Sustained Prognostic Value of Dobutamine Stress Echocardiography for Late Cardiac Events After Major Noncardiac Vascular Surgery
Background Late cardiac events after major noncardiac vascular surgery are an important cause of morbidity and mortality. We studied the prognostic value of preoperative dobutamine stress echocardiography, relative to clinical risk assessment, in predicting late cardiac events.
Methods and Results Three hundred sixteen patients undergoing major vascular surgery were studied. All patients underwent clinical evaluation for the presence of cardiac risk factors (smoking, hypertension, angina, diabetes, history of heart failure, previous infarction, and age >70 years) and dobutamine stress echocardiography. Left ventricular wall motion was evaluated at rest, and the extent and severity of stress-induced new wall motion abnormalities were quantified. The heart rate threshold at which new wall motion abnormalities occurred was noted. Patients were followed perioperatively and for 19±11 months postoperatively, and the occurrence of cardiac events was noted. Univariate and multivariate Cox proportional hazards regression models were used to identify predictors of late cardiac events. Thirty-two cardiac events occurred (11 cardiac deaths, 11 nonfatal myocardial infarctions, and 10 incidents of unstable angina). By multivariate regression analysis, the occurrence of extensive (three or more segments) or limited (one or two segments) stress-induced new wall motion abnormalities and previous infarction independently predicted late cardiac events, elevating the risk by 6.5-, 2.9-, and 3.8-fold, respectively. The severity of ischemia during stress and the heart rate threshold for ischemia were not independently predictive.
Conclusions Patients with a history of myocardial infarction or stress-induced ischemia have a high risk of fatal and nonfatal cardiac events after vascular surgery. Patients with both a history of infarction and extensive stress-induced ischemia are at especially high risk and deserve intensive management.
Coronary artery disease is a major cause of late morbidity and mortality in survivors of major vascular surgery. For example, Krupski et al1 found that late cardiac events occurred in 19% of 129 patients followed for 2 years after vascular surgery. This high incidence reflects the frequency and severity of underlying coronary artery disease. Hertzer et al2 obtained coronary angiograms in 1000 vascular surgery candidates and found severe correctable coronary artery disease in 36% of patients with aortic aneurysm and in 28% of those with lower extremity ischemia. Therefore, the preoperative evaluation of vascular surgery candidates should include an assessment of the risk of late postoperative cardiac events, because these will determine the likelihood that individual patients will survive to enjoy the benefits of successful surgery.
Several clinical and laboratory variables have been associated with an increased risk of late cardiac events after major surgery. Clinical predictors include a preoperative history of coronary artery disease or congestive heart failure.2 3 Of particular interest is the finding by Mangano et al4 that patients who survived a postoperative myocardial infarction or an episode of unstable angina had a 20-fold increase in the odds of a late cardiac event. Noninvasive laboratory indicators of late cardiac risk include left ventricular dilatation and thallium redistribution during dipyridamole-thallium myocardial perfusion scintigraphy,4 5 6 7 8 impaired left ventricular function on radionuclide ventriculography,9 and ischemic ST-segment changes on perioperative ambulatory ECG monitoring.10 11
We have found dobutamine stress echocardiography to be an extremely useful tool for the assessment of perioperative cardiac risk in patients undergoing major vascular surgery.12 13 Therefore, we hypothesized that this test would also predict the risk of late cardiac events after major vascular surgery. Dobutamine and atropine unfavorably alter myocardial oxygen balance and may induce ischemia in patients with coronary artery disease.14 15 Pharmacologically induced myocardial ischemia can be detected echocardiographically as new regional wall motion abnormalities. The test is safe, inexpensive, widely available, and applicable in patients who cannot exercise. It also provides information about resting left ventricular function. In this report, we summarize data from 316 patients who were followed for ≤36 months after major vascular surgery.
The study protocol was approved by the Hospital Ethics Committee, and all patients gave informed consent. Between 1991 and 1994, we studied 323 consecutive patients. β-Blockers were used by 72 patients (23%), and calcium antagonists were used by 89 (28%). Five patients had a fatal perioperative myocardial infarction, and 2 additional patients were lost to follow-up. The remaining 316 survivors form the basis of this report. These included 264 men and 52 women with an age range of 22 to 90 years (mean age, 67 years). They underwent aortic aneurysm resection (n=111), aortofemoral bypass (n=111), or infrainguinal arterial reconstruction (n=94). Preoperative clinical examination included a history, physical examination, and 12-lead ECG. We specifically searched for clinical cardiac risk factors: smoking, hypertension, angina, diabetes mellitus, history of congestive heart failure, age >70 years, and previous myocardial infarction (Q wave on ECG). The occurrence of a nonfatal perioperative cardiac event (ie, unstable angina or myocardial infarction) was also noted.
