Abstract
The aim of this study was to evaluate the effectiveness of a preoperative standardized cardiac assessment in patients undergoing major vascular surgery. From January 2005 to December 2006, 1446 elective interventions for major vascular diseases (carotid stenosis, CS; abdominal aortic aneurysm, AAA; peripheral arterial obstructive disease, PAOD) were performed; 1090 out of these patients underwent preoperative diagnostic assessment on an outpatient basis. Thirty-day results in terms of cardiac mortality and morbidity rates were recorded. Patients suffered from a CS in 578 cases (53%), an AAA in 303 cases (27.8%) and a PAOD in 209 cases (19.2%). Four hundred thirty-two patients (39.6%) underwent further evaluation of cardiac functional capacity with non-invasive stress testing. Sixteen patients were successfully treated prior to vascular surgery. Thirty-day cardiac mortality and morbidity rates were 0.2% and 3.9%, respectively. A positive preoperative non-invasive stress testing did not affect 30-day cardiac outcomes. In conclusion, the use of an accurate preoperative cardiac assessment allowed us to obtain satisfactory perioperative results in patients undergoing major vascular surgery. Routine preoperative evaluation with non-invasive stress testing did not seem to improve perioperative cardiac results.
Introduction
Cardiac complications (myocardial infarction, acute congestive heart failure, pulmonary edema, fatal arrhythmias) represent the major cause of perioperative morbidity and mortality in patients undergoing major vascular surgery. 1–4 This is particularly related to the frequent association of systemic localizations of atherosclerosis (carotid bifurcation, abdominal aorta, peripheral arteries) with coronary artery disease (CAD). 5,6 It has also been demonstrated that history of CAD, congestive heart failure, cerebrovascular disease, diabetes mellitus and chronic renal failure increase perioperative cardiac mortality and morbidity rates. 7,8
In order to reduce perioperative cardiac complications in patients undergoing major vascular surgery, an accurate cardiac assessment, consisting of electrocardiogram (ECG), cardiologist consultation, echocardiography and, in selected cases, non-invasive stress testing, has been advocated by several authors. 9–11 Non-invasive stress testing, is widely used in order to assess the risk of cardiac complications in the perioperative period, even if its predictive power is controversial; 10 furthermore, other authors suggested the use of preoperative cardiac testing only in patients considered at high surgical risk, omitting it in intermediate-risk patients. 12
The demonstration of a poor cardiac functional capacity makes necessary the adoption of measures to reduce perioperative cardiac risk. Perioperative beta-blocker medical therapy is associated with a reduced risk of in-hospital death among patients undergoing major non-cardiac surgery. 13
At the moment, there are no strong data to support prophylactic surgical or endovascular myocardial revascularization; 9,11,14 hence, myocardial revascularization prior to vascular surgery should be reserved for those few patients with unstable angina or stable angina with significant left main or three-vessel CAD 9 (American College of Cardiology/American Heart Association [ACC/AHA] guidelines) or for patients with proven ischemic heart disease subject to high-risk surgery (European Society of Cardiology [ESC] guidelines). 11
The aim of this study was to evaluate the effectiveness of a preoperative standardized cardiac assessment in the reduction of perioperative cardiac morbidity and mortality rates in patients undergoing major vascular surgery in a single-center experience.
Methods
From January 2005 to December 2006, 1446 elective open and endovascular interventions for major vascular diseases (carotid stenosis, CS; abdominal aortic aneurysm, AAA; peripheral arterial obstructive disease, PAOD) were performed at our Institution; 1090 out of these patients underwent preoperative diagnostic and cardiac assessment on an outpatient basis. We excluded from our series all patients without a complete preoperative assessment (operation in emergency/urgency, direct admission at our Department, transfer from other Departments).
Data from all preoperative evaluations in an outpatient basis were prospectively collected in a dedicated database, which included demographic data, preoperative risk factors, and clinical and diagnostic assessment. All these patients underwent open or endovascular intervention and perioperative (<30 days) results in terms of overall and cardiac morbidity and mortality rates were recorded.
