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


Diagnosis

Meta-analysis: Thallium scintigraphy is accurate for diagnosis and prognosis in coronary heart disease

ACP J Club. 1991 May-June;114:87. doi:10.7326/ACPJC-1991-114-3-087


Source Citations

Kotler TS, Diamond GA. Exercise thallium-201 scintigraphy in the diagnosis and prognosis of coronary artery disease. Ann Intern Med. 1990 Nov 1;113:684-702.

American College of Physicians. Efficacy of exercise thallium-201 scintigraphy in the diagnosis and prognosis of coronary artery disease. Ann Intern Med. 1990;113:703-4.


Abstract

Objective

To determine the sensitivity and specificity of exercise thallium-201 myocardial perfusion scintigraphy for the diagnosis and prognosis of patients with known or suspected coronary artery disease.

Data sources

A BRS MEDLINE search covering the period from 1966 to 1987 was done using the key words coronary disease, myocardial infarction, radionuclide imaging, and thallium. Peer-reviewed reports of original investigations from 15 general internal medicine and cardiology journals were included along with studies identified by experts and from bibliographies of relevant papers.

Study selection

Papers with relevant titles were reviewed and studies were selected if their data allowed unambiguous calculation of both sensitivity and specificity.

Data extraction

The pooled calculations used data from 33 studies and more than 2000 patients. Data extracted included type of disease, individual trial sensitivities and specificities, and number of patients studied.

Main results

The pooled sensitivity of planar thallium scintigraphy by visual analysis for the diagnosis of any coronary artery stenosis was 84% (95% CI 82% to 85%). For 1-, 2-, and 3-vessel disease, the pooled sensitivity was 78% (CI 72% to 83%), 89% (84% to 92%), and 92% (CI 89% to 95%), respectively. Pooled specificity was 87% (CI 85% to 89%), {and the likelihood ratio of a positive and negative test were 6.5 and 0.18.}* For patients with "inadequate" results on an exercise electrocardiography stress test (because of poor exercise capacity), thallium scintigraphy had a sensitivity of 78% (CI 68% to 86%), a specificity of 85% (CI 74% to 92%), {and the likelihood ratio of a positive and negative test were 6.5 and 0.18.}* For patients with "uninterpretable" exercise test results, thallium scintigraphy had a sensitivity of 74% (CI 59% to 86%) a specificity of 89% (CI 75% to 97%), {and the likelihood ratio of a positive and negative test were 6.7 and 0.29}.* The average sensitivity of thallium scintigraphy for identifying patients with "high risk anatomy" (triple vessel or left main disease) was 46% (CI 40% to 51%), the average specificity was 73% (CI 69% to 77%), {and the likelihood ratio of a positive and negative test were 1.7 and 0.74.}* A normal thallium scintigram was associated with a low rate of subsequent coronary events (approximately 1%/y).

Studies frequently had flaws that could have biased their results: by including patients whose diagnosis was already known or healthy volunteers, or by excluding patients with low exercise heart rates or equivocal electrocardiograph responses.

Conclusions

Patients with abnormal resting electrocardiograms, restricted exercise tolerance, and an intermediate probability of having coronary artery disease are most likely to benefit from thallium scintigraphy. Thallium scanning shows some promise for prognostic clinical use, but more longitudinal studies must be done to identify which patients would benefit most.

Source of funding: Not stated.

Address for article reprint: Dr. G.A. Diamond, Cedars-Sinai Medical Center, 8700 Beverly Boulevard, Becker Building B210, Los Angeles, CA 90048, USA.

*Numbers calculated from data in article.


Commentary

Meta-analyses of thallium-201 imaging have previously been published (1, 2). This truly comprehensive overview was developed as background to a position statement for the American College of Physicians (3).

This overview has some limitations. Studies are included only up to the end of 1987. Although the use of thallium scintigraphy after myocardial infarction is reviewed, thrombolysis is not mentioned. Little is said about discordant results from exercise electrocardiography and thallium imaging. Sex differences are not explored.

Kotler and Diamond attempted to link practical and theoretic concerns. For example, interpretation of the test depend heavily on the accuracy of the clinician's estimate of the patient's previous (pretest) likelihood of having coronary artery disease. Interpretation is further complicated in thallium scintigraphy by the concurrent or previous application of exercise electrocardiography testing and the availability of other diagnostic and prognostic data. Thus, 1 conclusion is that we need to know more about the additional information provided by thallium imaging.

What evidence, then, would support the widespread use of thallium-201 imaging? —Evidence that shows that patient outcomes are improved, with acceptable incremental costs and adverse effects. To obtain this evidence, large randomized trials must be done in which patient outcomes with and without use of the technology are compared. How many technologies have been subject to such scrutiny? Virtually none, and as Kotler and Diamond argue, despite some impressive data supporting its use, thallium imaging is still another partially validated technology.

C. David Naylor
Sunnybrook Health Science CentreToronto, Ontario, Canada


References

1. Detrano R, Lyons KP, Marcondes G, et al. Methodologic problems in exercise testing research. Are we solving them? Arch Intern Med. 1988;148:1289-95.

2. Detrano R, Janosi A, Lyons KP, et al. Factors affecting sensitivity and specificity of a diagnostic test: the exercise thallium scintigram. Am J Med. 1988;84:699-710.

3. American College of Physicians. Efficacy of exercise thallium-201 scintigraphy in the diagnosis and prognosis of coronary artery disease. Ann Intern Med. 1990;113:703-4.