Clinical research Coronary heart disease

Transcription

1 European Heart Journal (2007) 28, doi: /eurheartj/ehl390 Clinical research Coronary heart disease A clinical survival score predicts the likelihood to benefit from preoperative thallium scanning and coronary revascularization before major vascular surgery Giora Landesberg 1 *, Yacov Berlatzky 3, Moshe Bocher 4, Ron Alcalai 2, Haim Anner 3, Tatyana Ganon-Rozental 1, Myron H. Luria 2, Inna Akopnik 3, Charles Weissman 1, and Morris Mosseri 2 1 Departments of Anaesthesiology and Critical Care Medicine, Hebrew University, Hadassah Medical Center, Jerusalem 91120, Israel; 2 Cardiology Hebrew University, Hadassah Medical Center, Jerusalem 91120, Israel; 3 Vascular Surgery Hebrew University, Hadassah Medical Center, Jerusalem 91120, Israel; and 4 Nuclear Medicine Hebrew University, Hadassah Medical Center, Jerusalem 91120, Israel Received 23 June 2006; revised 9 October 2006; accepted 3 November 2006; online publish-ahead-of-print 27 November 2006 See page 519 for the editorial comment on this article (doi: /eurheartj/ehl523) KEYWORDS Preoperative thallium scanning; Coronary revascularization; Vascular surgery; Long-term survival Introduction Cardiac disease is a major cause of both perioperative and long-term morbidity and mortality in vascular surgery patients. 1 3 Patients with large-vessel peripheral artery disease are at 6 15-fold higher risk of long-term death from coronary artery disease (CAD) than patients without peripheral arterial disease. 4 Studies using routine preoperative coronary angiography showed that 60% of all vascular surgery patients have at least one major coronary artery with critical (.70%) stenosis and 14% have critical threevessel and/or left main (LM) disease that may warrant coronary revascularization (CR). 5 Therefore, an important goal of preoperative cardiac evaluation in elective vascular surgery patients should include a resource-effective and * Corresponding address. Tel: þ ; fax: þ address: gio@cc.huji.ac.il Aims Guidelines advocate selective non-invasive testing before major non-cardiac surgery, yet data defining who may benefit from such tests is lacking. We aimed to find the predictors that define patients who are most likely to benefit from preoperative cardiac testing and coronary revascularization (CR). Methods and results In 624 consecutive major vascular surgery patients, the preoperative thallium scanning (PTS) results and subsequent CRs were correlated with long-term (3 15 years) survival. Of all patients, 510 (80.6%) had PTS, 154 (24.7%) had moderate-severe ischaemia on PTS, and 96 (15.4%) underwent CR. Seven predictors: age 65, diabetes, cerebrovascular disease, ischaemic heart disease, congestive heart failure, ST-depression on preoperative ECG, and renal insufficiency, independently determined long-term survival. A long-term survival score (LTSS) comprised of these predictors, divided all patients into low, intermediate, and high-risk groups (0 1, 2 3, 4 predictors, respectively) with a 5-year survival of %, %, and %, respectively. Compared with low-risk patients, intermediate and high-risk patients had worse survival [HR (CI) ¼ 2.6 ( ) and 5.9 ( ), respectively, P, 0.001]. Yet, only the intermediate-risk group had better long-term survival following preoperative CR [HR ¼ 0.48 ( ), P ¼ 0.001]. The low-risk groups favourable survival and highrisk groups poor survival were not significantly affected by CR. Conclusion Intermediate-risk patients (LTSS 2 3) are most likely to have a long-term survival benefit from PTS and CR. safe approach to detecting the highest-risk patients and preventing not only perioperative but also long-term cardiovascular morbidity and mortality. The American College of Cardiology/American Heart Association (ACC/AHA) guidelines, 6 advocated non-invasive testing in patients with intermediate clinical risk criteria (angina pectoris, previous myocardial infarction, compensated heart failure, diabetes mellitus, and/or renal insufficiency) as well as in patients with minor risk criteria and poor functional capacity, which includes close to 80% of patients undergoing major vascular surgery. 7 In contrast, the recent Coronary Artery Randomization Prophylaxis (CARP) study 8 showed no effect of preoperative CR on perioperative or 3-years morbidity and mortality in patients undergoing major vascular surgery. This latter trial however, had some important limitations: preoperative screening was not performed according to the ACC/AHA, only 8.9% of all collected patients were eventually enrolled & The European Society of Cardiology All rights reserved. For Permissions, please journals.permissions@oxfordjournals.org