Preoperative diagnostic assessment
Regardless of the kind of vascular disease, all patients preoperatively underwent an extensive assessment of clinical history and a physical examination, a two-view chest X-ray, an ECG, laboratory tests including complete blood count, coagulative parameters and blood chemistries and a duplex ultrasound scanning of the indexed district. Moreover, all patients suffering from AAA or PAOD underwent duplex scanning of the supra-aortic vessels.
Patients assessed with CS preoperative underwent a computed tomography (CT) scan of cerebral parenchyma and CT angiography (CTA) of extracranial vessels. Degree of stenosis was determined by using the North American Symptomatic Carotid Endarterectomy Trial method.
Patients with AAA underwent CTA of the entire aorta to evaluate the presence of other aortic lesions and to determine the type of repair (assessment of criteria of feasibility for endovascular repair). In this subgroup, in patients with a history of chronic obstructive pulmonary disease or asthma, pulmonary functional capacity and response to bronchodilators were investigated and the presence of carbon dioxide retention through arterial blood gas analysis was evaluated.
In patients with PAOD, preoperative diagnostic evaluation was completed by CTA of the lower limbs, reserving the digital subtraction angiography in double projection in selected cases.
Finally, regardless of the response of ECG, all patients underwent cardiological consultation associated with an echocardiography with evaluation of left-ventricular ejection fraction (LVEF). Since most of the patients came from peripheral centers, the cardiologist recommended changes in medications, in order to obtain the best medical treatment in all patients, consisting of statins and antiplatelet agents in all patients and beta-blockers in selected cases. A resting heart rate of 60–65 beats/min was considered right.
Our series comprised patients undergoing carotid, aortic and peripheral surgery. On the basis of ACC/AHA and ESC guidelines, 9,11 patients undergoing open or endovascular carotid surgery should be considered at intermediate cardiac risk (between 1% and 5%) and patients undergoing open aortic and peripheral surgery should be considered at high cardiac risk (>5%); furthermore, patients undergoing endovascular aortic or peripheral surgery should be considered at intermediate cardiac risk (between 1% and 5%). The cardiologist performed the consultation before the surgeon had decided what was the kind of treatment for the patient (open or endovascular); hence, all patients subject to aortic or peripheral surgery have been considered at high surgical risk and the decision to make further preoperative tests was taken by the cardiologist on the basis of the results of history, physical examination, ECG and echocardiography.
In particular, the cardiologist suggested non-invasive stress testing in the presence of past or recent ischemic heart disease, compensated or prior heart failure, past or recent cerebrovascular disease, diabetes mellitus, chronic renal failure (history); alterations at cardiac auscultation (physical examination); significant arrhythmias, abnormal Q waves (ECG anomalies); and severe valvular diseases, abnormal areas of motility, LVEF less than 40% (echocardiographic anomalies). All non-invasive stress tests have been conducted after withdrawal of beta-blockers.
In patients with a negative cardiac stress test, surgery was performed without any other investigation. Patients with a positive stress test (a peak heart rate less than 85% of the maximum predicted at dobutamine stress echocardiography or reversible defects that involved more than 20% of myocardial segments at the radionuclide myocardial perfusion imaging) underwent new consultation and the cardiologist, also on the basis of the surgeon's decision, proposed higher levels of postoperative care in the intensive care unit (ICU) and further changes in medications with the adjunct of beta-blockers or suggested performing classical coronary angiography with possible therapeutic procedures (percutaneous transluminal coronary angioplasty [PTCA] or coronary artery bypass graft [CABG]). Coronary revascularization prior to vascular surgery was performed in patients subject to high-risk surgery, who had unstable angina or stable angina with three-vessel disease or a significant stenosis of the left main coronary artery.
In these patients the vascular intervention was consequently delayed. If PTCA was performed, patients received a single antiplatelet therapy (aspirin) for four weeks. If a stent was inserted, a dual antiplatelet therapy was administered (aspirin and clopidogrel) for four weeks (bare-metal coronary stents) or six months (drug-eluting coronary stents).
Definition of parameters
Our database included demographic data, preoperative risk factors, and clinical and diagnostic assessment. We used some criteria to define these parameters, as shown in Table 1.
