Abstract 322: Cyclooxygenase (COX) Inhibition Prior to CPR Decreases Heart Rate Variability in Pigs

Abstract

Background: Cyclooxygenase inhibitors (COX-1 and 2) are widely used and inhibit prostaglandin synthesis. COX inhibitors have been shown to increase overall cardiovascular risk. Heart rate variability (HRV) is a measure of the balance of autonomic nervous system and shown to be predictive of neurological outcome after cardiac arrest. Prostaglandins are cardioprotective and modulate HRV. We hypothesized that prostaglandin inhibition impacts short term CPR survival and outcomes based on HRV in swine. Methods: 24 animals (30±5kg) were randomized to pretreatment with indomethacin (I) 2mg/kg (COX-1 Inhibitor), Celecoxib (C) 2mg/kg (COX-2 inhibitor) or placebo (P). VF was induced and after 3 minutes all animals received chest compression and ventilation. After 18 minutes of VF, vasopressin given and defibrillation attempted. Electrocardiogram, echocardiogram and hemodynamic measurements done at baseline (BL), after infusions (Tx) and return of spontaneous circulation (ROSC) at 30 and 180 minutes. Results: ROSC was achieved; 3/8 (I) compared with 7/8 (C), and 7/8 (P). No differences in blood gases or hemodynamics pre, during or post CPR between groups. Echo showed decrease function post resuscitation in surviving animals but not significantly different among groups. COX-2 inhibition induced a significant decrease in linear (SDNN, RMSSD) and frequency (HF) measures of HRV towards greater sympathetic tone, post resuscitation. Compared to P, COX-2 inhibition increased Troponin I levels at 180 min after ROSC; P [84(4)] vs C [721(31)] mg/dl (p< 0.001). Conclusions: COX-1 inhibition decreases ROSC, whereas COX-2 inhibition significantly increases indices of myocardial tissue damage, and decreases HRV. The impact on long term outcome is unknown. Since many adults use COX-1 or COX-2 inhibitors, studies analyzing post resuscitation outcomes of patients should consider the effects of prostaglandin synthesis inhibitors as confounding variables.