.07e1.22 Gender Female (compared 1.43 1.26e1.63 with male) Preoperative comorbidities Cardiac 0.45 0.40e0.51 Pulmonary 1.21 0.4e0.51 Renal 1.53 1.07e2.18 Diabetes mellitus 0.95 0.77e1.16 Postoperative complications Neurologic 5.94 4.03e8.75 Cardiac 2.87 2.43e3.38 Respiratory 0.97 0.78e1.21 Renal 3.48 2.86e4.23 Elective versus emergent Emergent surgery 1.48 1.41e1.54
90 Open
90 Endo
95% CI
OR
95% CI
OR
95% CI
1.15 0.9 3.99
0.62e2.11 0.33e2.46 1.55e10.27
8.04 9.15 2.67
1.23e52.40 0.66e126 0.18e39.63
xx xx xx
xx xx xx
2.6
1.94e3.47
0.93
0.26e3.35
1.89
0.32e11.06
0.39 1.38 2.24 0.85
0.29e0.52 1.04e1.83 1.39e3.59 0.56e1.31
1.33 2.24 xx 0.67
0.35e4.99 0.59e8.47 xx 0.04e11.12
0.47 0.91 xx xx
0.1e2.25 0.09e8.93 xx xx
10.1 7.33 1.91 6.06
3.93e25.92 4.85e11.07 0.9e4.08 3.65e10.07
xx 3.6 0.87 5.8
xx 0.82e15.75 0.12e6.62 0.71e47.59
xx 11.2 4.02 xx
xx 1.14e110.87 0.02e661.81 xx
1.55
1.38e1.67
1.46
0.23e1.84
1.74
0.7e1.95
‘‘xx’’ refers to insufficient data for analysis.
patients.21 Unfortunately, our study indicates that this ethnic disparity is especially dramatic for the most elderly: blacks older than 90 years were 8 times more likely to die after an open AAA repair. The calculated CI for Hispanics suggests that the derived OR of 9 may have arrived by chance.
mortality, although significantly higher than that obtained for endovascular repair in patients <90 years, is nonetheless comparable with mortality noted in patients <90 years undergoing open AAA repair. REFERENCES
Limitations There are several limitations in our present study. The primary limitation is that this is based on administrative database. The NIS is the largest all-payer inpatient care database with approximately 8 million hospital stays each year from up to 40 states comprising a 20% stratified sample of US hospitals. This database, however, is limited to information collected during a particular hospitalization, so that quality of life data and long-term survival are not available. Other variables that could not be examined were endovascular cases that required reintervention, surgical technique, specific causes of death, and aneurysm size. In conclusion, this study representing the single largest evaluation of inpatient outcomes of elective AAA repair in patients 90 years shows that open repair of AAA’s in nonagenarians is associated with significantly high perioperative mortality, whereas endovascular repair is feasible with acceptable perioperative morbidity and mortality. This
1. He W, Muenchrath MN. 90+ in the United States: 2006e 2008. In: Bureau USC ed. Washington, DC: U.S. Government Printing Office, 2011. 2. Arias E, Smith BL. Deaths: preliminary data for 2001. Natl Vital Stat Rep 2003;51:1e44. 3. Prenner SB, Turnbull IC, Malik R, et al. Outcome of elective endovascular abdominal aortic aneurysm repair in octogenarians and nonagenarians. J Vasc Surg 2010;51:1354e9. 4. Tsilimparis N, Perez S, Dayama A, et al. Age-stratified results from 20,095 aortoiliac aneurysm repairs: should we approach octogenarians and nonagenarians differently? J Am Coll Surg 2012;215:690e701. 5. Baril DT, Palchik E, Carroccio A, et al. Experience with endovascular abdominal aortic aneurysm repair in nonagenarians. J Endovasc Ther 2006;13:330e7. 6. Bouwmeester MF, van Sterkenburg SM, Zeebregts CJ, et al. Endovascular abdominal aortic aneurysm repair in nonagenariansebeyond limits? Tex Heart Inst J 2009;36:345e8. 7. Jim J, Sanchez LA, Sicard GA, et al. Acceptable risk but small benefit of endovascular aneurysm repair in nonagenarians. Ann Vasc Surg 2010;24:441e6. 8. Park BD, Azefor NM, Huang C-C, et al. Elective endovascular aneurysm repair in the elderly: trends and outcomes from the nationwide inpatient sample. Ann Vasc Surg 2014;28:798e807.
