Piper Winthrop D, Malenka David J, Ryan Thomas J, Shubrooks Samuel J, O'Connor Gerald T, Robb John F, Farrell Karen L, Corliss Mary S, Hearne Michael J, Kellett Mirle A, Watkins Matthew W, Bradley William A, Hettleman Bruce D, Silver Theodore M, McGrath Paul D, O'Mears John R, Wennberg David E
Clinical Research Section, Department of Medicine, Dartmouth-Hitchcock Medical Center, Lebanon, NH 03756, USA.
Am Heart J. 2003 Jun;145(6):1022-9. doi: 10.1016/S0002-8703(03)00079-6.
Using a large, current, regional registry of percutaneous coronary interventions (PCI), we identified risk factors for postprocedure vascular complications and developed a scoring system to estimate individual patient risk.
A vascular complication (access-site injury requiring treatment or bleeding requiring transfusion) is a potentially avoidable outcome of PCI.
Data were collected on 18,137 consecutive patients undergoing PCI in northern New England from January 1997 to December 1999. Multivariate regression was used to identify characteristics associated with vascular complications and to develop a scoring system to predict risk.
The rate of vascular complication was 2.98% (541 cases). Variables associated with increased risk in the multivariate analysis included age >or=70, odds ratio (OR) 2.7, female sex (OR 2.4), body surface area <1.6 m(2) (OR 1.9), history of congestive heart failure (OR 1.4), chronic obstructive pulmonary disease (OR 1.5), renal failure (OR 1.9), lower extremity vascular disease (OR 1.4), bleeding disorder (OR 1.68), emergent priority (OR 2.3), myocardial infarction (OR 1.7), shock (1.86), >or=1 type B2 (OR 1.32) or type C (OR 1.7) lesions, 3-vessel PCI (OR 1.5), use of thienopyridines (OR 1.4) or use of glycoprotein IIb/IIIa receptor inhibitors (OR 1.9). The model performed well in tests for significance, discrimination, and calibration. The scoring system captured 75% of actual vascular complications in its highest quintiles of predicted risk.
Predicting the risk of post-PCI vascular complications is feasible. This information may be useful for clinical decision-making and institutional efforts at quality improvement.
通过使用一个大型的、当前的、区域性经皮冠状动脉介入治疗（PCI）登记系统，我们确定了术后血管并发症的危险因素，并开发了一种评分系统来评估个体患者的风险。
血管并发症（需要治疗的穿刺部位损伤或需要输血的出血）是PCI潜在可避免的结果。
收集了1997年1月至1999年12月在新英格兰北部连续接受PCI的18137例患者的数据。采用多变量回归来确定与血管并发症相关的特征，并开发一种评分系统来预测风险。
血管并发症发生率为2.98%（541例）。多变量分析中与风险增加相关的变量包括年龄≥70岁，比值比（OR）2.7；女性（OR 2.4）；体表面积<1.6 m²（OR 1.9）；充血性心力衰竭病史（OR 1.4）；慢性阻塞性肺疾病（OR 1.5）；肾衰竭（OR 1.9）；下肢血管疾病（OR 1.4）；出血性疾病（OR 1.68）；急诊优先级（OR 2.3）；心肌梗死（OR 1.7）；休克（1.86）；≥1处B2型（OR 1.32）或C型（OR 1.7）病变；三支血管PCI（OR 1.5）；使用噻吩吡啶类药物（OR 1.4）或使用糖蛋白IIb/IIIa受体抑制剂（OR 1.9）。该模型在显著性、区分度和校准测试中表现良好。评分系统在预测风险最高的五分位数中捕捉到了75%的实际血管并发症。
预测PCI术后血管并发症的风险是可行的。这些信息可能有助于临床决策和机构质量改进工作。
