- Email: [email protected]
Indication for Surgery, the Revised Cardiac Risk Index, and 1-Year Mortality
Indication for Surgery, the Revised Cardiac Risk Index, and 1-Year Mortality Vivek K. Moitra,1 Brigid C. Flynn,2 Michael Mazzeffi,2 Carol Bodian,3 David Bronheim,2 and John E. Ellis,4 New York, New York, and Philadelphia, Pennsylvania
Background: Patients who undergo vascular surgery are at increased risk of perioperative cardiovascular morbidity and mortality. The Revised Cardiac Risk Index (RCRI) is a validated and widely used bedside tool for estimating the risk of a perioperative major adverse myocardial event. We hypothesized that inclusion of the indication for surgery would add independent and prognostic information to the RCRI in predicting all-cause 30-day and 1-year mortality in open infrainguinal vascular surgical procedures. Methods: This was a retrospective study of 603 patients who underwent open infrainguinal bypass vascular surgery between January 2002 and January 2008 at a tertiary care medical center. RCRI and indication for surgery were determined. The primary outcomes of interest were all-cause 30-day mortality (which included all in-hospital mortality, regardless of time) and all-cause 1-year mortality. Results: Overall 30-day mortality was 32 (5.3%). Independent risk factors for early death were RCRI score, being of age 80 years, American Society of Anesthesiologists Physical Status classification ¼ 4, and emergency surgery. Overall 1-year mortality, including early deaths, was 114 (18.9%). Indication for surgery, RCRI score, age, American Society of Anesthesiologists Physical Status classification ¼ 4, female sex, and emergency surgery were all independent predictors of 1-year mortality. Conclusions: The RCRI score was associated with both 30-day and 1-year mortality in patients undergoing lower extremity bypass surgery. Indication for surgery was predictive of 1-year mortality but not of 30-day mortality.
INTRODUCTION Patients who undergo vascular surgery are at increased risk for perioperative cardiovascular morbidity and mortality. The American College of Cardiology/American Heart Association (ACC/ AHA) perioperative guidelines for noncardiac surgery classify peripheral vascular procedures as being high risk.1 These guidelines also advise careful evaluation of long-term outcomes in such cases. Given the probable comorbidities of this patient
1 Division of Critical Care Medicine, Department of Anesthesiology, College of Physicians and Surgeons of Columbia University, New York, NY. 2 Department of Anesthesiology and Critical Care, Mount Sinai School of Medicine, New York City, NY. 3
Division of Biostatistics, Department of Anesthesiology, Mount Sinai School of Medicine, New York City, NY. 4 Department of Anesthesiology and Critical Care, University of Pennsylvania School of Medicine, Philadelphia, PA.
902
population, understanding perioperative and long-term risk facilitates informed consent and the preoperative testing process. The Revised Cardiac Risk Index (RCRI) is a validated and widely used bedside tool for estimating a patient’s risk of a major adverse myocardial event within 30 days of surgery.2 However, a recent meta-analysis of the RCRI found that this index does not perform well in predicting death in the perioperative period after vascular surgery as compared with other noncardiac procedures.3 Although
Correspondence to: Vivek K. Moitra, MD, Department of Anesthesiology, College of Physicians and Surgeons of Columbia University, 630 West 168th Street, New York, NY 10032, USA, E-mail: vm2161@ columbia.edu Ann Vasc Surg 2011; 25: 902-908 DOI: 10.1016/j.avsg.2011.05.010 Ó Annals of Vascular Surgery Inc. Published online: August 8, 2011
Vol. 25, No. 7, October 2011
indication for lower extremity bypass surgery is associated with poor outcomes,4 the current ACC/AHA guidelines do not consider indication for vascular surgery to be important in stratifying risk. Indications for this surgery include chronic limb ischemia (claudication, ischemic rest pain, or tissue loss) and acute indications such as aneurysm repair and graft thrombosis. We hypothesized that inclusion of the indication for surgery would add independent and prognostic information to the RCRI score in predicting all-cause 30-day and 1-year mortality in patients undergoing open infrainguinal bypass surgery.
MATERIALS AND METHODS This was a retrospective study of vascular surgery patients performed at the Mount Sinai School of Medicine Hospital, New York, NY, an urban tertiary care teaching hospital. The Mount Sinai Hospital Institutional Review Board approved the study, and a waiver of informed consent was obtained. Between January 2002 and January 2008, a total of 603 adults who underwent open femoraledistal arterial bypass vascular surgery procedures and satisfied our inclusion criteria were included in the study. All patients aged >18 years who underwent emergent and nonemergent operations were included. Those undergoing suprainguinal vascular surgery were excluded. In cases involving multiple procedures performed during the study period, only the first one was included in the analysis. Six patients for whom indication for surgery could not be determined were excluded. Data Collection We obtained the following information from the electronic medical records: age, sex, body mass index, American Society of Anesthesiologists Physical Status (ASA-PS) classification, medications, and type of anesthesia. For each subject, we reviewed the chart for documented and self-reported history, and we calculated an RCRI score. One point was assigned for each of the following variables: (a) baseline serum creatinine of 2 mg/dL (to convert to micromoles per liter, multiply by 88.4), (b) history of congestive heart failure, (c) history of ischemic heart disease, (d) history of cerebral vascular disease, and (e) history of diabetes mellitus requiring insulin. In addition, the RCRI score routinely assigns one point for high-risk surgery. However, infrainguinal vascular procedures are not considered high risk for this purpose (in contrast to their classification as high risk by the ACC/AHA).1 Ischemic heart disease is
Indication for surgery, RCRI, 1-year mortality 903
defined as any of the following: (a) previous Q wave on electrocardiogram, (b) previous nitrate use, (c) previous coronary artery bypass or percutaneous intervention, (d) previous positive noninvasive test indicating ischemia, or (e) previous myocardial infarction. Cerebral vascular disease is defined as either previous cerebral vascular accident or transient ischemic attack. We recorded the indication for surgery by using the Rutherford grade classification, which categorizes symptoms of chronic limb ischemia. This classification includes claudication, rest pain, and tissue loss, that is, gangrene.5 Patients who presented with acute limb ischemia were categorized as acute. Outcomes and Measurement The primary outcomes of interest were all-cause 30day mortality (which included all in-hospital mortality, regardless of time) and all-cause 1-year mortality. We obtained this information from hospital discharge summaries and the Social Security Death Index, a public database that lists dates of death for anyone registered with the Social Security Administration. Statistical Methods Data were entered in an Excel file and transferred to SAS version 9.1 (SAS/STAT 9.1 User’s Guide, SAS Institute Inc., Cary, NC) for statistical analysis. Contingency tables were constructed to describe associations between factors. The primary objective of this analysis was to evaluate the RCRI score and indication for surgery as predictors of all-cause 30day and 1-year mortality in patients undergoing lower extremity bypass surgery. Whether indication for surgery provided independent information in these cases was of particular interest. Factors were first reviewed individually for a significant or nearsignificant ( p < 0.10) association with an outcome. Independent factors were identified in separate multiple logistic regression analyses. The joint effect of RCRI and indication for surgery (Rutherford classification) was tested first. Then, when appropriate, RCRI and/or indication for surgery were forced into a stepwise multiple logistic regression analysis to identify other factors that added independent information concerning outcome. Models that included the other independent factors were considered with and without the addition of RCRI and indication. Age, body mass index, and sex were included because of their association with poor outcome after vascular surgery.6-10 A diagnosis of diabetes mellitus was considered because the RCRI score includes only those with diabetes who are on insulin,