188 Hughes et al.
9. Indes JE, Tuggle CT, Mandawat A, et al. Age-stratified outcomes in elderly patients undergoing open and endovascular procedures for aortoiliac occlusive disease. Surgery 2010;148: 420e8. 10. Raval MV, Eskandari MK. Outcomes of elective abdominal aortic aneurysm repair among the elderly: endovascular versus open repair. Surgery 2012;151:245e60. 11. Endovascular aneurysm repair and outcome in patients unfit for open repair of abdominal aortic aneurysm (EVAR trial 2): randomised controlled trial. Lancet 2005;365:2187e92. 12. Sicard GA, Zwolak RM, Sidawy AN, et al. Endovascular abdominal aortic aneurysm repair: long-term outcome measures in patients at high-risk for open surgery. J Vasc Surg 2006;44:229e36. 13. Iannelli G, Monaco M, Di Tommaso L, et al. Endovascular vs. open surgery of abdominal aortic aneurysm in highrisk patients: a single center experience. Thorac Cardiovasc Surg 2005;53:291e4. 14. Samy AK, Murray G, MacBain G. Glasgow aneurysm score. Cardiovasc Surg 1994;2:41e4. 15. Steyerberg EW, Kievit J, de Mol Van Otterloo JC, et al. Perioperative mortality of elective abdominal aortic aneurysm
Annals of Vascular Surgery
16.
17.
18.
19.
20.
21.
surgery. A clinical prediction rule based on literature and individual patient data. Arch Intern Med 1995;155:1998e2004. 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: 2398e405. Becquemin JP, Kelley L, Zubilewicz T, et al. Outcomes of secondary interventions after abdominal aortic aneurysm endovascular repair. J Vasc Surg 2004;39:298e305. Osborne NH, Upchurch GR Jr, Mathur AK, et al. Explaining racial disparities in mortality after abdominal aortic aneurysm repair. J Vasc Surg 2009;50:709e13. Trivedi AN, Sequist TD, Ayanian JZ. Impact of hospital volume on racial disparities in cardiovascular procedure mortality. J Am Coll Cardiol 2006;47:417e24. Lemaire A, Cook C, Tackett S, et al. The impact of race and insurance type on the outcome of endovascular abdominal aortic aneurysm (AAA) repair. J Vasc Surg 2008;47: 1172e80. Vogel TR, Cantor JC, Dombrovskiy VY, et al. AAA repair: sociodemographic disparities in management and outcomes. Vasc Endovascular Surg 2008;42:555e60.
Background: The feasibility of abdominal aortic aneurysm (AAA) repair in nonagenarians on a national level is largely unknown. We undertook this study to determine the outcomes of open and endovascular AAA repair in this population on a national level. Methods: A retrospective review of the Nationwide Inpatient Sample Database was conducted to determine all patients 90 years and older who underwent either an open or endovascular repair of a nonruptured AAA from 1997 to 2008. Preoperative comorbidities and postoperative complications in the inpatient setting were recorded. The primary end point was mortality. Secondary end points were postoperative neurologic, cardiac, and respiratory complications. This group was then compared with all adult patients less than 90 years old (age, 18e89) who had undergone repair of a nonruptured AAA during this same period. Results: Four hundred twenty-three patients 90 years and older underwent repair of a nonruptured AAA (compared with 52,370 < 90). Of these, 132 patients underwent open repair (31%) and 291 (69%) underwent endovascular repair. Inpatient mortality was 18.3% for the 90 open, 4.6% for the <90 open, 3.1% for the 90 endovascular, and 1.2% for <90 endovascular group. Conclusions: Open repair of AAA’s in nonagenarians is associated with significantly high perioperative mortality, whereas endovascular repair is feasible with acceptable perioperative mortality. This mortality, although significantly higher than that obtained for endovascular repair in patients <90, is nonetheless not significantly different for the mortality noted for patients <90 undergoing open AAA repair.