2 534 G. Landesberg et al. in the randomization, only 44% of randomized patients had significant ischaemia on preoperative non-invasive testing and only 33% of randomized patients had triple-vessel CAD, while patients with left main CAD were excluded. Thus patients who were most likely to benefit from revascularization were a minority in that study. Currently, there is a consensus that the guidelines for preoperative cardiac assessment should be amended to reflect recent data, yet, the issue of whom to screen and how to screen preoperative patients is far from settled. 9 A most recent review article 10 advocated non-invasive testing in most patients with three or more Revised Cardiac Risk Index criteria. 11 Yet again, none of these reports was based on systematic preoperative testing and subsequent selective coronary intervention to support such recommendations. More importantly, all available guidelines are focused mainly on the immediate postoperative outcome rather than on the long-term survival of these high-risk patients. The present study aimed to define the vascular surgery patients who are most likely to have a long-term survival benefit from preoperative thallium scanning (PTS) and selective CR. Methods In accordance with the standards of the Hospital s Ethics Committee, all patients undergoing elective open repair of the abdominal aorta or lower-extremity arterial bypass surgery at Hadassah University Hospital, between 1990 and 2002 were studied. Patients who had re-operations were included in the study only once, in their first operation. Preoperative cardiac evaluation and CR During the period of the study, PTS was routinely performed in our institution in patients scheduled for major vascular surgery. Patients were exempt from PTS only if they had coronary angiography or revascularization within the last year with no change in symptoms, if they had no history of ischaemic heart disease (IHD) and had a recent normal exercise stress test, or if they had no history of IHD and a severe leg ischaemia prohibited further delay of the surgery. Our systematic regimen for preoperative cardiac evaluation has been previously described. 12 In brief, depending on their ability to exercise, patients were stressed by upright treadmill exercise test or by dipyridamole infusion (0.56 mg/kg BW over 4 min) and 3 min of isometric hand grip exercise. Two mci of thallium-201 were given at peak exercise. Immediate and 4 h delayed SPECT images were obtained, with additional 1.0 mci of thallium given prior to the delayed images. Defect size was determined based on a 9-sector model of the heart: anterior, lateral, inferior, and posterior walls, each divided into basal and apical regions, plus the apex. A defect larger than two sectors was defined as large, one to two sectors as moderate, and less than one as small size defect. Defect severity was evaluated based on the ratio of defect intensity to presumed normal myocardial area. A reduction of 40% in counts was considered moderate-severe defect. 12,13 Reversibility of perfusion defect was defined as an improvement of at least 10% in counts during delayed images as compared with stress images. Patients with moderate or severe reversible defects, even if partially reversible, or large areas (greater than two sectors) of even mildly reversible defects on thallium imaging, especially those coinciding with increased lung uptake, transient left ventricular (LV) dilatation, or ST-segment depression during stress, were defined as moderate-severe ischaemia and were referred to coronary angiography and possible revascularization by either PCI or CABG. Patients who had either no or mild ischaemia, or only fixed defects were allowed to undergo the vascular surgery without further cardiac investigation. Cardiovascular medications were continued perioperatively. Monitoring included intra-arterial blood pressure and pulse oximetry measurements. Central venous and pulmonary artery pressure monitoring was used almost exclusively in aortic surgery. After surgery, patients were observed in an intensive care area at least until the morning after surgery. Data collection The preoperative clinical findings and perioperative data were collected from patients files and hospital s computerized information system by investigators who were blinded to the process of data analysis. History of IHD was defined as previous infarction, angina pectoris, or CR. Congestive heart failure was defined as shortness of breath on exertion with documented reduced LV function by echocardiography, or a history of pulmonary oedema. Cerebrovascular disease was defined as previous stroke, transient ischaemic attack, or carotid endarterectomy surgery. Preoperative resting 12-lead ECGs were analysed for the presence of: pathological Q waves, ST-segment depression.0.5 mm (in any lead except L 3, avr or V 1 ), T-wave inversion, or LV hypertrophy according to the Sokolow-Lyon criteria, 14 and agreed upon by two investigators, as previously reported. 