Dobutamine Stress Echocardiography
Before surgery, each subject underwent dobutamine stress echocardiography. A resting two-dimensional transthoracic echocardiographic examination, including standard apical and parasternal views of the left ventricle, was performed and recorded on videotape. A resting 12-lead ECG was also obtained. Dobutamine was administered intravenously with an infusion pump, starting at 10 μg·kg−1·min−1 for 3 minutes (stage I). The infusion rate was increased by 10 μg·kg−1·min−1 every 3 minutes to a maximum of 40 μg·kg−1·min−1 (stage IV) and continued for 6 minutes. If signs and symptoms of ischemia were absent during stage I, and the patient did not reach a target heart rate ([220−age]×0.85 in men and [200−age]×0.85 in women), atropine was administered intravenously in 0.25-mg increments to a maximum of 1 mg while the dobutamine infusion was continued. During the test, the 12-lead ECG was monitored continuously and recorded each minute. Blood pressure was measured noninvasively (Accutorr A1, Datascope Corp) at rest and at each stage of the protocol. The two-dimensional echocardiogram was monitored continuously and recorded on videotape during the last minute of each stage. A quadscreen video display, for side-by-side comparison of resting and stress images, was used during the last 150 examinations (Vingmed CFM 800, Diasonics Sonatron). The criteria for stopping the test included a decline in systolic blood pressure of >40 mm Hg from the resting value or a systolic blood pressure of <100 mm Hg, significant cardiac arrhythmias, severe chest pain, horizontal or downsloping ECG ST depression of ≥0.2 mV measured 80 ms after the J point, ST-segment elevation of ≥0.2 mV in the absence of Q waves, and severe new echocardiographic wall motion abnormalities in multiple locations.
Off-line assessment of echocardiographic images was performed by two investigators who were aware of the doses of dobutamine and atropine that had been used but were blinded to clinical information. The left ventricular wall was divided into 16 segments,16 and wall motion was subjectively scored on a four-point scale: 1 indicates normal; 2, hypokinetic; 3, akinetic; and 4, dyskinetic. The test was considered positive when wall motion in any segment deteriorated one or more grades. Agreement between the two investigators was required for a test to be designated as positive. In the event of disagreement, a third investigator resolved the dispute.
For each patient, a wall motion score index (total score divided by the number of assessable segments) was calculated at rest and during peak stress. The severity of ischemia was defined as the difference between these two values. The extent of ischemia was defined as the number of segments exhibiting deteriorating wall motion during stress. The ischemic threshold was defined as the heart rate at which new echocardiographic wall motion abnormalities occurred divided by the maximal age-related heart rate ([220−age] in men and [200−age] in women).
All patients left the hospital within 1 month after surgery. Thereafter, they visited our hospital out-patient clinic at regular intervals for 19±11 months (mean±SD; range, 1 to 36 months). If a patient did not appear for a scheduled visit, his or her family physician was contacted. The physician performing the follow-up was blinded to the results of preoperative dobutamine-atropine stress echocardiography because the test was still considered investigational at that time. Therefore, postoperative management was not influenced by preoperative stress testing.
The occurrence of all late cardiac events (cardiac death, nonfatal myocardial infarction, and myocardial revascularization) was documented. “Hard cardiac events” were defined as nonfatal myocardial infarction or cardiac death. The occurrence of stroke and noncardiac death was also noted. Cardiac death was determined through a review of the clinical presentation, including cardiac isoenzymes, 12-lead ECG, and autopsy results when these were available. Nonfatal myocardial infarction was defined as elevated cardiac isoenzymes (creatinine kinase >110 U/L with MB isoenzymes >10%) and ECG changes (new Q waves >0.03 s in duration). Myocardial revascularization included coronary artery bypass graft surgery and percutaneous transluminal coronary angioplasty. When more than one cardiac event occurred in the same patient, the most serious event was considered for analysis (ie, death > myocardial infarction > revascularization).
Univariate and multivariate Cox proportional hazards regression models were used to identify independent predictors of late cardiac events. The risk associated with a given variable was expressed by a hazard ratio (HR) with corresponding 95% confidence interval (CI). Variables were considered to be significant if the null hypothesis of no contribution could be rejected at the .05 probability level. In addition, likelihood ratios were computed to describe the predictive usefulness of significant risk factors. Kaplan-Meier life-table analysis of cardiac events was used to assess the prognostic importance of stress-induced new wall motion abnormalities with respect to event-free survival. Receiver-operator characteristic (ROC) curves were used to determine the “optimal” cutoff point for prediction of late events with respect to (1) the number of segments with resting wall motion abnormalities, (2) the number of ischemic segments at peak stress, (3) the ischemic heart rate threshold, and (4) the severity of ischemia at peak stress. The best cutoff was defined as the point with the highest sum of sensitivity and specificity.
