Frequency of complications depending on the CIED type and the hospital's implantation spectrum: a subgroup analysis of 250,000 cases
P Vitkauske, M G Maurilio Gutzeit, H J Heike Jaegers, H A Heinz-Juergen Adomeit, S H Susanne Macher-Heidrich, Z K Zana Karosiene, F H Fuad Hasan, M S Markus Zarse, H B Harilaos Bogossian, U W Uwe Wiegand, A K Axel Kloppe, F S Fabian Schiedat, C I Carsten Israel, B L Bernd LemkeAbstract
Background
Our primary study on the volume–outcome analysis (VOA) of complication rates in cardiac implantable electronic device (CIED) implantations demonstrated a statistically significant association between implantation volume and the occurrence of complications (p < 0.001, AUC 0.7113). In the current subgroup analysis, we investigated the dependence of complications on the implanted system and the implantation spectrum of the hospital.
Methods
The analysis is based on data from all first implantations of a CIED between January 1st, 2010, and December 31st, 2020, in any hospital in North Rhine-Westphalia (NRW). Data were collected from the mandatory registry of the German Institute for Quality Assurance and Transparency in Healthcare (IQTIG). All documented peri- and postoperative complications were analysed using the statistical software R, based on the report "Data Analysis on Minimum Case Volumes" (IQTIG 2020). The method includes risk adjustment and accounts for statistical uncertainties associated with small case numbers.
Results
A total of 249,118 cases from 250 hospitals were analysed, 85 implanting only pacemakers (PM), 38 implanting PMs and implantable cardioverter defibrillators (ICD) and 127 implanting all types of CIED including cardiac resynchronization therapy (CRT). Of these patients, 56.3% received a DDD-PM, 17.0% an ICD, 13.9% a VVI-PM, 11.3% a CRT system, and 1.5% another type of device. The overall complication rate was 2.4% (2.9% for DDD-PMs, 1.4% for ICDs, 1.8% for VVI-PMs, and 2.1% for CRTs). The VOA showed a statistically significant association between implantation volume and the occurrence of complications for all implanted systems. This association was stronger in the DDD-PM group (p < 0.001, AUC 0.7025), with an exponential increase in complication rates at an annual implantation volume below 30 (Fig. 1, a). At this minimum threshold, approximately 12% of complications could be avoided, and around 10% of patients should be transferred to other hospitals. For VVI-PM, a U-shaped relationship was observed, with an exponential increase in complications at an implantation volume below 10 and again above 75 (Fig. 1, b). For ICDs, and even more markedly for CRTs, there was an almost linear relationship between implantation volume and complication rate, with complications decreasing from 3% at 14 annual CRT implantations to 1% at an implantation volume of 90.
When analysing outcomes in relation to procedural diversity of centres, complication rates in hospitals performing a broader spectrum of implantations were significantly lower than in those with a more limited range of devices (Table).
Conclusion
Even 50 years after its development, the implantation of a DDD-PM still represents a challenge for some hospitals. The centre’s implantation spectrum plays a decisive role, with the lowest complication rates observed in hospitals implanting all CIED systems.Fig. 1 a and b. VOA for DDD- and VVI-PMsComplications by implantation spectrum