BackgroundTo be used optimally, new interventional devices (stent, lasers, atherectomy catheters) must be carefully evaluated in terms of optimal patient and lesion selection, technique of use, expected acute success and complications, and long-term results. Sources for that information include singlecenter and multicenter (single-device) reports, although randomized trials may then be performed to provide a more definitive picture of any clinical benefits. One interim option, however, consists of carefully collected registry data. The purpose of this article is to review data collected in the National Heart, Lung, and Blood Institute-funded New Approaches to Coronary Intervention (NACI) Registry and to compare them with existing reports.Methods and ResultsNACI is an independent, investigatordriven effort that seeks to collect uniform data on patients undergoing treatment with one of several investigational devices and thereby provide an unbiased report of procedure outcome. Between November 1990 and November 1992, 36 participating centers treated a total of 3201 lesions in 2835 patients, using one of seven study devices: directional atherectomy (1084 lesions), transluminal extraction atherectomy (240 lesions), rotational atherectomy (349 lesions), Palmaz-Schatz stent (674 lesions), Gianturco-Roubin stent (213 lesions), and the Advanced Interventional Systems (474 lesions) or Spectranetics (167 lesions) excimer lasers. Data on each procedure were recorded on a unique modular database that captured the reason for (and interim result after) each device use. Device success (defined here as stenosis improvement by ≥20% and a residual stenosis <50% after new device use) was 66.5% overall. Adjunctive angioplasty was used in 75.5% of lesions, either before (25.9%) or after (43.5%) new device use, yielding an overall lesion success (≥20% stenosis improvement with a final residual stenosis <50% after all devices) of 92.2%. Adjunctive angioplasty after new device use produced further enlargement in minimal lumen diameter (from 2.2 to 2.7 mm) and further reduction in residual stenosis (26.4% to 16.1%) compared with the results present after use of the new devices themselves. Major complications consisting of death (1.6%), Q-wave myocardial infarction (1.3%), or emergency bypass surgery (1.7%) occurred in 4.0% of patients range, 2.6% to 8.7% across devices). Procedural success, defined as lesion success in all new device-treated lesions without a major complication, was achieved in 90.8% of patients, with a median length of hospital stay of 4 days.ConclusionsNACI illustrates the type of information that can be obtained in a registry format that examines the acute angiographic and clinical results of new devices according to uniform definitions. Although no registry can substitute for formal interdevice trials, registries such as this can supplement earlier single-center and multicenter reports. In doing so, they can help focus subsequent randomized interdevice comparisons on lesion types for which two or more devices have promising acute results. Given the substantial interdevice differences in baseline patient and lesion characteristics found in NACI, simple “head-to-head” comparison of the results of different devices might give misleading impressions and should be avoided unless such comparisons are restricted to carefully matched patient and lesion subgroups.