Clinical Profile and Resting Echocardiogram
One hundred seventy-two patients (54%) were >70 years of age; smoking was present in 104 (33%), hypertension in 64 (20%), typical angina in 57 (18%), and previous myocardial infarction in 92 (29%); and there was a history of heart failure in 13 (4%) and of diabetes mellitus in 32 (10%). Twenty-two patients (7%) experienced a nonfatal perioperative cardiac event.
Resting wall motion abnormalities were present in 102 patients (32%). In 48 patients (15%), more than three segments were abnormal at rest.
Heart rate increased from 73±14 to 132±15 bpm (P=.0001). Systolic blood pressure increased during dobutamine infusion from 144±25 to 153±32 mm Hg (P=.001). Diastolic blood pressure decreased from 83±14 to 80±17 mm Hg at the maximum dose of dobutamine (P=.01).
Stress Test Results
During the stress test, 25 patients (8%) experienced typical chest pain, 74 (23%) experienced ST-segment changes, and 84 (25%) experienced new wall motion abnormalities. The mean dobutamine dose was 36 μg·kg−1·min−1, and atropine was administrated to 111 patients (35%). The stress test was inconclusive (prematurely stopped because of side effects without signs or symptoms of myocardial ischemia) in 4 patients. Two of these patients developed intolerable chills, 1 developed marked hypertension (240193/ mm Hg), and 1 developed severe hypotension.
There were no fatal complications. Other than chest pain, hypotension and cardiac arrhythmias were the most important side effects of dobutamine stress testing. Cardiac arrhythmias occurred in 7 patients (2%), and hypotension occurred in 12 (4%). Cardiac arrhythmias consisted of ventricular fibrillation in 1 patient, nonsustained ventricular tachycardia in 3, and paroxysmal atrial fibrillation in 3. The patient who developed ventricular fibrillation during the recovery phase of the test was successfully resuscitated with one countershock. No ECG or enzymatic evidence of a new myocardial infarction was subsequently apparent. Hypotension was usually mild (decline in systolic blood pressure of >20 mm Hg compared with baseline). Stress test results were available in all of these patients because cardiac arrhythmias occurred during the recovery phase of the test and patients with mild hypotension were able to continue the test.
Thirty-two cardiac events occurred during follow-up in 316 patients (10.1%). There were 11 cardiac deaths and 11 nonfatal infarctions, and coronary revascularization was performed in 10 patients. Eight patients had a stroke, and there were 7 noncardiac deaths (mostly secondary to infected prosthetic grafts). Among 22 patients who experienced a nonfatal perioperative cardiac event, 11 late cardiac events occurred during follow-up. Two of these 11 events were fatal. There were no late cardiac events in the 4 patients in whom the stress test was prematurely discontinued because of side effects.
Prediction of Late Cardiac Events
ROC-curve analysis defined the following optimal cutoff point for predicting of late events: resting wall motion abnormalities, two or more segments; extent of ischemia, three or more segments; heart rate threshold, echocardiographically detected new wall motion abnormalities at a heart rate of <70% of the maximal age- and sex-related value; and severity of ischemia (wall motion score at peak stress minus wall motion score at rest divided by the number of segments) >0.14 wall motion score units.
The univariate predictors of late cardiac events (all cardiac events and hard cardiac events) are presented in Tables 1 and 2⇓⇓. A history of angina, myocardial infarction, and the occurrence of a nonfatal perioperative cardiac event were significant predictors of both all and hard late cardiac events. Diabetes mellitus and a history of heart failure were significant predictors of all cardiac events only. Stress test results were also predictive. Left ventricular dysfunction at rest (two or more segments), extensive new wall motion abnormalities during stress (three or more segments), any new wall motion abnormalities during stress, the heart rate threshold for ischemia (>70% of the maximal age- and sex-related heart rate), the severity of stress-induced ischemia and stress-induced angina, or ST-segment changes also predicted all late cardiac events. With the exception of ST-segment changes and severity of ischemia during stress, these test variables were also predictive for hard cardiac events. Stroke could not be predicted by clinical risk factors or stress test results.
