Detection of restenosis after percutaneous transluminal coronary angioplasty by an angiographic score

Background. Restenosis, or "Achilles heel" of balloon angioplasty, remains in a range near 30% and cannot be predicted with certainty. To find adequate methods for its foresight is a challenge. Objectives. To determine risk factors and an angiographic score to predict the appearance of restenosis after one site percutaneous transluminal coronary angioplasty (PTCA). Methods. We restudied prospectively 315 (239 men, 76 women, age range from 29 to 78, 53.6 +/- 9.5 years) of 360 patients who underwent PTCA to a native coronary artery. The study didn't include patients with left main disease, total occlusion, side-branch involvement, ostial stenosis, acute phase of myocardial infarction or those who repeat PTCA. Two hundred and twenty-eight patients underwent PTCA of the left anterior descending artery, 56 of the right coronary artery and 31 of the left circumflex artery. Results. Restenosis, defined as a luminal renarrowing > 50% at follow-up, was present in 82 (26%) patients between 1 and 8 months after the procedure. Univariate and multivariate analysis revealed four vascular factors related to restenosis (p < 0.05): a) lesion length and; b) irregularity of the lesion borders before PTCA; c) perivascular and/or endovascular haziness and; d) intensity of residual stenosis after PTCA. To construct the score, a zero was given to a lesion length < 8 mm; to smooth lesion borders; to residual stenosis up to 20%; and to absence of haziness after PTCA. A one was given to a lesion length between 8 and 10 mm; and to a lesion with irregular borders. A two was given to a lesion length > 10 mm. A three was given to a residual stenosis > 20%; and to the presence of haziness after PTCA. The sum of all terms was considered the final score. So, it could oscillate from 0 to 9. Calculated score from 0 to > 5 showed respective restenosis rates of (%): 5.2; 15.1; 24.0; 39.4; 44.8; 60.8; and 84.2. The calculated correlation coefficient (0.98) among the scored values and the correspondent restenosis rates was highly significant (p < 0.001). Conclusions. It is concluded that restenosis is primarily a multifactorial problem based on vascular factors and may be predicted with a high degree of probability by the proposed score.