904 Moitra et al.
and diabetes has been proposed as an explanation for poor long-term outcome after infrainguinal bypass surgery.11 Dyslipidemia was evaluated because of its association with atherosclerosis and coronary artery disease. Anesthesia type was studied because neuraxial analgesia has been suggested to decrease catecholamine levels and hypercoagulability and to improve outcomes after vascular surgery.1,12 The use of beta-blockers, clonidine, aspirin, and 3-hydroxy-3-methylglutaryl-coenzyme A (HMGCoA) reductase inhibitors was analyzed because of their association with improvements in perioperative outcome.13-20 ASA-PS classification and emergency surgery were considered because of their association with postoperative outcome. A KaplaneMeier life table was used to describe the cumulative incidence of death over the course of the first year.
RESULTS Patient demographics and clinical characteristics are shown in Table I. Patients with claudication were younger (median age, 66.0 years) than those who presented with other indications (median age, 74.0 years). The median RCRI score for patients who had claudication, rest pain, and acute surgery was 1, whereas those who presented with gangrene had a median RCRI score of 2. At the time of surgery, 61% of patients were taking a beta-blocker and 35% were taking an HMG-CoA reductase inhibitor. Among those with tissue loss, 61% were found to be taking a beta-blocker, but only 32% were taking an HMG-CoA reductase inhibitor (data not shown). Thirty-Day Mortality In our patient population, the overall 30-day mortality was 32 (5.3%). There were no early deaths reported in patients with an RCRI score of 0, regardless of surgical indication. In univariate analysis, early death was associated with RCRI score ( p < 0.001), age ( p ¼ 0.02), ASA-PS class ( p ¼ 0.0004), and emergency procedure ( p ¼ 0.004). Indication for surgery was not associated with early death. When the RCRI score and indication for surgery were considered jointly, the RCRI score, but not indication for surgery, was associated with death at 30 days. Therefore, only the RCRI score was forced into the stepwise model to identify other significant factors. Independent risk factors for early death were as follows: (a) RCRI score, (b) being of age 80 years, (c) ASA-PS classification ¼ 4, and (d) emergency surgery (Table II).
Annals of Vascular Surgery
Table I. Patient demographics and clinical characteristics (n ¼ 603)a Characteristics
Frequency
%
32 73 124 189 182
5.33 12.17 20.67 31.50 30.33
313 290
51.91 48.09
35 83 426 59
5.80 13.76 70.65 9.78
98 184 179 98 38 6 0
16.25 30.51 29.68 16.25 6.30 1.00 0
2 13 399 189 0
0.33 2.16 66.17 31.34 0
366 11 205 210 285 384 22
60.70 1.82 34.00 34.82 47.26 63.68 3.65
506 94
84.33 15.67
a
Age group , years <50 50-59 60-69 70-79 80 Gender Male Female Indication for surgery Claudication Rest pain Tissue loss Acute RCRI score 0 1 2 3 4 5 6 Classification ASA-PS 1 ASA-PS 2 ASA-PS 3 ASA-PS 4 ASA-PS 5 Medications Beta-blocker Clonidine Aspirin HMG-CoA reductase inhibitor ACE-I/ARB Diabetes mellitus Emergency Anesthesia typea General Regional
ACE-I/ARB, angiotensin-converting enzyme inhibitor/angiotensin receptor blocker; ASA-PS, American Society of Anesthesiologists Physical Status; HMG-CoA, 3-hydroxy-3-methylglutarylcoenzyme A; RCRI, Revised Cardiac Risk Index. a Missing values: age (3); anesthesia type (3).
One-Year Mortality Overall 1-year mortality, including early deaths, was 114 (18.9%). In univariate analysis, 1-year mortality was significantly associated with (a) indication for surgery ( p ¼ 0.0007), (b) RCRI score ( p < 0.0001), (c) age ( p ¼ 0.0004), (d) ASA-PS classification ( p < 0.0001), (e) sex ( p ¼ 0.0008), and (f) emergency surgery ( p ¼ 0.05).
Vol. 25, No. 7, October 2011
Indication for surgery, RCRI, 1-year mortality 905
Table II. Multivariate analysis for in-hospital mortality or death within 30 daysa Patient characteristic
OR
95% CI
p value
RCRI score Age 80 years ASA-PS ¼ 4 Emergency
1.7 2.2 2.2 7.3
1.2-2.4 1.1-4.7 1.0-5.0 2.3-24.3
0.002 0.04 0.05 0.0008
CI, confidence interval; OR, odds ratio. a There were 32 deaths among 568 patients with rest pain, tissue loss, and acute indications for surgery. There were no in-hospital deaths among 35 patients who presented with claudication.
This analysis revealed trends toward increasing risk of death with increasing RCRI score at each Rutherford level, and also with increasing Rutherford score at each RCRI level. The 1-year mortality rates cross-classified by RCRI score and indication for surgery are shown in Table III. The 1-year mortality stratified by indication for surgery was as follows: (a) claudication, 3%; (b) rest pain, 7%; (c) tissue loss, 22%; and (d) acute indication, 20%. Cumulative 1-year survival stratified by indication of surgery is shown in Figure 1. In multivariate analysis, indication for surgery, RCRI score, age, ASA-PS classification ¼ 4, female sex, and emergency surgery were independent predictors for 1-year mortality (Table IV). Among those patients who presented with claudication, none experienced early death and one died within 1 year. Because the risk of death was so low for these patients, a separate analysis that excluded them (N ¼ 35) was also conducted. However, results of this analysis were similar to the analysis of the group as a whole.