15 All preoperative thallium scans, coronary angiograms, PCI, or CABG data were collected. The extent of CAD was classification as single-, double-, or triple-vessel disease based on the number of three major coronary arteries with 70% stenosis and left main disease was defined as a stenosis of 50% of its internal diameter. Survival was recorded from the hospital s information system, which is updated monthly from the Israeli Ministry of Interior Affairs with all newly deceased individuals. Statistical analyses Logistic regression analyses, Kaplan Meier log-rank tests, and Cox proportional hazards models identified predictors of survival. Independent predictors of long-term survival were identified using stepwise, backward, likelihood-ratio selection Cox proportional hazards multivariable models. Cox s multivariable analysis was obtained twice: first time all patients and all variables except the thallium scanning results (to avoid missing data in patient without thallium scanning) were subjected to the analysis; the second time only patients who had PTS were included. Internal validation of the long-term survival score (LTSS) was performed by bootstrapping technique, using 1000 random resamples from the entire data set. Bootstrapping was applied to both logistic regression and Cox proportional hazards analyses predicting 3-years and long-term survival, respectively. The mean and bias-corrected CIs of the bootstrapped ORs and HRs were reported. A two-sided significance level of 0.05 was used in all analyses. Statistical analyses were performed using SPSS version 11.0 (SPSS Inc., Chicago, IL, USA) and S-PLUS version 7.03 (Insightful Co., Seattle, WA, USA). Results The cohort consisted of 624 consecutive patients. Of those, 510 (80.6%) patients had PTS, 154 (30.2%) had moderatesevere ischaemia on thallium scanning, and 96 (15.4%) underwent preoperative CR, the majority of them, 91 (94.8%) patients, following moderate-severe ischaemia on PTS. The 95 CRs included 54 percutaneous coronary interventions (PCIs) (41 with balloon angioplasty only, 13 with bare metal stents) and 42 coronary artery bypass grafting operations (CABG). The mean duration between preoperative CR and the vascular surgery was days (median ¼ 36 days). Among patients with preoperative CR, 18 (19%) had left main CAD, 55 (57%) had triple-vessel

3 Survival after revascularization in vascular surgery 535 disease, and 73% had left main and/or triple-vessel CAD. Patients with moderate-severe ischaemia who did not eventually undergo preoperative CR (63 patients) consisted of two main groups: 28 patients who did not undergo coronary angiography based on the clinical judgment of the treating cardiologists that their overall risk for surgery was low even considering the thallium scanning results; 35 patients in whom the coronary angiography revealed either mild CAD or such that it is not amenable for revascularization. 12 As per the ACC/AHA guidelines, 374 (59.9%) of the patients had intermediate clinical risk predictors, i.e. angina pectoris, prior myocardial infarction, diabetes mellitus, and/or renal insufficiency (creatinine.2 mg/dl), and 553 (89.9%) had either minor or intermediate clinical risk predictors. Table 1 summarizes the preoperative demographic and clinical feature of all patients, patients with moderate-severe ischaemia on thallium scanning, and those with preoperative CR. Long-term Survival Score During a mean follow-up period of years [range: 3 15 years, median: 5.4, IQR: years], 305 (48.9%) Table 1 Preoperative demographic and clinical data All patients (n ¼ 624) patients died: 59 (9.5%) within 1 year, 95 (15.2%) within 2 years, and 129 (20.7%) within 3 years of follow-up. Age 65, diabetes mellitus, cerebrovascular disease, IHD, congestive heart failure, renal insufficiency, ST-depression on preoperative ECG, and chronic b-blockade were the independent predictors of worse long-term survival while preoperative CR and statins were independent predictors of improved survival (Table 2). An LTSS ranging from 0 to 7 points comprised of the number of preoperative clinical predictors of mortality: age 65, diabetes mellitus, cerebrovascular disease, IHD, congestive heart failure, ST-depression on preoperative 12-lead ECG, and renal insufficiency, strongly predicted survival (Figure 1). Based on their LTSS, patients were divided into three groups: low-risk (LTSS 0 1), intermediate-risk (LTSS 2 3), and high-risk (LTSS 4). Eight (2.9%) low-risk patients, 19 (6.9%) intermediate-risk patients, and 10 (14.7%) high-risk patients died within the first 6 months after surgery. The corresponding 3-years mortality in the three groups were 11.5, 24.2, and 44.1%, respectively. Altogether, 89 (31.9%) low-risk, 160 (57.8%) intermediate-risk, and 56 (82.4%) highrisk patients died during the entire follow-up period. Compared with low-risk patients, intermediate- and high-risk Patients with moderate-severe ischaemia on thallium scan (n ¼ 154) Patients with preoperative CR (n ¼ 96) Age Gender (male/female) 469 (75.2)/155 (24.8) 112 (72.7)/42 (27.3) 77 (80.2)/19 (19.8) Surgery Infra-inguinal bypass procedures 403 (64.6) 97 (63.0) 58 (60.4) Abdominal aortic surgery 221 (35.4) 57 (37.0) 38 (39.6) Diabetes mellitus 203 (32.5) 67 (43.5) 37 (38.5) Non-insulin dependent 146 (23.4) 45 (29.2) 27 (28.1) Insulin dependent 57 (9.1) 22 (14.3) 10 (10.4) Hypertension 316 (50.6) 79 (51.3) 45 (46.9) Smoking history 350 (56.1) 91 (59.1) 60 (62.5) History of IHD 288 (46.2) 108 (70.1) 81 (84.4) History of myocardial infarction 174 (27.9) 66 (42.9) 44 (45.8) Angina pectoris 93 (14.9) 41 (26.6) 29 (30.2) Congestive heart failure 47 (7.5) 23 (14.9) 17 (17.7) Cerebrovascular disease 78 (12.5) 24 (15.6) 10 (10.4) Preoperative ECG Pathological Q-waves 178 (28.5) 66 (42.9) 47 (49.0) ST-depression (.0.5 mm) 127 (20.4) 48 (31.2) 25 (26.0) T-wave inversion 143 (22.9) 59 (38.3) 34 (35.4) LVH (voltage criteria) 96 (15.4) 31 (20.1) 15 (15.6) PTS 510 (80.6) 154 (100) 91 (94.8) Moderate-severe fixed defects 134 (26.2) a 64 (40.9) 35 (36.4) Moderate-severe reversible defects 154 (30.2) a 154 (100) 91 (94.6) Preoperative CR 96 (15.4) 91 (59.1) 96 (100) CABG 42 (6.7) 41 (26.6) 42 (43.7) PCI 54 (8.6) 50 (32.5) 54 (56.3) Kidney disease (creatinine 2 mg/dl) 33 (5.3) 6 (3.9) 3 (3.1) Medications Calcium blockers 201 (32.2) 64 (41.6) 47 (49.0) b-blockers 104 (16.7) 29 (18.8) 19 (19.8.) ACE-inhibitors 156 (25.0) 50 (32.5) 34 (35.4) Diuretics 89 (14.3) 30 (19.5) 16 (16.7) Digoxin 26 (4.2) 10 (6.5) 8 (8.3) Aspirin 307 (49.1) 255 (48.3) 52 (54.1) Hypolipidaemic agents (statins) 95 (15.2) 24 (15.6) 22 (22.9) a Percentage of patients who had PTS.

4 536 G. Landesberg et al. Table 2 Long-term survival analysis (Cox proportional hazards) Univariate analysis Multivariate analysis HR (95% CI) P-value HR (95% CI) P-value Age ( ), ( ),0.001 a 1.86 ( ),0.001 b Male sex 0.86 ( ) 0.23 Type of surgery Low-extremity bypass vs. Aortic 1.47 ( ) Diabetes mellitus 2.08 ( ), ( ),0.001 a 1.88 ( ),0.001 b Non-insulin dependent 1.52 ( ) Insulin dependent 1.66 ( ) Hypertension 1.29 ( ) Smoking history 0.82 ( ) Cerebrovascular disease 1.93 ( ), ( ),0.001 a 1.88 ( ),0.001 b IHD 1.60 ( ), ( ) a 1.49 ( ) b History of myocardial infarction 1.42 ( ) History of angina pectoris 1.34 ( ) Congestive heart failure 1.95 ( ), ( ) a 1.41 ( ) b Kidney disease (creatinine 2 mg/dl) 2.53 ( ), ( ),0.001 a 2.19 ( ) b Preoperative ECG: Q-waves 1.24 ( ) ST-depression (.0.5 mm) 2.59 ( ), ( ),0.001 a 1.72 ( ),0.001 b T-wave inversion 1.80 ( ),0.001 LV hypertrophy (V) 1.80 ( ),0.001 PTS Moderate-severe fixed defect 1.26 ( ) Moderate-severe reversible defect 1.45 ( ) Preoperative CR 0.89 ( ) ( ) b Medications Calcium blockers 1.14 ( ) 0.27 b-blockers 1.27 ( ) ( ) b ACE-inhibitors 1.18 ( ) 0.21 Diuretics 1.75 ( ),0.001 Digoxin 1.94 ( ) Aspirin 0.97 ( ) 0.76 Hypolipidaemic agents 0.64 ( ) ( ) a 0.50 ( ) b a Variables selected by the model when all 624 patients and all predictors are included in the multivariate Cox regression analysis, except thallium scanning results. b Variables selected by the model when only 510 patients with PTS are included in the multivariate Cox regression analysis. patients had significantly worse survival: [HR (CI) ¼ 2.6 ( ) and 5.9 ( )], respectively, P, Table 3 summarizes the results of bootstrap analyses performed on both 3-years (logistic regression analyses) and long-term mortality (Cox proportional hazards analyses). It shows the strong prognostic value of LTSS in predicting mortality. Preoperative thallium scanning and coronary revascularization The proportion of patients with moderate-severe ischaemia on thallium scanning was 12.8, 33.6, and 39.7% and the proportion of CRs was 6.8, 22.7, and 20.6% in low-risk, intermediate-risk, and high-risk patients, respectively (Figure 2). CR was independently associated with improved long-term survival when subjected with all preoperative risk predictors, including the thallium scanning results into the multivariable survival analysis [HR ¼ 0.67 ( ), P ¼ 0.029, Table 2]. However, in a subgroup analysis of both 3-years (logistic regression) and long-term (Cox s proportional hazards) mortality comparing patients with and without CR, only the intermediate-risk group had significantly better survival following CR (Table 4, Figure 3). When only patients with moderate-severe ischaemia on PTS were included in the analysis, the intermediate-risk patients had even better long-term survival if treated preoperatively by CR [HR ¼ 0.40 ( ), P ¼ 0.001, respectively], with no significant difference in either low- or high-risk patients [HR ¼ 1.86 ( ) and 0.89 ( ), P ¼ 0.33 and 0.80, respectively]. No statistically significant effect of preoperative CR was noted on short-term, 6-months, or even 1-year mortality in none of the risk groups.