INTRODUCTION
1 Department of Surgery, Howard University College of Medicine and Hospital, Washington, DC. 2 Department of Internal Medicine, Howard University College of Medicine and Hospital, Washington, DC. 3 Department of Internal Medicine, University of Texas Medical Branch, Galveston, TX. 4 Department of Surgery, Upstate Medical University, Syracuse, NY. Correspondence to: Kakra Hughes, MD, FACS, Department of Surgery, Howard University College of Medicine and Hospital, 2041 Georgia Avenue, NW, # 4B.04, Washington, DC 20060, USA; E-mail: [email protected]
Ann Vasc Surg 2015; 29: 183–188 http://dx.doi.org/10.1016/j.avsg.2014.07.037 Ó 2015 Elsevier Inc. All rights reserved. Manuscript received: March 17, 2014; manuscript accepted: July 30, 2014; published online: October 15, 2014.
Abdominal aortic aneurysms (AAAs) are largely a disease of the elderly. As the nation’s population continues to agedthe US Census Bureau data from 2008 predict that by the year 2050, 20 million individuals, representing approximately 5% of the population, will be older than 85 years1,2dit will be increasingly necessary to treat AAA’s in an elderly population. Kazmers et al. have shown that open aneurysm repair may be beneficial in octogenarians despite the increased mortality associated with their age.2,3 Furthermore, more recent studies have reported the feasibility of endovascular AAA repair in the elderly.4e8 Most of these reports, however, have been single institution studies, with results not necessarily generalizable to the entire 183
184 Hughes et al.
Annals of Vascular Surgery
Table I. Bivariate analysis of patient demographics, comorbidities, and timing of surgery Variable
Gender, % (n) Male Female Ethnicity, % (n) White Black Hispanic Asian Comorbidities, % (n) Cardiac disease COPD Kidney disease Diabetes Elective, % (n) Emergent, % (n)
<90 Open, 59% (N ¼ 30,924)
<90 Endo, 40% (N ¼ 21,033)
90 Open, 0.3% (N ¼ 132)
90 Endo, 0.6% (N ¼ 291)
77.2 (23,874) 22.8 (7,050)
83.2 (17,502) 16.8 (3,531)
67.4 (89) 32.6 (43)
74.9 (218) 25.1 (73)
92.8 3.4 2.8 1.1
(28,687) (1,050) (849) (338)
92.6 3.5 2.7 1.3
(19,475) (730) (560) (268)
87.9 4.6 2.3 5.3
(116) (6) (3) (7)
92.1 3.4 2.4 2.1
(268) (10) (7) (6)
71.8 35.4 1.6 10.4 78.4 21.6
(22,197) (10,945) (503) (3,209) (21,595) (5,958)
78.3 31.4 3.6 15.2 85.7 14.3
(16,467) (6,611) (758) (3,202) (16,035) (2,677)
68.2 21.2 1.5 6.1 69.5 30.5
(90) (28) (2) (8) (82) (36)
69.8 20.3 8.3 6.5 77.8 22.2
(203) (59) (24) (19) (196) (56)
COPD, chronic obstructive pulmonary disease.
Fig. 1. Age distribution of the study population.
nation. We undertook the present study to determine, on a national level, the outcomes of AAA repair in patients 90 years and older.
METHODS The Nationwide Inpatient Sample (NIS) Database was queried to identify all patients who underwent an open or endovascular nonruptured AAA repair between 1998 and 2007 (International Classification of Disease 9 Clinical Modification, ICD 9 CM codes: 38.44 and 39.71, respectively). Patients younger than 18 years were excluded. Patient demographics, comorbidities, and postoperative complications during the inpatient period were compared between patients 90 years and older and patients younger than 90 years. Preoperative comorbidities were defined as cardiac (a history of previous coronary artery bypass grafting or percutaneous coronary intervention, coronary artery disease, and/or congestive heart failure); respiratory (a diagnosis of chronic obstructive pulmonary
disease); renal (chronic kidney disease); and diabetes mellitus. Postoperative complications assessed included: mortality, neurologic (stroke), cardiac, respiratory, and renal complications. Bivariate analysis was conducted using the Pearson chi-squared test for categorical variables and Student’s t-test for continuous variables with a level of significance set at less than 0.05%. Multivariate regression was performed, controlling for patient age, sex, ethnicity, insurance status, and preoperative comorbidities.