Parameters used to elaborate the perioperative cardiac risk
ECG, electrocardiogram; CABG, coronary artery bypass graft; PTCA, percutaneous transluminal coronary angioplasty
Statistical analysis
Statistical analysis was performed by using SPSS 15.0 for Windows (SPSS Inc, Chicago, IL, USA). General characteristics, cardiovascular risk factors, results of preoperative diagnostic assessment and different surgical approaches for perioperative (<30 days) cardiac mortality and morbidity rates were analyzed and compared by χ 2 test or Fisher's exact test when necessary. Statistical significance was defined at the P < 0.05 level.
Results
Study group
Patients were predominantly men (831, 76.2%), with a mean age of 72.4 years (range 32–92). They suffered from CS in 578 cases (53%), AAA in 303 cases (27.8%) and PAOD in 209 cases (19.2%). Demographic data, preoperative risk factors and clinical assessment are shown in Table 2.
Demographic data, preoperative risk factors and clinical assessment
CABG, coronary artery bypass graft; PTCA, percutaneous transluminal coronary angioplasty; CS, carotid stenosis; PAOD, peripheral arterial obstructive disease; AAA, abdominal aortic aneurysm
Preoperative cardiac assessment
According to our decisional algorithm, 658 patients (60.4%) without anomalies at ECG, cardiological consultation and echocardiography underwent surgical vascular intervention without any other diagnostic assessments. The remaining 432 patients (39.6%) underwent further evaluation of cardiac functional capacity with a dobutamine stress echocardiography in 272 cases (63%), a radionuclide myocardial perfusion imaging in 140 cases (32.4%) and a maximal treadmill exercise test according to the Bruce protocol in 20 patients (4.6%).
Of these 432 patients, 326 (75.5%) had a negative test, while the remaining 106 patients (24.5%) had a positive one. In 73 (68.9%) out of these 106 patients, perioperative cardiological risk was judged acceptable by the consultant and beta-blocker medical therapy and postoperative ICU stay were suggested. In the remaining 33 patients (31.1%), a coronary angiography was performed; no significant CAD was diagnosed in 17 cases and the patients were operated on, while the remaining 16 patients had significant coronary artery lesions and were successfully treated (15 PTCA and 1 CABG) prior to vascular surgery (Table 3).
Preoperative cardiac assessment and operative management
CABG, coronary artery bypass graft; PTCA, percutaneous transluminal coronary angioplasty; CS, carotid stenosis; PAOD, peripheral arterial obstructive disease; AAA, abdominal aortic aneurysm; ICU, intensive care unit
Operative management
Concerning the CS group, a standard carotid endarterectomy according to the technique of early distal control of the internal carotid artery 15 with extensive use of patch and selective use of shunt was performed in 552/578 cases (95.5%), while the remaining 26 patients (4.5%) underwent carotid artery stenting (CAS). For the AAA group, traditional open repair with a standard transperitoneal approach was carried out in 180/303 patients (59.4%), while in the other 123 cases (40.6%) AAA was excluded through the endovascular implantation of a stent-graft. Finally, in PAOD patients the endovascular approach was performed in 101/209 cases (48.3%).
Overall, 125 patients (11.5%) required a higher level of postoperative monitoring in ICU.
A beta-blocker therapy was continued in all patients who received it prior to the cardiological assessment. Furthermore, with regard to cardiac management, the surveillance for perioperative myocardial infarction was obtained with the measurement of troponin T in the recovery room after operation and on the next two postoperative mornings; if indicated, the troponin T monitoring was continued. The diagnosis of myocardial infarction was determined by troponin T values combined with either changes on ECG or symptoms referred by the patient.
Early results (<30 days)
In our series, two open-treated patients died in the early postoperative period for an acute congestive heart failure with pulmonary edema. As a consequence, 30-day cardiac mortality rate was 0.2% (2/1090 patients). Analyzing the three categories of patients, there were no differences in terms of 30-day cardiac mortality risk (Table 4).
