Changes in the practice of percutaneous coronary intervention: a comparison of enrollment waves in the National Heart, Lung, and Blood Institute (NHLBI) Dynamic Registry.

The National Heart, Lung, and Blood Institute Dynamic Registry includes 15 clinical sites in wave 1, and 16 sites in wave 2 as well as a data-coordinating center. The first wave of enrollment began in July 1997 and was completed in February 1998. The second wave began in February 1999 and ended in June 1999. There were a total of 2,526 patients in wave 1 and 2,109 patients in wave 2. Comprehensive pre-, intra-, and postprocedure (in-hospital) data were analyzed for changes between recruitment waves. Patients in wave 2 were more frequently nonwhite (p < or = 0.001), hypertensive by history (p < or = 0.001), had more significant noncardiac comorbidity (p < or = 0.01), and had more frequently undergone prior percutaneous coronary intervention (p < 0.05). Patients in wave 2 underwent percutaneous coronary intervention in a setting of acute coronary syndromes more frequently than wave 1 patients (p < or = 0.001). However, most interventions in both waves were performed on 1 vessel, irrespective of the extent of disease. Attempted lesions in wave 2 were longer (p < or = 0.001), less frequently totally occluded (p < or = 0.001), and more frequently in vessels with a prior stent (p < or = 0.01). Using the American Heart Association/American College of Cardiology lesion classification scheme, attempted lesions in wave 2 were less complex than those in wave 1 (p < or = 0.001). Stent use increased significantly from wave 1 (67%) to wave 2 (79%, p < or = 0.001) as did the use of platelet glycoprotein IIb/IIIa antagonists (wave 1, 24%; wave 2, 32%: p < 0.001). Procedural outcomes (angiographic success without major in-hospital adverse events) were excellent in both waves 1 (94.6%) and 2 (95.6%) and were not significantly different. However, the frequency of significant procedural coronary dissection and in- and out-of-laboratory abrupt closure were significantly less in wave 2 (p < or = 0.001) Discharge medications were more likely to include angiotensin-converting enzyme inhibitors, beta-adrenergic blocking agents, and hypolipidemic treatment in wave 2 than in wave 1 (p < or = 0.001). These data indicate a continuing aggressive approach to patient care over the time interval analyzed. Although overall procedural outcomes are excellent, procedural safety has been further enhanced. There is also a growing awareness of the importance of secondary prevention among interventional cardiologists.