DISCUSSION Previous studies have shown that the RCRI score is an independent indicator of longer term outcome in patients undergoing a variety of procedures, including open and endovascular repair.19-21 These analyses, however, included a variety of vascular procedures with different mortality rates and did not consider indication for surgery. In our analysis, the RCRI score was significantly associated with risk of death at 1 year after open infrainguinal bypass surgery. However, indication for surgery, as well as factors such as age, ASA-PS status, sex, and emergency surgery, provided additional significant and prognostic information. Although the overall long-term mortality rate (18.9%) in this study was similar to the one that was previously reported (16.3%), mortality rates
varied considerably by surgical indication.22 In our study, the overall 1-year mortality was only 3% in patients who had claudication. Those who presented with limb tissue loss, in contrast, had an overall 1-year mortality of 22%. Open infrainguinal bypass surgery is considered high risk by the ACC (in contrast to the RCRI classification), and careful consideration of the indication for surgery is a prudent first step for risk assessment. The 1-year mortality rate of patients with rest pain or claudication was lower than that of patients with lower extremity tissue loss among those with an RCRI of >2 (15 vs. 35%, p ¼ 0.22), as well as among patients with an RCRI of 2 (5 vs. 17%, p ¼ 0.002). These data can be used in the informed consent process when patients are considering further surgery. Because vascular surgery is considered a high-risk procedure, preoperative stress testing is often performed to guide further management before surgery. Given the pain associated with claudication, this patient population may not be able to provide a history of achieving four metabolic equivalents (the equivalent of walking up two flights of stairs). Our data suggest that patients who present with claudication have a low risk of perioperative death or long-term mortality. Further studies should evaluate whether stress testing is prudent or cost-effective in this subgroup of vascular surgery patients, provided medical management has been optimal. Guidelines from the ACC/AHA suggest that clinicians should focus on the long-term management of patients undergoing vascular surgery. Consideration of indication for surgery may prompt the clinician to identify high-risk vascular surgery patients and ensure adequate medical follow-up outside the immediate perioperative period.1 These guidelines discuss magnitude of surgery and surgery type as important variables for immediate perioperative outcome. Our work demonstrates that indication for surgery should also be considered in the analysis of long-term outcome. Differences in 30-day and 1-year predictors of death in patients undergoing noncardiac surgery have been reported.21 Our data show that the RCRI is associated with both 30-day mortality and 1-year mortality in patients undergoing lower extremity bypass surgery. In contrast, indication for surgery is associated with 1-year mortality, but not with 30-day mortality. Our data are consistent with previously reported studies showing that patients who present for urgent and emergent surgery are at increased risk for perioperative morbidity at 30 days. At 1 year, emergent surgery continues to be a risk factor for poor outcome.
906 Moitra et al.
Annals of Vascular Surgery
Table III. One-year mortality rates cross-tabulated by RCRI score and indication for surgery Indication for surgery, number (%) RCRI
Claudication
Rest pain
Tissue
0 1 2 3 4 5 Total
0/14 1/14 0/6 0/1 0/0 0/0 1/35
0/14 2/32 2/25 2/10 0/2 0/0 6/83
3/50 21/118 29/138 25/80 13/34 4/6 95/426
(0) (7) (0) (0) (0) (0) (3)
(0) (6) (8) (20) (0) (0) (7)
Acute
(6) (18) (21) (31) (38) (67) (22)
1/20 5/20 3/10 2/7 1/2 0/0 12/59
Total
(5) (25) (30) (29) (50) (0) (20)
4/98 29/184 34/179 29/98 14/38 4/6 114/603
(4) (16) (19) (30) (37) (67) (18.9)
(%) ¼ mortality rate per group.
Fig. 1. Survival from service date to earlier of death or 1 year.
Indication for surgery may be a more accurate reflection of a patient’s overall health status compared with the RCRI. The reliability of the RCRI depends on the diagnosis of comorbid conditions. In patients with limited access to health care, an accurate score may not be possible. Furthermore, patients who present with gangrene may be less compliant with medication or may have limited access to preventive healthcare resources, compared with those who present with claudication. About 65% of patients with tissue loss have been found to be taking a beta-blocker, but only 32% were taking an HMG-CoA reductase inhibitor. For this reason, indication for lower extremity surgery, that is, gangrene, may reflect a patient’s undiagnosed comorbidities, serving as a useful objective
adjunct to quantify long-term risk. An atherosclerotic burden may manifest itself as an indication for vascular surgery. The influence of sex on outcomes in these patients has been reported.9,10,22 Our results show that female sex is not associated with in-hospital mortality, but is definitely associated with death after 1 year. This finding may be explained by anatomical differences and difficulty with vascular intervention, extent of disease, and access to health care for preventing complications.8,23,24 Chronic cardiac medication did not show an independent association with outcome in this study. Our investigation has several limitations. First, it was a retrospective study that relied on medical records based on patient reporting and physician
Vol. 25, No. 7, October 2011
Indication for surgery, RCRI, 1-year mortality 907
Table IV. Multivariate analysis for 1-year mortality Patient characteristic
OR
CI
Tissue loss or acute indicationa RCRI score Age ASA-PS ¼ 4 Female Emergency
3.7
1.6-8.5
1.5 1.04 2.7 2.1 3.0
1.2-1.8 1.02-1.06 1.7-4.3 1.3-3.3 1.1-8.1
important intellectual content: V.K.M., B.C.F., C.B., M.M., D.B., J.E.E.; statistical analysis: C.B.; administrative, technical, or material support: M.M.; and study supervision: V.K.M., C.B. p value
0.0058 REFERENCES 0.0002 0.0004 <0.0001 0.002 0.03
There were 113 deaths among 568 patients with rest pain, tissue loss, and acute indications for surgery. There was one 1-year death among 35 patients with claudication. a Factors offered in stepwise regression but not selected included dyslipidemia, beta-blockers, and statins, each of which had a p value between 0.07 and 0.10 in univariate analysis.
documentation. Second, it is possible that patients with limited access to the healthcare system are unaware of their own comorbidities. Third, although cardiac complications may be associated with noncardiac mortality, cause of death could not be determined in our study population.
