Background: Practitioners often assume a close relation between angiographic coronary artery stenosis and patient functional capacity. To test this unproven hypothesis, we analyzed the relation between coronary artery stenosis measured by different methods and maximal treadmill exercise tolerance in patients with single-vessel disease before and after intervention by percutaneous transluminal coronary angioplasty (PTCA).
Methods and results: Coronary angiography and maximal exercise testing off anti-ischemic medication were performed before random assignment of 227 patients with single-vessel coronary artery disease to PTCA or drug therapy. Six months later, angiography and exercise testing were repeated with patients assigned to PTCA off anti-ischemic therapy so that the altered coronary stenosis was the only consistent variable. Patients assigned to drug therapy were exercised on drug therapy. Coronary stenosis was assessed visually by the local investigator and quantitatively by blinded caliper and computer methods in central laboratories. Variabilities of caliper and computer measurements were established in a subset read twice. Visually estimated stenosis > or = 90% at baseline was associated with shorter exercise duration (7.9 versus 9.2 minutes, P < .04). Similar segregation at baseline was not observed with caliper or computer methods. Regardless of the method of measurement used, correlation between changes of lesion severity and exercise duration from baseline to follow-up was poor. Patients were angiographically classified as "better," "unchanged," or "worse" if follow-up stenosis was below, within, or above 2 SD of mean technical variability from baseline (+/- 18.8%, caliper, +/- 14.6%, computer). Exercise duration for PTCA patients improved among those with better lesions (+2.4 minutes, n = 50, P = .001) but also among those with unchanged lesions (+1.9 minutes, n = 41, P < or = .001). Unchanged medically treated patients improved less (+0.5 minutes, n = 86, P = .04). Results were similar when patients were angiographically classified by minimum lumen diameter.
Conclusions: Handheld calipers and quantitative coronary angiography are equivalent techniques for making anatomic measurements. Neither method identified patients having reduced exercise capacity at baseline as well as visual estimation. The relation between changes of coronary stenosis and exercise duration is highly variable, at least in part because of the insensitivity of angiographic methods for detecting small but potentially important changes. Minimal anatomic improvement 6 months after PTCA does not preclude a good functional outcome. Contrary to common belief, angiographic stenosis does not correlate well with functional capacity, even in patients with single-vessel disease.