Tu and colleagues are to be commended for their thought-provoking article [1]. However, two important points of difference were omitted from analysis and from discussion in the paper.

Tu and colleagues do mention, in one sentence, that 90% of the patients in New York State had Canadian Cardiovascular Society class 3 or 4 angina. However, they have based their paper on the objective finding that the severity of coronary artery disease can be defined on the basis of angiographic abnormality. Although anatomic categorization is important, it can be only an intermediate step in the management of ischemic heart disease. In a recent review of advances in coronary angioplasty [2], Bittl points out that correlation is poor between the angiographic severity of coronary artery disease and the risk for ischemic complications. He states that angiographically mild stenoses not conventionally targeted for treatment are inherently more likely than severe lesions to cause myocardial infarctions.

Unfortunately, defining the severity of angina involves subjective criteria, and the need for intervention may well be related to the physician eliciting the symptoms. An important difference was seen in the rates of performance of coronary angiography in New York and Ontario: Coronary angiography was performed much more frequently (2.20 times more often) in New York than in Ontario. Tu and colleagues do not identify the difference in the ratio of cardiologists to family physicians in New York and Ontario. This ratio is much higher in New York than in Ontario. and cardiologists can be readily seen by patients in New York who have not been first seen by family physicians. Once a patient is referred to a cardiologist, it can be expected that coronary angiography will be ordered when the subjective symptoms that define angina pectoris are being evaluated. This may be why coronary angiograms are ordered so much more frequently in New York than in Ontario. On the other hand, rates of percutaneous transluminal coronary angioplasty or coronary artery bypass graft (CABG) surgery performed per angiography in the two regions differ little; this demonstrates concordance on the criteria for these interventions. The difference in the number of cardiologists in New York compared with Ontario is very important and must be considered in attempts to evaluate differences in rates of CABG surgery.