CONCLUSIONS Our study confirms that the RCRI is a useful tool for predicting short- and long-term outcomes. We believe that to improve on the RCRI, the indication for surgery should also be considered in the risk assessment of patients undergoing open lower extremity bypass surgery. Mortality rates are significantly different among patients presenting with claudication, rest pain, tissue loss, and acute indications. Furthermore, indication for surgery may guide an informed consent process and further preoperative testing. The lack of prognostic value of indication for surgery for 30-day mortality is of small significance compared with its utility for longer term outcomes. This outcome may be of interest to the perioperative physician and patient alike. This investigation should prompt further studies in perioperative medicine to consider indication for surgery as an independent risk factor for perioperative and long-term mortality.
Support for this study was provided solely from institutional and/ or departmental sources. V.K.M. had full access to all the data in the study and takes responsibility for the integrity of these data and the accuracy of their analysis. Study concept and design: V.K.M., B.C.F.; acquisition of data: M.M., B.C.F.; analysis and interpretation of data: V.K.M., C.B., J.E.E.; drafting of the manuscript: V.K.M., B.C.F., C.B., M.M., J.E.E.; critical revision of the manuscript for
1. Fleisher LA, Beckman JA, Brown KA, et al. ACC/AHA 2007 guidelines on perioperative cardiovascular evaluation and care for noncardiac surgery: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines (Writing Committee to Revise the 2002 Guidelines on Perioperative Cardiovascular Evaluation for Noncardiac Surgery): developed in collaboration with the American Society of Echocardiography, American Society of Nuclear Cardiology, Heart Rhythm Society, Society of Cardiovascular Anesthesiologists, Society for Cardiovascular Angiography and Interventions, Society for Vascular Medicine and Biology, and Society for Vascular Surgery. Circulation 2007;116:e418-e499. 2. Lee TH, Marcantonio ER, Mangione CM, et al. Derivation and prospective validation of a simple index for prediction of cardiac risk of major noncardiac surgery. Circulation 1999;100:1043-1049. 3. Ford MK, Beattie WS, Wijeysundera DN. Systematic review: prediction of perioperative cardiac complications and mortality by the revised cardiac risk index. Ann Intern Med 2010;152:26-35. 4. Goodney PP, Nolan BW, Schanzer A, et al. Factors associated with death 1 year after lower extremity bypass in Northern New England. J Vasc Surg 2010;51:71-78. 5. 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. 6. Giles KA, Hamdan AD, Pomposelli FB, Wyers MC, Siracuse JJ, Schermerhorn ML. Body mass index: surgical site infections and mortality after lower extremity bypass from the national surgical quality improvement program 2005-2007. Ann Vasc Surg 2009 (in press). 7. Parlani G, Verzini F, Zannetti S, et al. Does gender influence outcome of AAA endoluminal repair? Eur J Vasc Endovasc Surg 2003;26:69-73. 8. Egorova N, Giacovelli JK, Gelijns A, et al. Defining high-risk patients for endovascular aneurysm repair. J Vasc Surg 2009;50. 1271e1279.e1. 9. Eugster T, Gurke L, Obeid T, Stierli P. Infrainguinal arterial reconstruction: female gender as risk factor for outcome. Eur J Vasc Endovasc Surg 2002;24:245-248. 10. Henke PK, Proctor MC, Zajkowski PJ, et al. Tissue loss, early primary graft occlusion, female gender, and a prohibitive failure rate of secondary infrainguinal arterial reconstruction. J Vasc Surg 2002;35:902-909. 11. Krupski WC, Layug EL, Reilly LM, Rapp JH, Mangano DT. Comparison of cardiac morbidity rates between aortic and infrainguinal operations: two-year follow-up. Study of Perioperative Ischemia Research Group. J Vasc Surg 1993;18: 609-615. discussion 615-617. 12. Christopherson R, Beattie C, Frank SM, et al. Perioperative Ischemia Randomized Anesthesia Trial Study Group. Perioperative morbidity in patients randomized to epidural or general anesthesia for lower extremity vascular surgery. Anesthesiology 1993;79:422-434. 13. Ellis JE, Drijvers G, Pedlow S, et al. Premedication with oral and transdermal clonidine provides safe and efficacious postoperative sympatholysis. Anesth Analg 1994;79:1133-1140.
908 Moitra et al.
14. Wallace AW, Galindez D, Salahieh A, et al. Effect of clonidine on cardiovascular morbidity and mortality after noncardiac surgery. Anesthesiology 2004;101:284-293. 15. Mangano DT, Layug EL, Wallace A, Tateo I. Multicenter Study of Perioperative Ischemia Research Group. Effect of atenolol on mortality and cardiovascular morbidity after noncardiac surgery. N Engl J Med 1996;335:1713-1720. 16. Poldermans D, Boersma E, Bax JJ, et al. Dutch Echocardiographic Cardiac Risk Evaluation Applying Stress Echocardiography Study Group. The effect of bisoprolol on perioperative mortality and myocardial infarction in highrisk patients undergoing vascular surgery. N Engl J Med 1999;341:1789-1794. 17. Poldermans D. Statins and noncardiac surgery: current evidence and practical considerations. Cleve Clin J Med 2009;76(Suppl. 4):S79-S83. 18. Schouten O, Bax JJ, Dunkelgrun M, et al. Statins for the prevention of perioperative cardiovascular complications in vascular surgery. J Vasc Surg 2006;44:419-424. 19. Poldermans D, Bax JJ, Kertai MD, et al. Statins are associated with a reduced incidence of perioperative mortality in
Annals of Vascular Surgery
20.
21.
22.
23.
24.
patients undergoing major noncardiac vascular surgery. Circulation 2003;107:1848-1851. Benedetto U, Sciarretta S, Roscitano A, et al. Preoperative angiotensin-converting enzyme inhibitors and acute kidney injury after coronary artery bypass grafting. Ann Thorac Surg 2008;86:1160-1165. Kaafarani HM, Itani KM, Thornby J, Berger DH. Thirty-day and one-year predictors of death in noncardiac major surgical procedures. Am J Surg 2004;188:495-499. Roddy SP, Darling RC 3rd, Maharaj D, et al. Genderrelated differences in outcome: an analysis of 5880 infrainguinal arterial reconstructions. J Vasc Surg 2003;37: 399-402. Sigvant B, Wiberg-Hedman K, Bergqvist D, Rolandsson O, Wahlberg E. Risk factor profiles and use of cardiovascular drug prevention in women and men with peripheral arterial disease. Eur J Cardiovasc Prev Rehabil 2009;16:39-46. Cheanvechai V, Harthun NL, Graham LM, Freischlag JA, Gahtan V. Incidence of peripheral vascular disease in women: is it different from that in men? J Thorac Cardiovasc Surg 2004;127:314-317.
