Current issues of ACP Journal Club are published in Annals of Internal Medicine


Risk stratification for patients having peripheral vascular surgery

ACP J Club. 1992 Sept-Oct;117:55. doi:10.7326/ACPJC-1992-117-2-055

Source Citation

Wong T, Detsky AS. Preoperative cardiac risk assessment for patients having peripheral vascular surgery. Ann Intern Med. 1992 May 1;116:743-53.



To evaluate the accuracy of clinical risk indexes, exercise stress testing, dipyridamole-thallium scanning, gated blood pool ejection fraction measurement, and electrocardiographic ST-segment monitoring in the preoperative cardiac risk stratification of patients having peripheral vascular surgery.

Data sources

All English-language studies published before August 1991 were identified using a MEDLINE search on the following keywords: vascular surgery, coronary disease, preoperative care, and postoperative complication. Bibliographies of identified articles were also reviewed.

Study selection

All clinical studies evaluating methods used for preoperative cardiac risk stratification of patients having peripheral vascular surgery.

Data extraction

Data extracted included the criteria for inclusion and exclusion of study patients, the techniques used for testing, the definitions of positive results, and the clinical outcomes. A Bayesian conceptual framework was used to determine the accuracy of the various stratification techniques.

Main results

The 3 clinical indexes (Goldman, Detsky, and Eagle) were limited by their inability to predict postoperative complications in patients with low or moderately low scores (low sensitivity). Exercise stress testing failed as a general screening test primarily because many patients (approximately 70%) failed to reach their target heart rate. A negative test result did not significantly lower post-test probability of cardiac complications (low sensitivity), particularly in patients with limited exercise tolerance. Recent studies suggest that resting gated blood pool scanning is not useful in risk stratification. Some studies suggest that absence of redistribution on dipyridamole-thallium scintigraphy rules out high-risk status; however, several recent reports challenge this finding. Perioperative ST-segment monitoring for ischemia shows promise for predicting cardiac risk, but recent data suggest that it may not be sensitive enough to identify patients at low risk.


Patients identified clinically to be at high risk for cardiac complications after peripheral vascular surgery are unlikely to benefit from further risk stratification. Dipyridamole-thallium scanning and ST-segment monitoring may be helpful in risk stratification for those patients with intermediate clinical risk, but are unreliable for those patients at low clinical risk.

Source of funding: Health and Welfare Canada.

Address for article reprint: Dr. T. Wong, St. Boniface General Hospital, Department of Medicine, 5C, 409 Tache Avenue, Winnipeg, Manitoba R2H 2A6.


In the review by Wong and Detsky, exercise testing was found to have a limited role in preoperative cardiac risk assessment because patients requiring peripheral vascular surgery frequently cannot achieve the necessary workload. Left ventricular ejection fraction (gated blood pool scanning) was predictive of future events in only some studies. Dipyridamole-thallium imaging (especially semiquantitative) was predictive of perioperative cardiac events in many studies, although a recent study reported limited sensitivity (1). Electrocardiographic ischemia monitoring was predictive of future events in 4 of 5 studies. The heterogeneity of predictive accuracies of these tests may be a function of variation in how tests were done in various centers, in different patient populations, and in patient selection. Publication bias may also exist because studies reporting value from preoperative screening tests are more likely to be published than those studies showing negative results.

We suggest the following for the clinician's approach to preoperative risk assessment in peripheral vascular surgery patients: Stratify patients into low-, medium-, and highrisk groups based on clinical markers (2); for the medium- and high-risk groups, choose the noninvasive test most feasible for the patient. For those who cannot exercise, we prefer pharmacologic stress testing because semiquantitation may allow for the identification of patients with advanced coronary disease, especially left-main or 3-vessel disease (3). The very-high-risk patient needs individual consideration of the options: coronary angiography with possible bypass surgery first; peripheral vascular surgery with maximal perioperative medical therapy; or cancellation of peripheral vascular surgery. Finally, optimal perioperative care requires collaboration among the internist, the cardiologist, the anesthesiologist, and the surgeon.

Sumita D. Paul, MD, MPH
Kim A. Eagle, MD Massachusetts General Hospital Boston, Massachusetts