5 Survival after revascularization in vascular surgery 537 Discussion Our principal findings: (1) Seven predictors independently determine long-term (up to 15 years) mortality after major vascular surgery: age 65, diabetes mellitus, cerebrovascular disease, IHD, congestive heart failure, ST-segment depression (0.5 mm) on preoperative ECG and renal insufficiency (creatinine 2 mg/dl). A simplified LTSS consisting of one point for each one of the seven predictors strongly predicts mortality and divides all patients into low-risk (LTSS 0 1), intermediate-risk (LTSS 2 3), and highrisk (LTSS 4) groups with significantly distinct long-term survival rates (Figure 1). (2) Intermediate-risk patients, with 2 3 of the seven predictors of long-term mortality, are most likely to have better long-term survival if treated by preoperative CR [HR ¼ 0.46 ( ), P, 0.001]. In contrast, low-risk patients (LTSS 0 1) have favourable survival and high-risk patients (LTSS 4) have poor survival (15 20% deaths/year) not significantly affected by preoperative CR (Figure 3). These results are consistent with and expand on two previous reports with significant impact on preoperative management of high-cardiac-risk patients undergoing non-cardiac surgery. One is the ACC/AHA consensus document 6 whose guidelines for preoperative non-invasive testing are based on extensive review of the literature, yet have not been tested or validated by a large series of high-risk patients. The other is the study by Lee et al. 16 that derived and validated a simplified Revised Figure 1 Kaplan Meier survival curves of all patients grouped according to their LTSS. The long-term survival of each group is significantly different from all other groups (P, 0.01). See online supplementary material for a colour version of this figure. Cardiac Risk Index (RCRI), but lacked specific correlation with preoperative testing, CR, or long-term outcome data. The present study that includes the largest cohort of high-risk non-cardiac surgery patients treated according to a systematic regimen of preoperative non-invasive testing and selective CR, bridges between those two works. The LTSS formulated in this study matches Lee s RCRI 17 criteria (high-risk type of surgery, IHD, congestive heart failure, cerebrovascular disease, insulindependent diabetes, and kidney failure), yet with the addition of age 65,ST-segmentdepressiononpreoperativeresting ECG, 18 and both insulin-dependent and non-insulin-dependent diabetes mellitus as equally important independent predictors of long-term mortality. We have previously reported in a smaller cohort, that CR in vascular surgery patients with moderate-severe ischaemia on PTS is associated with improved long-term survival. 14 The present study expands on that report by further defining the patients who are most likely to benefit from preoperative cardiac evaluation and revascularization. Recently, Hertzer et al. reported improved long-term survival following preoperative CR in a cohort of 855 patients undergoing abdominal aortic aneurysmectomy. 16 Back et al. 17 showed that patients who had CR less than 5 years before major vascular surgery, had better survival than those with older CRs (more than 5 years before the surgery) and better than high-risk patients with no CR at all (73 vs. 58% and 62%, respectively). These results as well as ours 19 differ, however, from the CARP trial that showed no difference in perioperative or 3-years outcome in vascular surgery patients randomized to CR vs. medical therapy. 8 Yet, as previously noted, only 33% of the patients included in the CARP trial had triple-vessel CAD (patients with LM CAD were excluded), much lower proportion than the 73% patients with LM and/or triple-vessel disease that had CR in the present cohort. This difference in the proportion of patients with highgrade CAD, who are most likely to benefit from CR and the main aim of any preoperative cardiac investigation, may explain the difference in results between the studies. 