Comorbidities vs age for the prediction of excess of length of in-hospital stay in patients undergoing new pacemaker implantation

Abstract

Abstract Background Traditionally, advanced age has been considered a major factor for increased length of in-Hospital stay (LOS) and health-care costs. However, recent studies suggest that comorbidity burden may have an equal, if not greater, effect on LOS as age in a variety of clinical scenarios. Purpose We aimed to determine the impact of comorbidity on LOS, as compared to age, in patients undergoing new pacemaker (PM) implantation. Methods In this retrospective study, patients that underwent new PM implantation (both elective and non-elective) in our centre between January 2015 and June 2018 were admitted for chart-review. We excluded patients that underwent PM replacement and lead removal procedures. Furthermore, patients that underwent other invasive procedures, such as percutaneous coronary intervention, coronary bypass or heart valve surgery, were excluded in order to minimize confounding factors. Excess in LOS was defined as an in-hospital stay >3 days. Multivariate logistic regression models were generated in order to evaluate the impact of comorbidity burden (as assessed by the Charlson Comorbidity Index (CCI)) and age on LOS. Furthermore, we compared the excess of LOS in elderly patients (i.e. ≥75 years) and low comorbidity burden (i.e. CCI ≤1) with younger patients. Results 423 elective and non-elective patients with a mean age of 82 were included in the study. The median follow-up time was 34 (15–46) months. Baseline clinical characteristics are detailed in the table. A CCI >1 resulted to be a significant predictor for the excess of LOS, whereas age did not significantly predict excess of LOS (HR: 11.7 (7.3–18.9), p<0.001); HR: 1.04 (0.98–1.04), p=0.43, respectively). When analyzed separately, according to elective (n=150) vs non-elective (n=270) indication, CCI remained a significant predictor (HR: 7.1 (2.1–24.2), p=0.002; HR: 13.8 (7.6–25.1), p<0.001, respectively). Age remained not significant. Elderly patients (age ≥75 years) with low comorbidity burden (CCI≤1) had a significantly shorter LOS compared to a younger overall population (2 (1–3) vs 2 (2–4) days, p=0.016) (Figure). Conclusion Comorbidity burden, rather than age, significantly predicts excess of LOS and should therefore be the driving factor in the approach of patients undergoing new PM implantation. Funding Acknowledgement Type of funding sources: None. Figure 1