Background: Patients who undergo vascular surgery are at increased risk of perioperative cardiovascular morbidity and mortality. The Revised Cardiac Risk Index (RCRI) is a validated and widely used bedside tool for estimating the risk of a perioperative major adverse myocardial event. We hypothesized that inclusion of the indication for surgery would add independent and prognostic information to the RCRI in predicting all-cause 30-day and 1-year mortality in open infrainguinal vascular surgical procedures. Methods: This was a retrospective study of 603 patients who underwent open infrainguinal bypass vascular surgery between January 2002 and January 2008 at a tertiary care medical center. RCRI and indication for surgery were determined. The primary outcomes of interest were all-cause 30-day mortality (which included all in-hospital mortality, regardless of time) and all-cause 1-year mortality. Results: Overall 30-day mortality was 32 (5.3%). Independent risk factors for early death were RCRI score, being of age 80 years, American Society of Anesthesiologists Physical Status classification ¼ 4, and emergency surgery. Overall 1-year mortality, including early deaths, was 114 (18.9%). Indication for surgery, RCRI score, age, American Society of Anesthesiologists Physical Status classification ¼ 4, female sex, and emergency surgery were all independent predictors of 1-year mortality. Conclusions: The RCRI score was associated with both 30-day and 1-year mortality in patients undergoing lower extremity bypass surgery. Indication for surgery was predictive of 1-year mortality but not of 30-day mortality.
INTRODUCTION Patients who undergo vascular surgery are at increased risk for perioperative cardiovascular morbidity and mortality. The American College of Cardiology/American Heart Association (ACC/ AHA) perioperative guidelines for noncardiac surgery classify peripheral vascular procedures as being high risk.1 These guidelines also advise careful evaluation of long-term outcomes in such cases. Given the probable comorbidities of this patient
1 Division of Critical Care Medicine, Department of Anesthesiology, College of Physicians and Surgeons of Columbia University, New York, NY. 2 Department of Anesthesiology and Critical Care, Mount Sinai School of Medicine, New York City, NY. 3
Division of Biostatistics, Department of Anesthesiology, Mount Sinai School of Medicine, New York City, NY. 4 Department of Anesthesiology and Critical Care, University of Pennsylvania School of Medicine, Philadelphia, PA.
902
population, understanding perioperative and long-term risk facilitates informed consent and the preoperative testing process. The Revised Cardiac Risk Index (RCRI) is a validated and widely used bedside tool for estimating a patient’s risk of a major adverse myocardial event within 30 days of surgery.2 However, a recent meta-analysis of the RCRI found that this index does not perform well in predicting death in the perioperative period after vascular surgery as compared with other noncardiac procedures.3 Although
Correspondence to: Vivek K. Moitra, MD, Department of Anesthesiology, College of Physicians and Surgeons of Columbia University, 630 West 168th Street, New York, NY 10032, USA, E-mail: vm2161@ columbia.edu Ann Vasc Surg 2011; 25: 902-908 DOI: 10.1016/j.avsg.2011.05.010 Ó Annals of Vascular Surgery Inc. Published online: August 8, 2011
Vol. 25, No. 7, October 2011
indication for lower extremity bypass surgery is associated with poor outcomes,4 the current ACC/AHA guidelines do not consider indication for vascular surgery to be important in stratifying risk. Indications for this surgery include chronic limb ischemia (claudication, ischemic rest pain, or tissue loss) and acute indications such as aneurysm repair and graft thrombosis. We hypothesized that inclusion of the indication for surgery would add independent and prognostic information to the RCRI score in predicting all-cause 30-day and 1-year mortality in patients undergoing open infrainguinal bypass surgery.
MATERIALS AND METHODS This was a retrospective study of vascular surgery patients performed at the Mount Sinai School of Medicine Hospital, New York, NY, an urban tertiary care teaching hospital. The Mount Sinai Hospital Institutional Review Board approved the study, and a waiver of informed consent was obtained. Between January 2002 and January 2008, a total of 603 adults who underwent open femoraledistal arterial bypass vascular surgery procedures and satisfied our inclusion criteria were included in the study. All patients aged >18 years who underwent emergent and nonemergent operations were included. Those undergoing suprainguinal vascular surgery were excluded. In cases involving multiple procedures performed during the study period, only the first one was included in the analysis. Six patients for whom indication for surgery could not be determined were excluded. Data Collection We obtained the following information from the electronic medical records: age, sex, body mass index, American Society of Anesthesiologists Physical Status (ASA-PS) classification, medications, and type of anesthesia. For each subject, we reviewed the chart for documented and self-reported history, and we calculated an RCRI score. One point was assigned for each of the following variables: (a) baseline serum creatinine of 2 mg/dL (to convert to micromoles per liter, multiply by 88.4), (b) history of congestive heart failure, (c) history of ischemic heart disease, (d) history of cerebral vascular disease, and (e) history of diabetes mellitus requiring insulin. In addition, the RCRI score routinely assigns one point for high-risk surgery. However, infrainguinal vascular procedures are not considered high risk for this purpose (in contrast to their classification as high risk by the ACC/AHA).1 Ischemic heart disease is
Indication for surgery, RCRI, 1-year mortality 903
defined as any of the following: (a) previous Q wave on electrocardiogram, (b) previous nitrate use, (c) previous coronary artery bypass or percutaneous intervention, (d) previous positive noninvasive test indicating ischemia, or (e) previous myocardial infarction. Cerebral vascular disease is defined as either previous cerebral vascular accident or transient ischemic attack. We recorded the indication for surgery by using the Rutherford grade classification, which categorizes symptoms of chronic limb ischemia. This classification includes claudication, rest pain, and tissue loss, that is, gangrene.5 Patients who presented with acute limb ischemia were categorized as acute. Outcomes and Measurement The primary outcomes of interest were all-cause 30day mortality (which included all in-hospital mortality, regardless of time) and all-cause 1-year mortality. We obtained this information from hospital discharge summaries and the Social Security Death Index, a public database that lists dates of death for anyone registered with the Social Security Administration. Statistical Methods Data were entered in an Excel file and transferred to SAS version 9.1 (SAS/STAT 9.1 User’s Guide, SAS Institute Inc., Cary, NC) for statistical analysis. Contingency tables were constructed to describe associations between factors. The primary objective of this analysis was to evaluate the RCRI score and indication for surgery as predictors of all-cause 30day and 1-year mortality in patients undergoing lower extremity bypass surgery. Whether indication for surgery provided independent information in these cases was of particular interest. Factors were first reviewed individually for a significant or nearsignificant ( p < 0.10) association with an outcome. Independent factors were identified in separate multiple logistic regression analyses. The joint effect of RCRI and indication for surgery (Rutherford classification) was tested first. Then, when appropriate, RCRI and/or indication for surgery were forced into a stepwise multiple logistic regression analysis to identify other factors that added independent information concerning outcome. Models that included the other independent factors were considered with and without the addition of RCRI and indication. Age, body mass index, and sex were included because of their association with poor outcome after vascular surgery.6-10 A diagnosis of diabetes mellitus was considered because the RCRI score includes only those with diabetes who are on insulin,