20,21 Study limitations Its retrospective nature and the fact that the results are based on a single institution without an external validation group are obvious limitations. However, these data are unique in the sense that they are derived from an institution with long-term adherence to a strict clinical protocol for preoperative cardiac assessment using PTS, selective coronary angiography, and CR to patients with significant inducible ischaemia on preoperative testing and high-grade CAD amenable for revascularization. External validation could support the generalizeability of our LTSS, however, studies Table 3 Bootstrap validation of the value of LTSS in predicting 3-years and long-term mortality LTSS Mortality in 3 years logistic regression analysis Long-term mortality Cox proportional hazards analysis Observed OR Bootstrap OR mean (95% CI) P-value Observed HRs Bootstrap HRs mean (95% CI) P-value ( ) ( ), ( ) ( ), ( ), ( ), ( ), ( ),0.001 CI denotes bias-corrected confidence intervals.

6 538 G. Landesberg et al. have also shown that internal validation by bootstrap resampling on the original data is as effective as having a separate test sample twice as large. 22 Moreover, the LTSS is an assembly of seven well known predictors of long-term survival each one of them was previously confirmed and validated in numerous clinical settings. Five of the seven predictors were validated by Lee et al. 11 as predicting early postoperative complications. Age and ST-segment depression on resting ECG are additional well recognized determinants of long-term mortality. 23,24 Our study lacked the implementation of guidelines for prophylactic perioperative b-blockade, mainly because these guidelines were published after the enrolment of most of our patients. 25 While this is potentially a major limitation, it does not refute our main result that intermediate-risk patients (LTSS 2 3) are the most likely group of patients among all vascular surgery patients to benefit from preoperative cardiac investigation and revascularization. Furthermore, the strength of the evidence for routine use of prophylactic perioperative b-blockade is still intensely debated. 26 Two recent studies failed to show a favourable effect of prophylactic perioperative betablockers, 27,28 and the whole subject await the results of another large scale randomized controlled trial (POISE). 26 Figure 2 Bar plot demonstrating the distribution of all patients, patients with PTS, patients with moderate-severe ischaemia on thallium scanning, and patients with CR, in relation to their LTSS. See online supplementary material for a colour version of this figure. Moreover, the DECREASE study itself, 25 which provided the main data on the beneficial effect of perioperative b-blockade, specifically excluded from randomization all patients with evidence during dobutamine stress echocardiography (DSE) for LM or three-vessel CAD who were most likely to benefit from preoperative CR. In our study chronic beta-blockade was associated with worse long-term survival, probably denoting sicker patients. This is in contrast to statins which were associated with improved long-term survival, independent of the effect of CR, and in agreement with numerous previous publications. Cholesterol-lowering medications are currently one of the mainstays in therapy of cardiovascular patients and it is hard to predict how our long-term results with CR would have been if all our patients were treated according to the current evidence-based lipid-lowering therapy. Conclusions Vascular surgery patients with intermediate-risk LTSS 2 3 are most likely to have a long-term benefit from CR if highgrade CAD amenable for revascularization are detected following preoperative cardiac testing. CR is not likely to affect low-risk patients favourable survival or to improve high-risk patients poor survival, determined mainly by their Figure 3 Kaplan Meier long-term survival curves of patients with vs. without preoperative CR in each one of the three groups: low-risk (LTSS 0 1); intermediate-risk (LTSS 2 3); high-risk (LTSS 4). Only the intermediate-risk patients had significantly better long-term survival with preoperative CR (P ¼ 0.001). See online supplementary material for a colour version of this figure. Table 4 The effect of LTSS and CR on 3-years and long-term mortality 3-years mortality logistic regression Long-term mortality Cox s proportional hazards All patients No CR, no. (%) With CR, no. (%) OR (CI) P-value HR (CI) P-value Low-risk LTSS /279 (11.5%) 32/260 (12.3) 0/19 NA 0.14 a 1.09 ( ) 0.81 Intermediate-risk LTSS /277 (24.4%) 58/241 (27.1) 9/63 (14.3) 0.45 ( ) ( ) High-risk LTSS 4 30/68 (44.1%) 25/54 (46.3) 5/14 (35.7) 0.89 ( ) ( ) 0.73 Total 129/624 (20.7%) 115/528 (21.8) 14/96 (14.6) 0.65 ( ) ( ) 0.50 a Fisher s exact test.