RESULTS There were 52,380 patients who were identified in the NIS as having undergone an AAA repair from 1998 to 2007. The population was primarily men (79%, n ¼ 41,683) and white (93%). Open AAA repair was performed in 59% of patients (Table I). Patients aged 90 years or older represented 0.8% of the population (n ¼ 423). Most patients in the study group were in their 70s (46% of the population), followed by patients in their 60s (27% of the population; Fig. 1). Four groups were identified patients <90 years old undergoing open repair represented 59% of the population (n ¼ 30,924); patients < 90 years undergoing endovascular repair represented 40% of the population (n ¼ 21,033); patients 90 years undergoing open repair represented 0.3% of the population (n ¼ 132); and patients 90 years undergoing endovascular repair comprised 0.6% of the population (n ¼ 291; Table I). In all the 4
Vol. 29, No. 2, February 2015
AAA repair in nonagenarians
185
Table II. Bivariate analysis of postoperative complications Complication
<90 Open, 59% (N ¼ 30,924)
<90 Endo, 40% (N ¼ 21,033)
90 Open, 0.3% (N ¼ 132)
90 Endo, 0.6% (N ¼ 291)
Death, % (n) Neurologic, % (n) Cardiac, % (n) Respiratory, % (n) Renal, % (n)
4.6 0.5 6.7 6.4 3.5
1.2 0.2 2.1 1.2 1.4
18.3 0 15.2 7.6 4.5
3.1 0.3 3.1 2.1 3.1
(1,414) (165) (2,076) (1,977) (1,086)
(243) (45) (439) (253) (294)
(24) (0) (20) (10) (6)
(9) (1) (9) (6) (9)
Table III. Adjusted multivariate analysis of postoperative complications <90 Endo
90 Open
90 Endo
Complication
<90 Open
OR
95% CI
OR
95% CI
OR
95% CI
Mortality Neurologica Cardiacb Respiratoryc Renald
Referent Referent Referent Referent Referent
0.29 0.44 0.3 0.19 0.39
0.25e0.34 0.32e0.62 0.27e0.33 0.17e0.22 0.35e0.45
4.32 0 2.44 1.26 1.27
2.73e6.85 0 1.51e3.94 0.66e2.41 0.56e2.88
0.66 0.66 0.43 0.34 0.77
0.33e1.30 0.92e4.73 0.22e0.85 0.15e0.76 0.39e1.50
a
Post operative stroke, cerebral hypoxia, and anoxic brain damage. Heart failure, cardiac arrest, and cardiac insufficiency. c Ventilatory-associated pneumonia, pneumonia, and aspiration pneumonitis. d Acute renal failure, acute renal insufficiency, acute tubular necrosis, oliguria, and anuria. b
groups, the most common comorbidity was cardiac disease, representing 71% of the open <90, 78% of the endovascular <90, 68% of the open 90, and 69% of the endovascular 90 patients. Additional preoperative comorbidities are listed in Table I. Open AAA repair in nonagenarians had the highest rate of inpatient postoperative complications. The mortality rate for this group was 18%. Mortality for <90 open group was 4.6%; the <90 endovascular 1.2%; and the 90 endovascular 3.1% (Table II). Multivariate analysis adjusting for sex, ethnicity, and comorbidities was done using the <90 open group as a reference (Table III). There was a 4-fold increased mortality identified in patients 90 years undergoing open repair compared with patients <90 years (odds ratio [OR], 4.32; 95% confidence interval [CI], 2.73e6.85). Endovascular repair in patients 90 years was associated with no increased risk of inpatient mortality as compared with open repair in patients <90 years (OR, 0.66; 95% CI, 0.33e1.3; Fig. 2). Table IV depicts the association of various demographic variables with postoperative outcomes and the likelihood of presenting comorbidities to result in mortality. Of note, blacks undergoing open repair were more likely to die as compared with whites, and dramatically so in the >90 years age group (OR, 8 for blacks). A postoperative neurologic or renal complication was noted to greatly increase the chances of mortality.
