Should We Check It? Assessing Interrogation of Cardiac Implantable Electronic Devices in the Emergency Department-The CHECK-ED Study: Implications for Service Planning and Care Delivery

Abstract

Background: Requests from the emergency department (ED) for cardiac implantable electronic device (CIED) checks constitute a large workload for cardiac electrophysiology services. We sought to determine the yield of, and clinical characteristics associated with, clinically relevant (remarkable) issues from ED CIED checks. Methods Consecutive CIED checks from our ED over a 12-month period were studied. A remarkable issue (RI) was defined as arrhythmia relating to the presentation or device/lead issue requiring reprogramming or intervention. The association between the presenting complaint and an RI was assessed using regression analysis. Multivariable regression model was used to identify pre-specified patient-level characteristics that were predictive of a RI. Results: A RI was found in 28% (n=98) of 354 ED CIED checks for 306 patients (76616 yrs, 59% male). Most patients had no RI (n=224, 73%). One third of checks occurred after-hours and these had a higher yield of RIs than those during routine clinic hours (35% vs 23%, p=0.018). Presenting with a perceived ICD shock was predictive of a RI (odds ratio [OR] 6.0, 95% CI=1.8–20.0). Syncope/presyncope was five-fold less likely to be predictive of a RI (OR 0.19, 95% CI=0.13–0.28) despite being the most common indication for CIED check (51%, n=180 checks). Only history of AF was predictive of RI while advancing age was predictive of not finding a RI. Conclusion: Almost three-quarters of ED CIED checks did not yield any RI. Patient-reported ICD shock and history of AF were predictive of RI, while syncope/presyncope was not. New models of care especially during afterhours, may help to reduce the burden on cardiac electrophysiology services and health care costs.