Abstract Body (Do not enter title and authors here): Introduction: Over 6 million patients (pts) present to US emergency departments annually with chest pain (CP), of which the majority are found to have no serious disease. Evaluation of these pts results in substantial costs for unnecessary hospitalization and extensive testing. We evaluated the utility of early discharge of selected low-risk (LR) CP pts from a chest pain unit (CPU) in which no predischarge testing or risk scores were used.

Methods: This retrospective study analyzed 1,037 consecutive LR CP pts from a prospectively recorded database. LR was based on normal examination, stable hemodynamics, normal electrocardiograms (ECG), and negative cardiac troponin I, without pre-discharge functional or anatomic cardiac testing or risk scores. We assessed demographics, comorbidities, medications, and major cardiac events at 30 d and 6 mos post-discharge.

Results: The study group of 1037 pts comprised 26% of the 4010 pts admitted to the CPU during the study interval from May 2005 to March 2015. Mean patient age was 55 yrs, 56% (n=575) were women, and comorbid conditions were frequent: hypertension (64.1%), dyslipidemia (46.1%), diabetes (25.7%), documented coronary artery disease (19.3%), previous revascularization (20.6%), previous myocardial infarction (10.1%). Length of stay (LOS) in the CPU to discharge was 10.4 hrs. Women received more discharge cardiac medications than men: antiplatelet agents, statins, beta blockers, ACE inhibitors, angiotensin II blockers, calcium channel blockers, and nitrates (p=0.0002 - 0.04).
Follow-up (F/U) at 30 d was 91% (n=948) complete and revealed 0.3% (n=3) cardiac deaths, 0.6% (n=6) acute coronary syndromes (ACS), and 6.2% (n= 64) receiving revascularization. F/U at 6 mos was 90% (n=936) complete and total cumulative cardiac deaths were 0.9% (n=9), 0.7% (n=7) ACS and 6.3% (n=65) revascularization.

Discussion: Cardiac events at both 30 d and 6 mos were very low and did not differ in men and women (P=0.8). LOS in CPU was minimized and patient safety was maintained. These selected LR pts remained at reduced risk for cardiac events despite a high rate of comorbidities. Early discharge of selected LR CP pts based on history, examination, ECG, and biomarker evaluation was safe and effective.

Conclusion: This approach in selected LR pts has the potential to reduce unnecessary diagnostic testing and CPU LOS. This strategy could lead to substantial savings in healthcare costs without compromising patient safety.