Routine chest X-ray post cardiac implantable electronic device implantation: is it really required? Experience from a tertiary centre

Abstract

Abstract Introduction The 2021 EHRA expert consensus statement on pacemaker (PPM) and transvenous implantable cardioverter defibrillator (ICD) implant technique proposes a chest X-ray (CXR) prior to hospital discharge in all patients. This recommendation serves the dual purpose of excluding pneumothorax following subclavian or axillary access (SAA) and documenting lead position. Purpose Given that cephalic venous access (CA) precludes pneumothorax, the utility of CXR as a tool for identifying early lead displacement (LD) in these patients warrants further examination. We studied the yield of post implant CXR, device interrogation (DI), or both for the identification of pneumothorax and LD in patients undergoing transvenous PPM or ICD implant via SAA or CA. Methods 4,051 consecutive PPM or ICD devices were implanted at a tertiary centre between 2016-2019. Lead position was documented fluoroscopically on conclusion of all procedures. A subsequent post implant CXR was performed in cases of SAA, plus in those CA patients requiring imaging for another indication (e.g. pulmonary oedema). Post implant DI was only performed in patients with a high anticipated burden of ventricular pacing (e.g. high-grade atrioventricular block). All patients underwent clinical review with repeat DI at 30 days post implant. LD was defined as a clinically significant translocation of a lead requiring repositioning or reprogramming. Results For the whole cohort, the incidence of LD at 30 days was 2.2% (n=89), and the incidence of pneumothorax (within the SAA group) was 0.9% (n=11). The venous access method used, post implant investigations, and the incidence of LD is shown in Figure 1. 1,282 patients (31.6%) required SAA; CXR was performed in all, and 9 LDs were identified (0.7%). However, four patients (0.3%) with ‘normal’ CXRs had evidence of LD on inpatient DI, and a further 17 LDs (1.3%) were confirmed on the 30 day DI despite a ‘normal’ CXR. 2,769 patients (68.4%) underwent CA; 309 (11%) required CXR. This identified LD in four patients (1.3%), with three further patients (1%) found to have LD on subsequent inpatient DI despite a ‘normal’ CXR. A further four LDs (1.3%) were diagnosed at 30 day DI. The majority (89%, n=2,460) of CA patients were discharged without CXR. 30% (n=825) of these patients required inpatient DI; this identified 13 LDs (1.6%). Accordingly, 59% (n=1,635) of CA patients had no post procedural investigations; 24 (1.5%) were found to have LD at 30 day DI. Conclusions In patients who developed LD within 30 days of a PPM or ICD implant, a post implant CXR identified this in only 32% of cases (n=13/41), and the overall sensitivity of CXR for LD was low (50%). Additionally, the majority of LDs (54%) occurred after hospital discharge. We suggest that the utility of post implant CXR for identifying LD is limited, and that CXR should not be considered an essential investigation in CA device implants.Figure 1