DISCUSSION The concept behind elective treatment of AAA’s is to prevent potential rupture with an attendant high mortality. This benefit must be carefully weighed against the inherent risks associated with the proposed operation. Nonagenarians and their associated comorbidities make them regularly considered unfit for repair and a decision not to treat has frequently been the sole management for this group of high-risk patients in the past. Nonagenarians, however, represent a growing demographic within the United States. From 1990 to 2010, individuals aged 90 to 94 years increased by 1.4% according to the US Census Bureau.1 The calculated life expectancy at 90 years of age is 4.8 years and 3.6 years for individuals who reach the age of 95 years.2 This reality has influenced attitudes in recent years with several studies demonstrating reasonable results after AAA repair in this age group.3e8 Our study demonstrates that in patients 90 years there are differences in inpatient morbidity and mortality compared with patients <90. Nonagenarians have an overall mortality risk of 18.3% for open repair and 3.1% for endovascular repair compared with a mortality of 4.6% for open and 1.2% for endovascular in patients <90 years. As expected, mortality was significantly lower in nonagenarians undergoing endovascular repair of AAA (EVAR) validating the attractiveness of EVAR as a
186 Hughes et al.
Fig. 2. Odds ratios of mortality compared by age and type of repair.
Fig. 3. (A) Postoperative morbidity and mortality by age groupdendovascular. (B) Postoperative morbidity and mortality by age groupdopen.
preferred modality for this age group. Tsilimparis et al.4 demonstrated similar mortality rates as found in our study when comparing open with endovascular repair in nonagenarians. The finding that open repair is associated with worse outcomes when compared with endovascular repair in the elderly has also been previously reported by others.9,10 Not surprisingly, we noted that inpatient postoperative complications increased with increasing age, although at a lower rate for EVAR than for open repair (Fig. 3). The key finding in this study is that nonagenarians undergoing EVAR had an inpatient mortality that was at least as good as that obtained for patients <90 undergoing open repair (OR, 0.66; 95% CI, 0.33e1.3; Table III). Beyond the immediate postoperative period, several studies have attempted to determine if EVAR offers a survival benefit in high-risk groups.
Annals of Vascular Surgery
The EVAR 2 trial, conducted in the United Kingdom, randomized patients to endovascular repair or no intervention and concluded that there was no survival benefit for endovascular repair in patients deemed too high risk to undergo an open repair.11,12 In a similar study, Iannelli et al.13 randomized highrisk patients to either endovascular repair or open repair and found no significant difference in late mortality. In contrast, in a study conducted by the Society for Vascular Surgery Outcomes Committee, using data from 5 multicenter trials to evaluate high-risk patients as defined by the EVAR 2 trial, the authors concluded that endovascular repair was safe and effective with procedural and clinical outcomes that were superior to that reported by the EVAR 2 trial.12 Enthusiastic as we may be, the benefits of EVAR should not compromise proper surgical judgment regarding appropriate