Early results (<30 days)
CS, carotid stenosis; PAOD, peripheral arterial obstructive disease; AAA, abdominal aortic aneurysm; NS, non-significant
Between AAA group and the other two groups
At univariate analysis, in the AAA group, valvular heart disease and history of previous congestive heart failure represented factors significantly affecting 30-day cardiac mortality (Table 5). Multivariate analysis did not confirm the significance for any of the examined parameters. Moreover, at 30 days, 81 patients (7.4%) had a major complication, requiring a prolonged hospital stay or a re-intervention: 45 had at least one cardiac complication (22 myocardial infarctions requiring a PTCA in 11 cases, 16 episodes of atrial fibrillation and 7 of acute congestive heart failure) with a 30-day cardiac morbidity rate of 4.1% (Table 4). In the 22 patients with a postoperative myocardial infarction, none of the examined factors showed an increased risk at uni- and multivariate analysis.
AAA group: univariate analysis for cardiac mortality at 30 days
NS, non-significant; AAA, abdominal aortic aneurysm
Furthermore, the AAA group showed an increased risk of overall and cardiac 30-day morbidity rates with respect to other patients. In particular, univariate analysis showed that at ages older than 80 years, chronic renal failure and valvular heart disease significantly affect 30-day cardiac morbidity (Table 6); chronic renal failure and valvular heart disease maintained their significance also at multivariate analysis (P = 0.01, 95% CI: 1.3–12.1; P = 0.002, 95% CI: 1.8–15.5, respectively).
AAA group: univariate analysis for cardiac morbidity at 30 days
NS, non-significant; AAA, abdominal aortic aneurysm
Discussion
Historically, many attempts to obtain a precise cardiac risk-score following major vascular surgery have been made; 7,16, 17 in these studies, history of CAD, congestive heart failure, cerebrovascular disease, diabetes mellitus and chronic renal failure increased the risk of perioperative cardiac adverse events. Also in our patients undergoing aortic surgery, several preoperative risk factors affected cardiac mortality and morbidity (valvular heart disease, history of previous congestive heart failure, age older than 80 years and chronic renal failure).
In 1996, the ACC/AHA published the first clinical guidelines for preoperative cardiac risk stratification in patients undergoing non-cardiac surgery. These guidelines were recently updated in 2007. 9 Other societies published their own guidelines 11,18,19 and numerous discrepancies exist between them. 20
In particular, comparing the most updated guidelines (ACC/AHA 2007 and ESC 2009), 9,11 there were no substantial differences in the stratification of the risk in major vascular surgery; in fact, both considered at high surgical risk all patients undergoing open aortic and peripheral surgery and at intermediate surgical risk all patients subject to carotid surgery (open or endovascular) or to endovascular aortic/peripheral procedures. Our approach did not substantially differ from these classifications, but our patients underwent cardiological consultation before the surgeons and the patients decided the best type of treatment and so we considered all patients suffering from aortic/peripheral diseases as high surgical risk patients. This represents a real problem in clinical practice, which is often omitted in the official guidelines.
Furthermore, ACC/AHA 2007 and ESC 2009 guidelines recommended the selective use of supplemental preoperative non-invasive cardiac testing in order to provide an objective measure of cardiac functional capacity, to identify the presence of significant myocardial ischemia and to evaluate the risk of perioperative cardiac adverse outcomes; however, some differences exist. In fact, American guidelines recommended the use of non-invasive stress testing in patients with active cardiac conditions and in patients subject to all vascular procedures with three or more clinical factors, and suggested its utilization in patients subject to all vascular procedures with at least one clinical factor. On the other hand, European guidelines recommended a supplemental stress testing in patients subject to high-risk surgery (open aortic and peripheral diseases) with three or more clinical factors and suggested its utilization in patients considered at intermediate risk (carotid procedures). Our data were prospectively collected from January 2005 to December 2006; hence, we could not follow any of these guidelines. However, our approach was quite similar to the recommendations and suggestions reported by ACC/AHA guidelines.
