In this large, contemporary multicenter cohort, we observed substantial differences in clinical profiles, hospitalization patterns, and peri-procedural risk among patients receiving single-chamber ICDs, dual-chamber ICDs, and CRT-Ds. The main findings were as follows: (1) indications for hospital admission differed across device types, with a higher rate of emergency admissions in the dual-chamber ICD group compared with the single-chamber ICD and CRT-D groups; (2) patients receiving dual-chamber ICDs had higher clinical acuity and required more intensive post-procedural monitoring; (3) CRT-D recipients were older and had a greater burden of comorbidities but were managed safely; and (4) patients receiving single-chamber ICDs experienced the lowest procedural burden and the most favorable immediate safety profile.

First, dual-chamber ICD recipients consistently demonstrated the highest intensity of post-procedural care—including the longest overall hospitalization stay, the highest ICU admission rate, and prolonged ICU stay. This pattern likely reflects the underlying case mix: the dual-chamber ICD group had the highest proportion of secondary prevention indications such as prior cardiac arrest and ventricular arrhythmias. These patients typically require closer hemodynamic monitoring and more frequent ventilatory support. Previous large-scale analyses have similarly shown higher complication rates with dual-chamber devices, largely due to additional lead implantation and more complex device programming, although without translating into increased mortality [5, 7]. Our study confirms this trend, showing higher complication rates but preserved short-term survival.

Second, despite being older and having a significantly higher burden of comorbidities—including diabetes, chronic kidney disease, permanent atrial fibrillation, and conduction disease—CRT-D recipients experienced short ICU stays, low mechanical ventilation requirements, and in-hospital mortality comparable to the other device groups. These findings reinforce prior registry data indicating that CRT-D implantation, although technically more complex, can be performed safely in high-risk patients when standardized patient selection, peri-procedural protocols and operator experience are in place. The favourable short-term outcomes observed in the CRT-D group may reflect careful pre-implantation optimization, strict adherence to guideline-directed indications, and the avoidance of emergent implantation scenarios.

Third, single-chamber ICD recipients exhibited the lowest rates of peri-procedural complications and the lowest in-hospital mortality. The simpler procedural workflow, reduced lead manipulation, and shorter fluoroscopy and procedure durations likely contribute to these favourable outcomes. These findings are aligned with prior reports showing fewer acute complications in single-chamber systems [7], supporting their preferential use in patients who do not require atrial sensing or pacing.

These results also complement recent studies identifying renal dysfunction, frailty, functional status, and atrial arrhythmias as key determinants of short-term mortality after ICD or CRT implantation [6, 8, 9]. In our population, despite having more comorbid conditions, CRT-D patients did not experience higher short-term mortality. This underscores the importance of optimized peri-procedural management and appropriate use of CRT-D in guideline-indicated patients, even when baseline risk is high.

Interestingly, only 46.3% of CRT recipients presented with complete LBBB. Current cardiac device and heart failure guidelines recommend CRT primarily in patients with LBBB, LVEF ≤ 35%, and a QRS duration ≥ 150 ms, whereas CRT is not indicated in patients in sinus rhythm without a pacing indication who have a QRS duration < 130 ms despite reduced LVEF [10]. The relatively high proportion of non-LBBB patients may be explained by the use of a “pace-and-ablate” strategy. In patients with heart failure and atrial fibrillation undergoing atrioventricular node ablation, CRT may be considered irrespective of baseline QRS duration [10, 11].

For the main endpoint, defibrillator type and age were identified as adjusted predictors for MACCE. While age represents a continuous risk factor, defibrillator type may be considered also a decisive determinant. Patients receiving dual-chamber defibrillator systems experience sudden cardiac arrest more frequently. In addition, these procedures are usually not performed electively, in contrast to single-chamber ICDs or CRT-Ds. Age was also shown to be a strong predictor across several subgroups.

Regarding the second endpoint, for similar reasons, defibrillator type emerged as a significant risk factor for MACCE combined with ICU admission. In contrast, age was not identified as a relevant predictor for the combined second endpoint.

Taken together, this analysis provides a comprehensive overview of device-specific differences in early outcomes and may guide device selection and peri-operative resource planning. As ICD technology evolves—including the increasing use of subcutaneous systems and conduction system pacing—ongoing reassessment of implantation strategies and patient selection remains essential.

This study has several limitations. First, its retrospective design precluded assessment of long-term outcomes such as survival, arrhythmia recurrence, and rehospitalization. Key procedural and device-related variables—including device programming, procedure and fluoroscopy duration, lead-related complications, and 30-day post-discharge outcomes—were not available. Second, the analysis was based on ICD and OPS codes from 11 centers in Germany, limiting access to clinical details such as functional status, laboratory values, and echocardiographic parameters. No adjustment was made for clustering by center, operator volume, or operator experience—important confounders that may influence complication rates, ICU utilization, and resource use. Additionally, primary and secondary implantation indications could not be distinguished due to the coding-based dataset, and subcutaneous ICDs were not included. The analysis is restricted to in-hospital period. Further prospective studies are needed to better define the independent effects of these factors and to evaluate long-term clinical and functional outcomes.