Meeting Brochure and registration form      SMDM Homepage

Tuesday, October 23, 2007 - 9:15 AM
F-4

WHERE TO GO AND WHAT TO DO: A DECISION ANALYSIS OF AMBULANCE DIVERSION FOR MYOCARDIAL INFARCTION INTERVENTIONAL THERAPY

Henry Wang, MD, MS, University of Pittsburgh, Pittsburgh, PA, Oscar Marroquin, MD, University of Pittsburgh, Pittsburgh, PA, and Kenneth Smith, MD, MS, Section of Decision Sciences and Clinical Systems Modeling, Pittsburgh, USA USA USA.

OBJECTIVE: While national guidelines recommend diverting ambulance patients with acute myocardial infarction (MI) to primary percutaneous coronary intervention (PCI) centers, the effect of this strategy on survival is not known. We evaluated the hypothesis that ambulance diversion to PCI centers increases MI survival over community hospital fibrinolytics (FL).

METHODS: We evaluated a decision tree model comparing nearest community hospital FL vs. PCI at a more distant center. We used published studies for baseline parameter values and ranges. We assumed that paramedics carried 12-lead ECGs, diagnosed MI accurately and identified FL contraindications. We excluded scenarios with FL contraindications or closer PCI centers. We modeled the relationships between symptom-to-intervention time (sum of chest pain-to-911 call, ambulance response, on-scene, transport to nearest community hospital, additional transport to PCI center, and door-to-drug (FL) or door-to-balloon (PCI) times) and 30-day MI survival. We considered 1) base-case FL and PCI efficacy, 2) best-case FL efficacy, 3) base-case paramedic administered FL, 4) best-case paramedic FL. We performed one-way and probabilistic sensitivity analyses evaluating parameter value variations on the relative risk of MI survival.

RESULTS: The base-case favored PCI over FL (94.4% vs. 92.7% survival, RR 1.02) except where additional PCI transport >163 min (base 11 min) or door-to-balloon >233 min (base 84 min). Best-case FL favored FL when chest pain-to-911 was <60 or >220 min (base 81), additional PCI transport >15 min, door-to-drug <10 min (base 31) or door-to-balloon >220 min. Base-case paramedic FL favored PCI except where additional PCI transport >157 min or door-to-balloon >223 min. Best-case paramedic FL favored FL in nearly all situations. On-scene and transport to nearest community hospital times did not affect survival in any scenario. Probabilistic sensitivity analysis of the base-case favored PCI in 99.2% of iterations (RR 1.02; 95% range: 1.00-1.05), but did not indicate a preferred strategy for other scenarios.

CONCLUSIONS: While ambulance diversion of select MI patients to PCI centers may offer a small survival benefit over community hospital FL, in most scenarios neither is superior.


See more of Concurrent Abstracts F: Simulation and Clinical Strategies
See more of The 29th Annual Meeting of the Society for Medical Decision Making (October 20-24, 2007)