Abstract 4505: Presentation To Non-PCI Center Is Major Modifiable Factor Associated With Delayed Door-to-balloon Time In HORIZONS-AMI
Introduction: Many pts with STEMI receive reperfusion later than recommended by current guidelines. Hypothesis: Factors associated with delayed door-to-balloon times (D2B) can identified; some may be modifiable.
Methods: In the HORIZONS-AMI study, 3602 pts within 12 hours of onset of STEMI undergoing primary PCI were prospectively enrolled. Times were recorded for onset of symptoms, arrival at PCI center, arrival in catheterization laboratory, and first balloon inflation. Pts transferred from a non-PCI center to a PCI center were considered separately from pts presenting initially to a PCI center. Logistic regression using 29 demographic/historic variables and 6 pre-PCI adverse events (arrhythmias, CPR, hypotension, respiratory failure) identified factors associated with (1) first door-to-balloon time > 90 minutes, and (2) for each time interval, the highest quintile.
Results: Complete time data was available for 2639 pts. Of 1848 pts presenting directly to a PCI center, 55% had D2B < 90 minutes. Of 781 pts transferred from a non-PCI center to a PCI center, 12% had D2B< 90 minutes. Presentation to a non-PCI center was the strongest predictor of D2B > 90 minutes (OR 9.1 compared to presentation to a PCI center, p < .0001). Other predictors of D2B > 90 minutes included history of congestive heart failure (OR 2.1, p = 0.0133), diabetes (OR 1.45, p = 0.002), prior MI (OR 1.43, p = 0.046), and baseline diuretic use (OR 1.36, p = 0.042), but not pre-PCI adverse events. Demographic factors correlated only weakly with delays in milestones between presentation and balloon inflation. Median D2B was similar for US (91 minutes) versus non-US pts (101 minutes, p = NS). After arrival at the cath lab, balloon inflation was delayed 1 hour (p < 0.001) in pts with prior CABG compared to pts without prior CABG.
Conclusions: In HORIZONS-AMI, D2B < 90 minutes was achieved in only one half of pts presenting to a PCI center and one tenth of pts presenting to a non-PCI center. The only potentially modifiable factor observed in this study was presentation to a non-PCI center. This data highlights the need for ambulance triage of pts with chest pain to bypass non-interventional centers and deliver pts directly to primary PCI centers of excellence.