selection of nonagenarians for AAA repair. Our report establishes clear risks of morbidity and mortality associated with EVAR, stressing the continued need for careful patient selection. Currently, there is no universally accepted standard to define ‘‘high-risk’’ patients. The assumption of ‘‘fit’’ is based on a clinical judgment of a patient’s cardiovascular, pulmonary, and renal comorbidities in relation with the operative risk. Several models such as the Glasgow aneurysm score14 and the modified Leiden score15 have been developed to predict postoperative mortality for open AAA repair and have also been used to evaluate patients being considered for EVAR. Other investigators have developed similar scoring models to predict outcomes. Unfortunately, all these models use age as a prominent predictor, so that age older than 80 years alone often places an individual at high risk. A potential downside of EVAR is the need for reintervention or secondary procedures reported as high as 27e35% over a 2-year period in some studies.16,17 Clearly, the ability of an elderly patient to tolerate a potentially necessary secondary procedure should factor in the clinical decision-making process.3 The finding that blacks are more likely to die after AAA repair has been previously reported by many.18e20 Osborne et al. demonstrated worse outcomes for blacks after AAA repair even after adjusting for type of repair. Postulated reasons have included socioeconomic factors and access to care including the fact that black patients tend to receive care in lower quality and low volume hospitals.18 Vogel et al. also reported increased risk-adjusted mortality rates in blacks compared with whites for elective AAA repair. In their study, Hispanics and uninsured subjects were also less likely to receive EVAR as compared with whites and insured
Vol. 29, No. 2, February 2015
AAA repair in nonagenarians
187
Table IV. Multivariate analysis of associated variables affecting mortality Odds ratio for mortality
Variable
<90 Open
<90 Endo
OR
OR
95% CI
Ethnicity (compared with white) Black 1.76 1.37e2.26 Hispanic 1.69 1.21e2.36 Asian 0.3 0.07e1.22 Gender Female (compared 1.43 1.26e1.63 with male) Preoperative comorbidities Cardiac 0.45 0.40e0.51 Pulmonary 1.21 0.4e0.51 Renal 1.53 1.07e2.18 Diabetes mellitus 0.95 0.77e1.16 Postoperative complications Neurologic 5.94 4.03e8.75 Cardiac 2.87 2.43e3.38 Respiratory 0.97 0.78e1.21 Renal 3.48 2.86e4.23 Elective versus emergent Emergent surgery 1.48 1.41e1.54
90 Open
90 Endo
95% CI
OR
95% CI
OR
95% CI
1.15 0.9 3.99
0.62e2.11 0.33e2.46 1.55e10.27
8.04 9.15 2.67
1.23e52.40 0.66e126 0.18e39.63
xx xx xx
xx xx xx
2.6
1.94e3.47
0.93
0.26e3.35
1.89
0.32e11.06
0.39 1.38 2.24 0.85
0.29e0.52 1.04e1.83 1.39e3.59 0.56e1.31
1.33 2.24 xx 0.67
0.35e4.99 0.59e8.47 xx 0.04e11.12
0.47 0.91 xx xx
0.1e2.25 0.09e8.93 xx xx
10.1 7.33 1.91 6.06
3.93e25.92 4.85e11.07 0.9e4.08 3.65e10.07
xx 3.6 0.87 5.8
xx 0.82e15.75 0.12e6.62 0.71e47.59
xx 11.2 4.02 xx
xx 1.14e110.87 0.02e661.81 xx
1.55
1.38e1.67
1.46
0.23e1.84
1.74
0.7e1.95
‘‘xx’’ refers to insufficient data for analysis.
patients.21 Unfortunately, our study indicates that this ethnic disparity is especially dramatic for the most elderly: blacks older than 90 years were 8 times more likely to die after an open AAA repair. The calculated CI for Hispanics suggests that the derived OR of 9 may have arrived by chance.