904 Moitra et al.
and diabetes has been proposed as an explanation for poor long-term outcome after infrainguinal bypass surgery.11 Dyslipidemia was evaluated because of its association with atherosclerosis and coronary artery disease. Anesthesia type was studied because neuraxial analgesia has been suggested to decrease catecholamine levels and hypercoagulability and to improve outcomes after vascular surgery.1,12 The use of beta-blockers, clonidine, aspirin, and 3-hydroxy-3-methylglutaryl-coenzyme A (HMGCoA) reductase inhibitors was analyzed because of their association with improvements in perioperative outcome.13-20 ASA-PS classification and emergency surgery were considered because of their association with postoperative outcome. A KaplaneMeier life table was used to describe the cumulative incidence of death over the course of the first year.
RESULTS Patient demographics and clinical characteristics are shown in Table I. Patients with claudication were younger (median age, 66.0 years) than those who presented with other indications (median age, 74.0 years). The median RCRI score for patients who had claudication, rest pain, and acute surgery was 1, whereas those who presented with gangrene had a median RCRI score of 2. At the time of surgery, 61% of patients were taking a beta-blocker and 35% were taking an HMG-CoA reductase inhibitor. Among those with tissue loss, 61% were found to be taking a beta-blocker, but only 32% were taking an HMG-CoA reductase inhibitor (data not shown). Thirty-Day Mortality In our patient population, the overall 30-day mortality was 32 (5.3%). There were no early deaths reported in patients with an RCRI score of 0, regardless of surgical indication. In univariate analysis, early death was associated with RCRI score ( p < 0.001), age ( p ¼ 0.02), ASA-PS class ( p ¼ 0.0004), and emergency procedure ( p ¼ 0.004). Indication for surgery was not associated with early death. When the RCRI score and indication for surgery were considered jointly, the RCRI score, but not indication for surgery, was associated with death at 30 days. Therefore, only the RCRI score was forced into the stepwise model to identify other significant factors. Independent risk factors for early death were as follows: (a) RCRI score, (b) being of age 80 years, (c) ASA-PS classification ¼ 4, and (d) emergency surgery (Table II).
Annals of Vascular Surgery
Table I. Patient demographics and clinical characteristics (n ¼ 603)a Characteristics
Frequency
%
32 73 124 189 182
5.33 12.17 20.67 31.50 30.33
313 290
51.91 48.09
35 83 426 59
5.80 13.76 70.65 9.78
98 184 179 98 38 6 0
16.25 30.51 29.68 16.25 6.30 1.00 0
2 13 399 189 0
0.33 2.16 66.17 31.34 0
366 11 205 210 285 384 22
60.70 1.82 34.00 34.82 47.26 63.68 3.65
506 94
84.33 15.67
a
Age group , years <50 50-59 60-69 70-79 80 Gender Male Female Indication for surgery Claudication Rest pain Tissue loss Acute RCRI score 0 1 2 3 4 5 6 Classification ASA-PS 1 ASA-PS 2 ASA-PS 3 ASA-PS 4 ASA-PS 5 Medications Beta-blocker Clonidine Aspirin HMG-CoA reductase inhibitor ACE-I/ARB Diabetes mellitus Emergency Anesthesia typea General Regional
ACE-I/ARB, angiotensin-converting enzyme inhibitor/angiotensin receptor blocker; ASA-PS, American Society of Anesthesiologists Physical Status; HMG-CoA, 3-hydroxy-3-methylglutarylcoenzyme A; RCRI, Revised Cardiac Risk Index. a Missing values: age (3); anesthesia type (3).
One-Year Mortality Overall 1-year mortality, including early deaths, was 114 (18.9%). In univariate analysis, 1-year mortality was significantly associated with (a) indication for surgery ( p ¼ 0.0007), (b) RCRI score ( p < 0.0001), (c) age ( p ¼ 0.0004), (d) ASA-PS classification ( p < 0.0001), (e) sex ( p ¼ 0.0008), and (f) emergency surgery ( p ¼ 0.05).
Vol. 25, No. 7, October 2011
Indication for surgery, RCRI, 1-year mortality 905
Table II. Multivariate analysis for in-hospital mortality or death within 30 daysa Patient characteristic
OR
95% CI
p value
RCRI score Age 80 years ASA-PS ¼ 4 Emergency
1.7 2.2 2.2 7.3
1.2-2.4 1.1-4.7 1.0-5.0 2.3-24.3
0.002 0.04 0.05 0.0008
CI, confidence interval; OR, odds ratio. a There were 32 deaths among 568 patients with rest pain, tissue loss, and acute indications for surgery. There were no in-hospital deaths among 35 patients who presented with claudication.
This analysis revealed trends toward increasing risk of death with increasing RCRI score at each Rutherford level, and also with increasing Rutherford score at each RCRI level. The 1-year mortality rates cross-classified by RCRI score and indication for surgery are shown in Table III. The 1-year mortality stratified by indication for surgery was as follows: (a) claudication, 3%; (b) rest pain, 7%; (c) tissue loss, 22%; and (d) acute indication, 20%. Cumulative 1-year survival stratified by indication of surgery is shown in Figure 1. In multivariate analysis, indication for surgery, RCRI score, age, ASA-PS classification ¼ 4, female sex, and emergency surgery were independent predictors for 1-year mortality (Table IV). Among those patients who presented with claudication, none experienced early death and one died within 1 year. Because the risk of death was so low for these patients, a separate analysis that excluded them (N ¼ 35) was also conducted. However, results of this analysis were similar to the analysis of the group as a whole.
