Cost Comparison Of Conventional Monitoring Strategies To Diagnose Syncope Patients With Pause Arrhythmias Relative To Insertable Cardiac Monitoring: A Simulation Approach

Abstract Background The infrequent nature of syncope can make diagnosis of cardiac pauses challenging with conventional monitoring (CM) strategies of intermittent external short-term ECG monitoring. Insertable cardiac monitors (ICMs) continuously monitor for arrhythmias and are well-established to have a higher likelihood of diagnosis compared to CM. It is not well understood whether the higher up-front cost of ICM is offset by the cost of repeat evaluation in a CM strategy, per diagnosed patient. Purpose The objective of this analysis was to simulate the cost per patient diagnosed with pause arrhythmias, between various CM strategies and ICM monitoring. Methods ICM device data from syncope patients was utilized to simulate patient pathways with CM. We assumed that detected true pause episodes (≥5 seconds) were symptomatic and prompted further evaluation: simulated inpatient or outpatient emergency department (ED) assessment, followed by external ECG monitoring of varying durations (24 or 48 hours, 14 or 30 days, or two 30-day monitors) beginning at random within the week after discharge. Subsequent true pause episodes in yet undiagnosed patients triggered additional rounds of CM. ECG diagnosis was considered successful if a pause episode occurred on the same day as simulated CM. Costs of evaluation and monitoring were accrued at each encounter. Inpatient and outpatient (ED) syncope evaluation costs (mean £3,746 and £367, respectively) were based on national episode statistics data and national average tariffs, and simulated at random from log-normal distributions; costs of external ECG monitors and ICM (including ICM device, insertion, remote monitoring, and explant) were fixed. Costs stopped accruing once a patient was diagnosed. We computed costs per diagnosed patient by dividing the total costs accrued for all patients by the number of patients diagnosed across 1,000 simulations. Longer pause definitions of ≥6–7 seconds were also evaluated. Results A total of 105 true pause episodes from 44 patients (mean (SD) age 66 (17), 48% male) were detected by ICM during 505 (333) days of follow-up. Patients experienced an average of 2.4 (2.7) pause episodes ≥5 seconds during follow-up. Relative to ICM-diagnosed patients, CM diagnosed between 13.8% (24-hour holter) to 30.2% (two 30-day monitors) of these patients. Cost per ICM-diagnosed patient was £2,985, whereas in the CM strategies the average cost per diagnosed patient ranged from £18,519 (£7,603) with 24-hour holter to £5,656 (£890) for two 30-day monitors (Figure). Costs per diagnosed patient further increased for pause durations of ≥6–7s, as the percent of patients diagnosed via CM decreased. Conclusion Relative to syncope patients diagnosed with pause arrhythmias via ICM, CM strategies diagnose fewer patients and incur significantly greater costs per diagnosed patient. Real-world suboptimal patient compliance with external monitoring would further decrease the cost-effectiveness of CM. Funding Acknowledgement Type of funding source: Private company. Main funding source(s): Medtronic