mortality, although significantly higher than that obtained for endovascular repair in patients <90 years, is nonetheless comparable with mortality noted in patients <90 years undergoing open AAA repair. REFERENCES
Limitations There are several limitations in our present study. The primary limitation is that this is based on administrative database. The NIS is the largest all-payer inpatient care database with approximately 8 million hospital stays each year from up to 40 states comprising a 20% stratified sample of US hospitals. This database, however, is limited to information collected during a particular hospitalization, so that quality of life data and long-term survival are not available. Other variables that could not be examined were endovascular cases that required reintervention, surgical technique, specific causes of death, and aneurysm size. In conclusion, this study representing the single largest evaluation of inpatient outcomes of elective AAA repair in patients 90 years shows that open repair of AAA’s in nonagenarians is associated with significantly high perioperative mortality, whereas endovascular repair is feasible with acceptable perioperative morbidity and mortality. This
1. He W, Muenchrath MN. 90+ in the United States: 2006e 2008. In: Bureau USC ed. Washington, DC: U.S. Government Printing Office, 2011. 2. Arias E, Smith BL. Deaths: preliminary data for 2001. Natl Vital Stat Rep 2003;51:1e44. 3. Prenner SB, Turnbull IC, Malik R, et al. Outcome of elective endovascular abdominal aortic aneurysm repair in octogenarians and nonagenarians. J Vasc Surg 2010;51:1354e9. 4. Tsilimparis N, Perez S, Dayama A, et al. Age-stratified results from 20,095 aortoiliac aneurysm repairs: should we approach octogenarians and nonagenarians differently? J Am Coll Surg 2012;215:690e701. 5. Baril DT, Palchik E, Carroccio A, et al. Experience with endovascular abdominal aortic aneurysm repair in nonagenarians. J Endovasc Ther 2006;13:330e7. 6. Bouwmeester MF, van Sterkenburg SM, Zeebregts CJ, et al. Endovascular abdominal aortic aneurysm repair in nonagenariansebeyond limits? Tex Heart Inst J 2009;36:345e8. 7. Jim J, Sanchez LA, Sicard GA, et al. Acceptable risk but small benefit of endovascular aneurysm repair in nonagenarians. Ann Vasc Surg 2010;24:441e6. 8. Park BD, Azefor NM, Huang C-C, et al. Elective endovascular aneurysm repair in the elderly: trends and outcomes from the nationwide inpatient sample. Ann Vasc Surg 2014;28:798e807.
188 Hughes et al.
9. Indes JE, Tuggle CT, Mandawat A, et al. Age-stratified outcomes in elderly patients undergoing open and endovascular procedures for aortoiliac occlusive disease. Surgery 2010;148: 420e8. 10. Raval MV, Eskandari MK. Outcomes of elective abdominal aortic aneurysm repair among the elderly: endovascular versus open repair. Surgery 2012;151:245e60. 11. Endovascular aneurysm repair and outcome in patients unfit for open repair of abdominal aortic aneurysm (EVAR trial 2): randomised controlled trial. Lancet 2005;365:2187e92. 12. Sicard GA, Zwolak RM, Sidawy AN, et al. Endovascular abdominal aortic aneurysm repair: long-term outcome measures in patients at high-risk for open surgery. J Vasc Surg 2006;44:229e36. 13. Iannelli G, Monaco M, Di Tommaso L, et al. Endovascular vs. open surgery of abdominal aortic aneurysm in highrisk patients: a single center experience. Thorac Cardiovasc Surg 2005;53:291e4. 14. Samy AK, Murray G, MacBain G. Glasgow aneurysm score. Cardiovasc Surg 1994;2:41e4. 15. Steyerberg EW, Kievit J, de Mol Van Otterloo JC, et al. Perioperative mortality of elective abdominal aortic aneurysm
Annals of Vascular Surgery
16.
17.
18.
19.
20.
21.
surgery. A clinical prediction rule based on literature and individual patient data. Arch Intern Med 1995;155:1998e2004. 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: 2398e405. Becquemin JP, Kelley L, Zubilewicz T, et al. Outcomes of secondary interventions after abdominal aortic aneurysm endovascular repair. J Vasc Surg 2004;39:298e305. Osborne NH, Upchurch GR Jr, Mathur AK, et al. Explaining racial disparities in mortality after abdominal aortic aneurysm repair. J Vasc Surg 2009;50:709e13. Trivedi AN, Sequist TD, Ayanian JZ. Impact of hospital volume on racial disparities in cardiovascular procedure mortality. J Am Coll Cardiol 2006;47:417e24. Lemaire A, Cook C, Tackett S, et al. The impact of race and insurance type on the outcome of endovascular abdominal aortic aneurysm (AAA) repair. J Vasc Surg 2008;47: 1172e80. Vogel TR, Cantor JC, Dombrovskiy VY, et al. AAA repair: sociodemographic disparities in management and outcomes. Vasc Endovascular Surg 2008;42:555e60.