A particular key point is the need to perform a supplemental preoperative cardiac testing in patients subject to carotid surgery (considered in both guidelines at intermediate risk); the DECREASE (Dutch Echographic Cardiac Risk Evaluation Applying Stress Echo) II trial 12 demonstrated that cardiac testing could safely be omitted in intermediate-risk patients, provided that beta-blocker therapy aiming at tight heart rate control is prescribed. In our series, we performed 135 cardiac stress testings in the 578 patients subject to carotid surgery, and according to the conclusions of the DECREASE II trial, these further evaluations could have been avoided, saving time and economical resources. Furthermore, only 31/135 patients (23%) had a positive stress test and five patients (3.7%) underwent preoperative myocardial revascularization; in these five patients no perioperative cardiac complications occurred, but we cannot know what could have happened in these patients if left untreated. Hence, our experience supports the conclusions of this randomized trial.
Several authors 21–24 demonstrated that positive predictive value of stress testing is very poor (5–25%) and they concluded that its use in order to predict perioperative cardiac events is very limited. Our experience supports these conclusions; in our population study, a positive preoperative cardiac stress testing does not represent a risk factor for postoperative cardiac events and its positive predictive power is very low.
As a consequence, a positive preoperative non-invasive cardiac stress testing seems to support the value of appropriate perioperative medical therapies in order to lower the cardiac risk rather than extensive use of myocardial revascularization. The guidelines 9,11 suggested that beta-blockers should be given to patients undergoing major vascular surgery, but the most recent POISE trial 25 minimized the benefits resulting in the assumption of beta-blocker medical therapy in the perioperative period. In our experience, beta-blocker medical therapy and a high level of postoperative care in ICU is the most used approach in patients with poor cardiac functional capacity undergoing endovascular treatment.
Controversial data exist also about benefits of myocardial revascularization in reducing perioperative cardiac adverse events in patients undergoing major vascular surgery; 9,11,12 ACC/AHA 2007 guidelines stated that coronary revascularization before non-cardiac surgery is useful in patients with stable angina and significant left main coronary artery stenosis or a three-vessel disease or in patients with unstable angina, while ESC 2009 guidelines recommended a prophylactic myocardial revascularization prior to high-risk surgery in patients with proven ischemic heart disease and not in patients at intermediate risk. In our series, about one-third of patients (33/106, 31.1%) with positive stress testing underwent coronary angiography; one of the treated patients had a postoperative myocardial infarction, while in the untreated group no cardiac complications occurred. So, preoperative coronary revascularization did not seem to be useful for reducing perioperative cardiac risk.
In the AAA subgroup, endovascular techniques allow the treatment of patients whose co-morbidities make conventional open repair difficult or high risk. For this reason, patients with a positive test could be treated with endovascular treatment when anatomically feasible. In our experience, endovascular treatment reduced cardiac morbidity rates compared with open surgical repair, even if these differences did not reach the statistical significance. However, this should encourage many surgeons to stress the indications for endovascular treatment.
Finally, this study has some limits: our results were carried out on a small monocentric sample size and we analyzed a heterogeneous population including patients undergoing vascular interventions at different levels of perioperative cardiac risk; in fact, patients undergoing abdominal aortic surgery showed higher cardiac morbidity and mortality rates than the other two groups. Furthermore, our overall 30-day outcomes in terms of mortality and major morbidity seemed to be better than others reported in previously published series. Maybe our good 30-day outcomes suffered from a selection bias; in fact, in our series we enrolled only the patients completely assessed in the preoperative period on an outpatient basis and we excluded all patients without a complete preoperative assessment. The strict respect of the worldwide accepted indications and the close collaboration with cardiologists and internists in the selection and the care of our patients may have contributed to our very good results in this subset of patients as well.
So, in conclusion, in our experience the use of an accurate preoperative cardiac assessment allowed us to obtain satisfactory perioperative results in patients undergoing major vascular surgery. Patients with AAAs represented a subgroup with higher perioperative cardiac morbidity rates. Routine preoperative evaluation with non-invasive stress testing did not seem to improve perioperative cardiac results in any subgroup treated for vascular diseases. However, the selective use of stress testing made it possible to identify a subgroup of higher risk patients who could benefit from coronary angiography and possible treatment of their CAD.
Footnotes
Acknowledgments
Financial disclosure of authors and reviewers: none reported.