[Coronary angioplasty--can the risk of recurrence be predicted on the day of surgery? A prospective study].
Data from a retrospective study defining seven parameters of increased risk of restenosis after successful transluminal coronary angioplasty (high-grade stenoses, long stenoses, eccentric stenoses, use of high pressure, extended time of balloon inflation, stenoses in obese patients, stenoses in patients without a history of smoking) were fed into a computer. A discriminant analysis was made and an algorithm for prediction of restenosis was defined. The validity of prediction was prospectively tested in 101 patients. In 80/101 (79.2%) prediction was possible; in 21/101 (20.8%) it was not possible. In 15/80 patients (18.8%) the prediction was: "restenosis probable"; in 65/80 patients (81.2%): "restenosis not probable". After 4.4 months 93/101 patients (92.1%) had an angiographic follow-up. The prediction "restenosis" proved to be correct in 13/15 patients (86.7%), and the prediction "no restenosis" was correct in 56/65 patients (86.2%). It is concluded that in the majority of patients the risk of restenosis can be predicted immediately after the intervention.