DISCUSSION Previous studies have shown that the RCRI score is an independent indicator of longer term outcome in patients undergoing a variety of procedures, including open and endovascular repair.19-21 These analyses, however, included a variety of vascular procedures with different mortality rates and did not consider indication for surgery. In our analysis, the RCRI score was significantly associated with risk of death at 1 year after open infrainguinal bypass surgery. However, indication for surgery, as well as factors such as age, ASA-PS status, sex, and emergency surgery, provided additional significant and prognostic information. Although the overall long-term mortality rate (18.9%) in this study was similar to the one that was previously reported (16.3%), mortality rates
varied considerably by surgical indication.22 In our study, the overall 1-year mortality was only 3% in patients who had claudication. Those who presented with limb tissue loss, in contrast, had an overall 1-year mortality of 22%. Open infrainguinal bypass surgery is considered high risk by the ACC (in contrast to the RCRI classification), and careful consideration of the indication for surgery is a prudent first step for risk assessment. The 1-year mortality rate of patients with rest pain or claudication was lower than that of patients with lower extremity tissue loss among those with an RCRI of >2 (15 vs. 35%, p ¼ 0.22), as well as among patients with an RCRI of 2 (5 vs. 17%, p ¼ 0.002). These data can be used in the informed consent process when patients are considering further surgery. Because vascular surgery is considered a high-risk procedure, preoperative stress testing is often performed to guide further management before surgery. Given the pain associated with claudication, this patient population may not be able to provide a history of achieving four metabolic equivalents (the equivalent of walking up two flights of stairs). Our data suggest that patients who present with claudication have a low risk of perioperative death or long-term mortality. Further studies should evaluate whether stress testing is prudent or cost-effective in this subgroup of vascular surgery patients, provided medical management has been optimal. Guidelines from the ACC/AHA suggest that clinicians should focus on the long-term management of patients undergoing vascular surgery. Consideration of indication for surgery may prompt the clinician to identify high-risk vascular surgery patients and ensure adequate medical follow-up outside the immediate perioperative period.1 These guidelines discuss magnitude of surgery and surgery type as important variables for immediate perioperative outcome. Our work demonstrates that indication for surgery should also be considered in the analysis of long-term outcome. Differences in 30-day and 1-year predictors of death in patients undergoing noncardiac surgery have been reported.21 Our data show that the RCRI is associated with both 30-day mortality and 1-year mortality in patients undergoing lower extremity bypass surgery. In contrast, indication for surgery is associated with 1-year mortality, but not with 30-day mortality. Our data are consistent with previously reported studies showing that patients who present for urgent and emergent surgery are at increased risk for perioperative morbidity at 30 days. At 1 year, emergent surgery continues to be a risk factor for poor outcome.
906 Moitra et al.
Annals of Vascular Surgery
Table III. One-year mortality rates cross-tabulated by RCRI score and indication for surgery Indication for surgery, number (%) RCRI
Claudication
Rest pain
Tissue
0 1 2 3 4 5 Total
0/14 1/14 0/6 0/1 0/0 0/0 1/35
0/14 2/32 2/25 2/10 0/2 0/0 6/83
3/50 21/118 29/138 25/80 13/34 4/6 95/426
(0) (7) (0) (0) (0) (0) (3)
(0) (6) (8) (20) (0) (0) (7)
Acute
(6) (18) (21) (31) (38) (67) (22)
1/20 5/20 3/10 2/7 1/2 0/0 12/59
Total
(5) (25) (30) (29) (50) (0) (20)
4/98 29/184 34/179 29/98 14/38 4/6 114/603
(4) (16) (19) (30) (37) (67) (18.9)
(%) ¼ mortality rate per group.
Fig. 1. Survival from service date to earlier of death or 1 year.
Indication for surgery may be a more accurate reflection of a patient’s overall health status compared with the RCRI. The reliability of the RCRI depends on the diagnosis of comorbid conditions. In patients with limited access to health care, an accurate score may not be possible. Furthermore, patients who present with gangrene may be less compliant with medication or may have limited access to preventive healthcare resources, compared with those who present with claudication. About 65% of patients with tissue loss have been found to be taking a beta-blocker, but only 32% were taking an HMG-CoA reductase inhibitor. For this reason, indication for lower extremity surgery, that is, gangrene, may reflect a patient’s undiagnosed comorbidities, serving as a useful objective
adjunct to quantify long-term risk. An atherosclerotic burden may manifest itself as an indication for vascular surgery. The influence of sex on outcomes in these patients has been reported.9,10,22 Our results show that female sex is not associated with in-hospital mortality, but is definitely associated with death after 1 year. This finding may be explained by anatomical differences and difficulty with vascular intervention, extent of disease, and access to health care for preventing complications.8,23,24 Chronic cardiac medication did not show an independent association with outcome in this study. Our investigation has several limitations. First, it was a retrospective study that relied on medical records based on patient reporting and physician
Vol. 25, No. 7, October 2011
Indication for surgery, RCRI, 1-year mortality 907
Table IV. Multivariate analysis for 1-year mortality Patient characteristic
OR
CI
Tissue loss or acute indicationa RCRI score Age ASA-PS ¼ 4 Female Emergency
3.7
1.6-8.5
1.5 1.04 2.7 2.1 3.0
1.2-1.8 1.02-1.06 1.7-4.3 1.3-3.3 1.1-8.1
important intellectual content: V.K.M., B.C.F., C.B., M.M., D.B., J.E.E.; statistical analysis: C.B.; administrative, technical, or material support: M.M.; and study supervision: V.K.M., C.B. p value
0.0058 REFERENCES 0.0002 0.0004 <0.0001 0.002 0.03
There were 113 deaths among 568 patients with rest pain, tissue loss, and acute indications for surgery. There was one 1-year death among 35 patients with claudication. a Factors offered in stepwise regression but not selected included dyslipidemia, beta-blockers, and statins, each of which had a p value between 0.07 and 0.10 in univariate analysis.
documentation. Second, it is possible that patients with limited access to the healthcare system are unaware of their own comorbidities. Third, although cardiac complications may be associated with noncardiac mortality, cause of death could not be determined in our study population.