Multivariate regression analysis was performed on all clinical risk factors and stress test results (Table 2⇑). Extensive echocardiographically detected ischemia during stress was the most powerful predictor of all as well as of hard late cardiac events. A previous myocardial infarction was also an independent predictor of fatal and nonfatal late cardiac events after surgery. Event-free survival curves for patients with a normal stress test, limited ischemia (one to two segments), and extensive ischemia (three or more segments) during stress are presented in Fig 1⇓. There was a significant decrease in event-free survival in patients with either limited or extensive new wall motion abnormalities.
The hazard ratios for the occurrence of late cardiac events are presented in Table 3⇓ and Fig 2⇓. Patients with no history of myocardial infarction and a negative test were at low risk for late cardiac events (HR, 1.0). Patients with either limited ischemia or a history of infarction were at relatively low risk (HR, 2.9 and 3.1, respectively). Patients with limited ischemia and a history of infarction or with extensive ischemia without previous infarction were at moderately high risk (HR, 8.8 and 10.3, respectively). Those with both extensive ischemia and a history of infarction were at extremely high risk (HR, 31.5).
Patients undergoing major vascular surgery represent a high-risk population with respect to the risk of both perioperative and late cardiac events.1 16 17 The decision to proceed with surgery should include a consideration of both of these risks. Ideally, a patient should not only survive the perioperative period intact but also live long enough to enjoy the benefits of successful surgery.18 Noninvasive preoperative risk stratification in patients scheduled for major vascular surgery may help in making appropriate decisions.
Pharmacological stress echocardiography is a promising tool in this context because it is widely available, generally applicable, and less expensive than perfusion scintigraphy.12 13 14 15 16 This study demonstrates that dobutamine stress echocardiography is the most powerful predictor of late cardiac events after major vascular surgery and is superior to simple clinical risk assessment. The negative prognostic value of extensive stress-induced ischemia (three or more segments) applies to both all and hard late cardiac events, increasing risk by 6.5- and 5.8-fold, respectively. Limited stress-induced ischemia increased the risk of all late events by 2.9-fold. The only independent preoperative clinical risk factor was a history of myocardial infarction. The presence of this risk factor increased the risk of all and hard late events by 3.8- and 2.8-fold, respectively.
A combination of semiquantitative stress echocardiography and the clinical history allows us to define the risk of all late cardiac events for patients in six categories (Table 3⇑ and Fig 2⇑). The absence of previous myocardial infarction combined with a negative stress test defines a large subset of patients (183 of 316, or 58%) with a very low rate of late events (HR, 1.0). Patients with either limited ischemia or a previous myocardial infarction were also at relatively low risk (HR, 2.9 to 3.1). Patients with both limited ischemia and a history of myocardial infarction were at relatively high risk (HR, 8.8), as were those with extensive ischemia without previous infarction (HR, 10.3). Patients with a history of infarction and extensive stress-induced ischemia were at extremely high risk (HR, 31.5).
This approach defines a relatively large subset of low- and relatively low-risk patients (247 of 316, or 78%) who need not undergo intensive postoperative follow-up. In patients with a relatively high risk of late cardiac events, meticulous follow-up is clearly indicated. In very high-risk patients with extensive stress-induced ischemia and a history of myocardial infarction, an aggressive anti-ischemic approach might prolong the period of event-free survival after successful surgery. However, this hypothesis remains untested, and the optimal anti-ischemic approach remains undefined.
We previously recommended preoperative dobutamine stress echocardiography for the assessment of perioperative cardiac risk in all patients with inadequate physical exercise capacity or any of the clinical risk factors defined previously12 19 (ie, age >70 years, Q wave on ECG, angina pectoris, diabetes mellitus, or a history of ventricular tachyarrhythmias. This recommendation also appears to be appropriate for preoperative assessment of the risk of late cardiac events. In our study, only 6 patients with an adequate physical exercise capacity and no risk factors as defined by Eagle et al had a positive stress test. None of these 6 patients experienced a late cardiac event. We recommend that dobutamine stress echocardiography be performed preoperatively in all patients with one or more of the risk factors of Eagle et al or an inadequate physical exercise capacity to define the risk of late cardiac events after surgery.
Previous studies in nonsurgical patients have demonstrated a relation between stress-induced wall motion abnormalities and the risk of subsequent cardiac events in patients with known or suspected coronary disease.20 21 However, patients presenting for major vascular surgery compose a very different population in whom the applicability of previous results has been undefined. The results of the present study in general confirm previous work in medical patients but provide unique insight into the role of pharmacological stress echocardiography for the assessment of late cardiac risk in patients presenting for major vascular surgery. The semiquantitative analysis of stress test results combined with a single clinical risk factor provides precise definition of late risk. By studying consecutive patients who underwent surgery, we avoided a referral bias. By blinding the attending physicians to test results, we avoided alterations in management based on presumptions about risk. For example, coronary arteriography and subsequent prophylactic myocardial revascularization were in large part avoided. These design features and the relative sample size increase the reliability of our conclusions.
