Impact of referral diagnosis on diagnostic yield in 1-day vs 7-day and 14-day extended wear holter

Abstract

Abstract Background Use of Extended Wear Holter (EWH) with monitoring durations up to 14 days has become common in clinical practice. When ordering a monitor, a referral diagnosis (RefDx) is provided to indicate the reason for the study. We previously presented data showing EWH monitors, irrespective of RefDx, provided a 2x diagnostic yield (DY) at 7d and 2.5x DY at 14d when compared to 1d Holter. Purpose To determine the impact of RefDx on DY for clinically significant arrythmias (CSA) using ePatch EWH monitoring for 1d, 7d, and 14d. Methods From October - December 2021, retrospective data was analyzed for RefDx (ICD-10 diagnosis code) and DY of CSA for ePatch with 1d, 7d, and 14d of analyzable data. RefDx were categorized as Common and Uncommon. Common RefDx included ICD-10 codes which represented at least 1.5% of total referrals on 1d monitors, while all RefDx with <1.5% were grouped together as Uncommon. CSA was defined as Bradycardia < 40 bpm, Pause ≥ 3 seconds, ≥ 2nd Degree AV Block, AF > 30 sec, SVT > 160 bpm for ≥ 3 beats, and VT > 100 bpm for ≥ 3 beats. DY was calculated as the percentage of patients with at least one CSA over the monitoring duration. Patients with Chronic AF were excluded from this analysis. Results RefDx for patients who wore ePatch for 1d (n=21,505), 7d (n=9,252), and 14d (n=11,269) are shown in the Table. Irrespective of monitoring duration, the most frequent RefDx was Palpitations (>35%). For each RefDx, the DY for any CSA at 1d vs 7d, 1d vs 14d, and 7d vs 14d was statistically significant (see Figure). Across all RefDx, DY increased 1.7x - 2.8x when comparing 1d to 7d, 1.9x - 4.2x when comparing 1d to 14d, and 1.1x – 1.5x when comparing 7d to 14d. For 1d, the lowest DY was for RefDx Chest Pain (13%) while the highest was for RefDx Bradycardia (43%). The DY benefit of extending monitoring from 1d to 7d was normally distributed across RefDX, while the DY benefit of extending monitoring from 7d to 14d was greatest (1.5x) for Chest Pain, and the DY benefit of extending from 1d to 14d was greatest (≥3.4x) for Chest Pain, Tachy Unspecified, Palpitations, Syncope/Collapse, and Dizziness/Giddiness. Notably, the most common RefDx of Palpitations demonstrated a DY of only 17% at 1d that increased 2.5x at 7d and 3.4x at 14d. Conclusion The impact of extending monitoring duration from 1d to either 7d or 14d was that a disparity in additive DY was observed when analyzing by RefDx. The 5 RefDx with the lowest DY for 1d of monitoring showed increases of ≥2.5x DY at 7d and ≥3.4x Dy at 14d of monitoring. Conversely, the 3 RefDx with the highest DY for 1d of monitoring showed <2X DY at 7d and <2.2x DY at 14d. Our data suggests that the RefDx can potentially be used to guide an optimal duration of telemetry monitoring.