Angiographic restenosis after angioplasty: comparison of definitions and correlation with clinical outcome

The aim of this study was to assess which of the currently used definitions of restenosis most closely indicates degree of recurrence and clinical status by 1) correlating percentage luminal renarrowing with restenosis defined according to each of four definitions, and 2) evaluating which definition was best predicted by clinical recurrence.

Quantitative angiography in 125 patients was undertaken either at time of early clinical presentation or at 6-month follow-up after percutaneous transluminal coronary angioplasty (PTCA). Absolute luminal diameters measured before and after PTCA and at follow-up were plotted as the percentage return from post-PTCA toward pre-PTCA value. All patients were also defined as restenosed or not restenosed according to each of the four definitions.

The angiographic restenosis rate varied from 31% to 47%. Other than for "loss of 50% absolute gain," all definitions defined restenosis in some patients, despite the degree of return from post-PTCA to pre-PTCA value being less than 50%. Early recurrent symptoms predicted angiographic restenosis best, irrespective of angiographic definition, whereas history of recurrent angina or positive exercise testing alone at follow-up were poor predictors (range, 0.46 to 0.54). The predictive value increased (0.75 to 0.87) when exercise testing was positive in patients complaining of angina. The definition "loss of 2 standard deviations" gave the lowest values for positive or negative predictive values irrespective of clinical parameter.

"Loss of 50% absolute gain" may be the best compromise definition. Patients admitted early with angina should undergo recatheterization, whereas exercise tests should be reserved for patients who complain of angina at routine follow-up.