(Comments in: N Engl J Med. 2001;345(4):294-5; discussion 295-6. N Engl J Med. 2001;345(4):294; discussion 295-6. N Engl J Med. 2001;345(4):294; discussion 295-6. N Engl J Med. 2001;345(4):295; discussion 295-6. N Engl J Med. 2001;344(9):645-54. N Engl J Med. 2001;344(9):677-8. Comment on: N Engl J Med. 2001;344(9):645-54.)

The purpose of this study was to test the hypothesis that patients who were classified as appropriate candidates for revascularization by an expert panel but did not undergo the procedure would have worse outcomes than those who did undergo it, independent of other clinical characteristics. Data for this study were drawn from the Appropriateness of Coronary Revascularization study in London. Patients were eligible for inclusion in the study if they were to undergo elective or emergency coronary angiography in specified areas between 1996 and 1997. Study patients were followed for 2.5 years. Appropriateness was rated using the RAND criteria.

Of the 2,552 patients analyzed, 908 had indications for which percutaneous transluminal coronary angioplasty (PCTA) was deemed appropriate, 1,353 had indications for which coronary-artery bypass surgery (CABG) was deemed appropriate, and 521 patients were deemed appropriate candidates for both procedures. PTCA was performed on 36% of the patients who were rated as appropriate candidates, on 22% of the patients who were rated as "uncertain," and on 6% of the patients who were rated as inappropriate. CABG was performed on 57% of the appropriate patients, on 21% of the uncertain patients, and on 8% of the inappropriate patients. Of the 308 patients who were appropriate candidates for PTCA, but who only received medical treatment, the recorded intention of the physician at the time of angiography was to use medical treatment for 89% of them. Of the 354 patients appropriate for CABG who received medical treatment, the recorded intention of the physician was to use medical treatment for 81% of them. Patients who did not receive appropriate procedures were more likely to be nonwhite, as well as have several clinical characteristics (this was statistically significant for CABG).

Patients who were classified as appropriate candidates for PTCA but who received medical treatment were more likely to have angina at follow-up (OR=1.97) than those who actually received the PTCA, but the two groups were equally likely to die or have a nonfatal MI during the follow-up. For CABG, those who received only medical treatment were again more likely to have angina (OR=3.03) and to die or have a nonfatal MI (OR=4.08).

Thus, this study documented that underuse of PTCA and CABG according to RAND appropriateness criteria has negative implications for 2.5 year outcomes for patients with heart disease. This study also demonstrated that, in the UK, as well as in the US, nonwhite racial groups are less likely to receive appropriate PTCA and CABG procedures.

In a letter by Barr DA, this study was criticized for not adjusting for social class. He argued that a larger number of non-whites were in the medical treatment group, and, since low social class and non-white race are associated, probably a larger proportion of the medical treatment group was of a lower social class. Thus, the poorer outcome of the medical treatment group might be due to poorer outcomes of lower social class patients (independent of treatment).