CONCLUSIONS Our study confirms that the RCRI is a useful tool for predicting short- and long-term outcomes. We believe that to improve on the RCRI, the indication for surgery should also be considered in the risk assessment of patients undergoing open lower extremity bypass surgery. Mortality rates are significantly different among patients presenting with claudication, rest pain, tissue loss, and acute indications. Furthermore, indication for surgery may guide an informed consent process and further preoperative testing. The lack of prognostic value of indication for surgery for 30-day mortality is of small significance compared with its utility for longer term outcomes. This outcome may be of interest to the perioperative physician and patient alike. This investigation should prompt further studies in perioperative medicine to consider indication for surgery as an independent risk factor for perioperative and long-term mortality.
Support for this study was provided solely from institutional and/ or departmental sources. V.K.M. had full access to all the data in the study and takes responsibility for the integrity of these data and the accuracy of their analysis. Study concept and design: V.K.M., B.C.F.; acquisition of data: M.M., B.C.F.; analysis and interpretation of data: V.K.M., C.B., J.E.E.; drafting of the manuscript: V.K.M., B.C.F., C.B., M.M., J.E.E.; critical revision of the manuscript for
1. Fleisher LA, Beckman JA, Brown KA, et al. ACC/AHA 2007 guidelines on perioperative cardiovascular evaluation and care for noncardiac surgery: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines (Writing Committee to Revise the 2002 Guidelines on Perioperative Cardiovascular Evaluation for Noncardiac Surgery): developed in collaboration with the American Society of Echocardiography, American Society of Nuclear Cardiology, Heart Rhythm Society, Society of Cardiovascular Anesthesiologists, Society for Cardiovascular Angiography and Interventions, Society for Vascular Medicine and Biology, and Society for Vascular Surgery. Circulation 2007;116:e418-e499. 2. Lee TH, Marcantonio ER, Mangione CM, et al. Derivation and prospective validation of a simple index for prediction of cardiac risk of major noncardiac surgery. Circulation 1999;100:1043-1049. 3. Ford MK, Beattie WS, Wijeysundera DN. Systematic review: prediction of perioperative cardiac complications and mortality by the revised cardiac risk index. Ann Intern Med 2010;152:26-35. 4. Goodney PP, Nolan BW, Schanzer A, et al. Factors associated with death 1 year after lower extremity bypass in Northern New England. J Vasc Surg 2010;51:71-78. 5. 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. 6. Giles KA, Hamdan AD, Pomposelli FB, Wyers MC, Siracuse JJ, Schermerhorn ML. Body mass index: surgical site infections and mortality after lower extremity bypass from the national surgical quality improvement program 2005-2007. Ann Vasc Surg 2009 (in press). 7. Parlani G, Verzini F, Zannetti S, et al. Does gender influence outcome of AAA endoluminal repair? Eur J Vasc Endovasc Surg 2003;26:69-73. 8. Egorova N, Giacovelli JK, Gelijns A, et al. Defining high-risk patients for endovascular aneurysm repair. J Vasc Surg 2009;50. 1271e1279.e1. 9. Eugster T, Gurke L, Obeid T, Stierli P. Infrainguinal arterial reconstruction: female gender as risk factor for outcome. Eur J Vasc Endovasc Surg 2002;24:245-248. 10. Henke PK, Proctor MC, Zajkowski PJ, et al. Tissue loss, early primary graft occlusion, female gender, and a prohibitive failure rate of secondary infrainguinal arterial reconstruction. J Vasc Surg 2002;35:902-909. 11. Krupski WC, Layug EL, Reilly LM, Rapp JH, Mangano DT. Comparison of cardiac morbidity rates between aortic and infrainguinal operations: two-year follow-up. Study of Perioperative Ischemia Research Group. J Vasc Surg 1993;18: 609-615. discussion 615-617. 12. Christopherson R, Beattie C, Frank SM, et al. Perioperative Ischemia Randomized Anesthesia Trial Study Group. Perioperative morbidity in patients randomized to epidural or general anesthesia for lower extremity vascular surgery. Anesthesiology 1993;79:422-434. 13. Ellis JE, Drijvers G, Pedlow S, et al. Premedication with oral and transdermal clonidine provides safe and efficacious postoperative sympatholysis. Anesth Analg 1994;79:1133-1140.
908 Moitra et al.
14. Wallace AW, Galindez D, Salahieh A, et al. Effect of clonidine on cardiovascular morbidity and mortality after noncardiac surgery. Anesthesiology 2004;101:284-293. 15. Mangano DT, Layug EL, Wallace A, Tateo I. Multicenter Study of Perioperative Ischemia Research Group. Effect of atenolol on mortality and cardiovascular morbidity after noncardiac surgery. N Engl J Med 1996;335:1713-1720. 16. Poldermans D, Boersma E, Bax JJ, et al. Dutch Echocardiographic Cardiac Risk Evaluation Applying Stress Echocardiography Study Group. The effect of bisoprolol on perioperative mortality and myocardial infarction in highrisk patients undergoing vascular surgery. N Engl J Med 1999;341:1789-1794. 17. Poldermans D. Statins and noncardiac surgery: current evidence and practical considerations. Cleve Clin J Med 2009;76(Suppl. 4):S79-S83. 18. Schouten O, Bax JJ, Dunkelgrun M, et al. Statins for the prevention of perioperative cardiovascular complications in vascular surgery. J Vasc Surg 2006;44:419-424. 19. Poldermans D, Bax JJ, Kertai MD, et al. Statins are associated with a reduced incidence of perioperative mortality in
Annals of Vascular Surgery
20.
21.
22.
23.
24.
patients undergoing major noncardiac vascular surgery. Circulation 2003;107:1848-1851. Benedetto U, Sciarretta S, Roscitano A, et al. Preoperative angiotensin-converting enzyme inhibitors and acute kidney injury after coronary artery bypass grafting. Ann Thorac Surg 2008;86:1160-1165. Kaafarani HM, Itani KM, Thornby J, Berger DH. Thirty-day and one-year predictors of death in noncardiac major surgical procedures. Am J Surg 2004;188:495-499. Roddy SP, Darling RC 3rd, Maharaj D, et al. Genderrelated differences in outcome: an analysis of 5880 infrainguinal arterial reconstructions. J Vasc Surg 2003;37: 399-402. Sigvant B, Wiberg-Hedman K, Bergqvist D, Rolandsson O, Wahlberg E. Risk factor profiles and use of cardiovascular drug prevention in women and men with peripheral arterial disease. Eur J Cardiovasc Prev Rehabil 2009;16:39-46. Cheanvechai V, Harthun NL, Graham LM, Freischlag JA, Gahtan V. Incidence of peripheral vascular disease in women: is it different from that in men? J Thorac Cardiovasc Surg 2004;127:314-317.