7 Survival after revascularization in vascular surgery 539 extensive co-morbidities. Since the benefit from preoperative CR is not statistically measurable in the immediate perioperative period, CR based on preoperative non-invasive testing may probably be safely postponed to the postoperative period in selected patients. Such a strategy in specific patients may prevent unnecessary delays in surgery, will avoid the need to stop the anti-platelet therapy currently given following coronary stenting, but may retain the long-term survival benefit of CR. Further studies, aimed particularly at the intermediate-risk patients are needed to validate these results in the context of current medical and interventional therapies. Supplementary material Supplementary material is available at European Heart Journal online. Conflict of interest: none declared. References 1. Farkouh ME, Rihal CS, Gersh BJ, Rooke TW, Hallett JW Jr, O Fallon WM, Ballard DJ. Influence of coronary heart disease on morbidity and mortality after lower extremity revascularization surgery: a population-based study in Olmsted County, Minnesota. J Am Coll Cardiol 1994;24: Biancari F, Leo E, Ylonen K, Vaarala MH, Rainio P, Juvonen T. Value of the Glasgow Aneurysm Score in predicting the immediate and long-term outcome after elective open repair of infrarenal abdominal aortic aneurysm. Br J Surg 2003;90: Eagle KA, Rihal CS, Mickel MC, Holmes DR, Foster ED, Gersh BJ; For the CASS investigators and University of Michigan Heart Care program. Cardiac risk of non-cardiac surgery: influence of coronary disease and type of surgery in 3368 operations. Circulation 1997;96: Criqui MH, Langer RD, Fronek A Feigelson HS, Klauber MR, McCann TJ, Browner D. Mortality over a period of 10 years in patients with peripheral arterial disease. N Engl J Med 1992;326: Paul SD, Eagle KA, Kuntz KM, Young JR, Hertzer NR. Concordance of preoperative clinical risk with angiographic severity of coronary artery disease in patients undergoing vascular surgery. Circulation 1996;94: Eagle KA, Berger PB, Calkins H, Chaitman BR, Ewy GA, Fleischmann KE, Fleisher LA, Froehlich JB, Gusberg RJ, Leppo JA, Ryan T, Schlant RC, Winters WL Jr, Gibbons RJ, Antman EM, Alpert JS, Faxon DP, Fuster V, Gregoratos G, Jacobs AK, Hiratzka LF, Russell RO, Smith SC Jr. Guideline update for perioperative cardiovascular evaluation for noncardiac surgery. J Am Coll Cardiol. 2002;39: Bursi F, Babuin L, Barbieri A, Politi L, Zennaro M, Grimaldi T, Rumolo A, Gargiulo M, Stella A, Modena MG, Jaffe AS. Vascular surgery patients: perioperative and long-term risk according to the ACC/AHA guidelines, the additive role of post-operative troponin elevation. Eur Heart J 2005;26: McFalls EO, Ward HB, Moritz TE, Goldman S, Krupski WC, Littooy F, Pierpont G, Santilli S, Rapp J, Hattler B, Shunk K, Jaenicke C, Thottapurathu L, Ellis N, Reda DJ, Henderson WG. Coronary-artery revascularization before elective major vascular surgery. N Engl J Med 2004;351: Mascucci M, Eagle KA. Coronary revascularization before non-cardiac surgery. N Engl J Med 2004;351: Wesorick DH, Eagle KA. The preoperative cardiovascular evaluation of the intermediate-risk patient: new data, changing strategies. Am J Med 2005;118:1413.e e Lee TH, Marcantonio ER, Mangione CM, Thomas EJ, Polanczyk CA, Cook EF, Sugarbaker DJ, Donaldson MC, Poss R, Ho KK, Ludwig LE, Pedan A, Goldman L. Derivation and prospective validation of a simple index for prediction of cardiac risk of major non-cardiac surgery. Circulation 1999;100: Landesberg G, Mosseri