Multivariate analysis indicated that the extent of ischemia was an independent predictor of late cardiac events. However, neither the severity of ischemia nor the heart rate at which ischemia occurred provided additional prognostic information. These results are in partial agreement with those of a study by Lette et al18 in which quantitative dipyridamole-thallium scintigraphy was used. The authors found that both the extent and the severity of reversible perfusion defects were significant predictors of late cardiac events. The difference between studies regarding the predictive power of the severity of ischemia is unexplained, but methodological differences are presumably involved.
In agreement with the studies of Mangano et al4 and Yeager et al,17 we found that patients who survived surgery but experienced a postoperative myocardial infarction or episode of unstable angina had a nearly 7-fold increase in the risk of a late cardiac event. In the study of Mangano et al, the negative prognostic value of a postoperative ischemic cardiac event was independent of the results of perioperative Holter monitoring. However, in the present study, the occurrence of a perioperative nonfatal infarction or unstable angina was not an independent predictor of adverse late outcome when multivariate analysis included the results of preoperative stress echocardiography. These differing results suggest that the prognostic power of stress echocardiography may be superior to that of perioperative Holter monitoring.
Preoperative dobutamine stress echocardiography is an important new technique for defining the risk of late cardiac events in patients undergoing major vascular surgery. The test should be performed preoperatively in all patients with inadequate exercise capacity or any of the clinical risk factors defined by Eagle et al.19 Extensive stress-induced ischemia is the most powerful predictor of late cardiac events, but a history of myocardial infarction also had independent negative prognostic value. These conclusions remain valid when only hard ischemic events such as cardiac death and myocardial infarction are considered. Preoperative semiquantitative stress echocardiography combined with information about previous myocardial infarction optimizes assessment of late risk. Patients presenting for vascular surgery with a history of myocardial infarction and extensive stress-induced ischemia are at extremely high risk. This information should be carefully considered when deciding on the advisability of surgery, appropriate anti-ischemic therapy, and the need for intensive follow-up.
- Received April 11, 1996.
- Revision received June 25, 1996.
- Accepted July 5, 1996.
- Copyright © 1997 by American Heart Association
Brown KA. Prognostic value of thallium-201 myocardial perfusion imaging: a diagnostic tool comes of age. Circulation. 1991:83:363-381.
Hendel RC, Chen MH, L'Italien GJ, Newell JB, Paul SD, Eagle KA, Leppo JA. Sex differences in perioperative and long-term cardiac event-free survival in vascular surgery patients: an analysis of clinical and scintigraphic variables. Circulation. 1995;91:1044-1051.
Mangano DT, London MJ, Tubau JF, Browner JS, Hollenberg M, Krupski W, Layug EL, Massie B. Dipyridamole thallium-201 scintigraphy as a preoperative screening test: a reexamination of its predictive potential. Circulation. 1991;84:493-502.
Poldermans D, Arnese M, Fioretti PM, Salustri A, Boersma E, Thomson IR, Roelandt JRTC, van Urk H. Improved cardiac risk stratification in major vascular surgery with dobutamine-atropine stress echocardiography. J Am Coll Cardiol. 1995,26:648-653.
Poldermans D, Fioretti PM, Forster T, Thomson IR, Boersma E, El-Said M, du Bois NAJJ, Roelandt JRTC, van Urk H. Dobutamine stress echocardiography for the assessment of perioperative cardiac risk in patients undergoing major noncardiac vascular surgery. Circulation. 1993;87:1506-1512.
Marcovitz PA, Armstrong WF. Accuracy of dobutamine stress echocardiography in detecting coronary artery disease. Am J Cardiol. 1992:69:1269-1273.
Eagle KA, Coley CM, Newell JB, Brewster D, Darling R, Strauss H, Guiney T, Boucher C. Combining clinical and thallium data optimizes preoperative assessment of cardiac risk before major vascular surgery. Ann Intern Med. 1989;110:859-866.
Poldermans D, Fioretti PM, Boersma E, Cornel JH, Borst F, Vermeulen EGJ, Arnese M, El-Hendy A, Roelandt JRTC. Dobutamine-atropine stress echocardiography and clinical data for predicting late cardiac events in patients with suspected coronary artery disease. Am J Med. 1994;97:119-125.
Mazeika PK, Nadazin A, Oakley CM. Prognostic value of dobutamine echocardiography in patients with high pretest likelihood of coronary artery disease. Am J Cardiol. 1993;71:33-39.