Evaluation of a prehospital chest pain protocol.

Abstract

To evaluate the diagnostic accuracy and outcomes for patients treated by use of a prehospital chest pain protocol (CPP). Consecutive case series for 1 year (1993) of prehospital nontrauma advanced life support (ALS) cases including hospital outcomes. Nonurban two-tiered emergency medical services system. Patients treated under the prehospital CPP or with hospital diagnoses of ischemic heart disease (IHD; ICD-9 between 410 and 414). Patients with cardiac arrest or dysrhythmias were excluded. Patients were given, by standing orders, ECG monitoring, i.v. access, and sublingual nitroglycerin. Further therapy was guided by on-line medical direction. Of 3,122 ALS nontrauma patients, 620 (20%) were treated with the CPP. All patients underwent ECG monitoring, i.v. access was started in 83%, and 61% received nitroglycerin. Only 55% of patients completed the entire CPP; patients who failed to complete the CPP had the same prevalence of IHD as those who completed it. When compared with hospital diagnosis of IHD, the CPP had a sensitivity of 69% (95% confidence interval [CI], 64% to 74%), a specificity of 87% (95% CI, 86% to 88%), and a positive predictive value of 42%. The positive likelihood ratio of CPP for IHD was 5.31, and the negative likelihood ratio was .36. The hospital mortality rate for all patients was 2.2%; for those with IHD, it was 1.6%. This prehospital ALS CPP had good diagnostic accuracy, but only half of patients completed it, and the hospital mortality rate was low. These data challenge the efficacy of the CPP.