M, Wolf YG, Bocher M, Basevitch A, Rudis E, Izhar U, Anner H, Weissman C, Berlatzky Y. Preoperative thallium scanning, selective coronary revascularization, and long-term survival after major vascular surgery. Circulation 2003;108: Kitsiou AN, Srinivasan G, Quyyumi AA, Summers RM, Bacharach SL, Dilsizian V. Stress-induced reversible and mild-to-moderate irreversible thallium defects: are they equally accurate for predicting recovery of regional left ventricular function after revascularization? Circulation 1998;98: Sokolow M, Lyon TP. The ventricular complex in left ventricular hypertrophy as observed by unipolar precordial and limb leads. Am Heart J 1960;21: Landesberg G, Einav S, Christopherson R, Beattie C, Berlatzky Y, Rosenfeld B, Eidelman LA, Norris E, Anner H, Mosseri M, Cotev S, Luria MH. Perioperative ischemia and cardiac complications in major vascular surgery: importance of the preoperative 12-lead ECG. J Vasc Surg 1997;26: Hertzer NR, Mascha EJ. A personal experience with factors influencing survival after elective open repair of infra-renal aortic aneurysms. J Vasc Surg 2005;42: Back MB, Leo F, Cuthbertson D, Johnson BL, Shames ML, Bandyk DF. Longterm survival after vascular surgery: specific influence of cardiac factors and implications for preoperative evaluation. J Vasc Surg 2004; 40: Ashley EA, Raxwal VK, Froelicher VF. The prevalence and prognostic significance of electrocardiographic abnormalities. Curr Probl Cardiol 2000; 25: Kertai MD, Bogar L, Gal J, Poldermans D. Pre-operative coronary revascularization: an optimal therapy for high-risk vascular surgery patients? Acta Anaesthesiol Scand 2006;50: Landesberg G, Mosseri M, Fleisher LA. Coronary artery revascularization prior to major vascular surgery. NEnglJMed2005;352: Moscucci M, Eagle KA. Coronary revascularization before non-cardiac surgery. N Engl J Med 2004;351: Breiman L. The little bootstrap and other methods for dimensionality selection in regression: X-fixed prediction error. J Am Stat Assoc 1992;87: Daviglus ML, Liao Y, Greenland P, Dyer AR, Liu K, Xie X, Huang CF, Prineas RJ, Stamler J. Association of non-specific minor ST-T abnormalities with cardiovascular mortality: the Chicago Western Electric Study. JAMA 1999;281: Larsen CT, Dahlin J, Blackburn H, Scharling H, Appleyard M, Sigurd B, Schnohr P. Prevalence and prognosis of electrocardiographic left ventricular hypertrophy, ST segment depression and negative T-wave; the Copenhagen City Heart Study. Eur Heart J 2002;23: Poldermans D, Boersma E, Bax JJ, Thomson IR, van de Ven LL, Blankensteijn JD, Baars HF, Yo TI, Trocino G, Vigna C, Roelandt JR, van Urk H. The effect of bisoprolol on perioperative mortality and myocardial infarction in high-risk patients undergoing vascular surgery. Dutch Echocardiographic Cardiac Risk Evaluation Applying Stress Echocardiography Study Group. N Engl J Med 1999;341: Devereaux PJ, Yusuf S, Yang H, Choi PT, Guyatt GH. Are the recommendations to use perioperative b-blocker therapy in patients undergoing non-cardiac surgery based on reliable evidence? Can Med J 2004;171: Brady AR, Gibbs JS, Greenhalgh RM, Powell JT, Sydes MR; POBBLE Trial Investigators. Perioperative beta-blockade (POBBLE) for patients undergoing infrarenal vascular surgery: results of a randomized double-blind controlled trial. J Vasc Surg 2005;41: Juul AB, Wetterslev J, Kofoed-Enevoldsen A, Callesen T, Jensen G, Gluud C. Randomized, blinded trial on perioperative metoprolol versus placebo for diabetic patients undergoing non-cardiac surgery. Circulation 2005; 111: