ABC | Volume 112, Nº5, May 2019

Original Article Warpechowski Neto et al Hospital readmission – implantable devices Arq Bras Cardiol. 2019; 112(5):491-498 Methods Study design This was a prospective single-center cohort study developed in a high-complexity cardiology hospital located in southern Brazil. Population Patients aged 18 years or older submitted to ICD implantation or CRT from February 2014 to July 2015 were consecutively included. The cases of changes in generators without lead implantation were excluded. Data collection Data collectionwas performed fromthemoment immediately prior to the device implantation and, subsequently, at the time of visits to the emergency unit for medical care. The obtained data, recorded in the electronic medical record, were exported to an Excel database. The analyzed variables were clinical ones and related to the implant. The clinical variables were: (1) LVEF at the two-dimensional echocardiogram (Teicholtz or Simpson method, when indicated); (2) HF etiology, defined as valvular, ischemic or non-ischemic. In case of more than one etiology, we selected the one considered to be predominantly accountable for the condition; (3) patient functional class, classified according to the New York Heart Association classification. The variables related to the CIED implant were: (1) operative wound infections; (2) operative wound pain; (3) need to replace the leads; (4) upper limb venous thrombosis; and (5) device pocket hematoma. Outcomes The primary outcomewas the occurrence of an unscheduled hospital visit related to the CIED implant. Visits resulting from HF worsening or progression were also considered for this analysis. The secondary outcome was all-cause mortality. Ethical Considerations This study was approved by the Research Ethics Committee of Instituto de Cardiologia – Fundação Universitária de Cardiologia do Rio Grande do Sul, under number 4983/14. Data analysis The Statistical Package for Social Sciences (SPSS), version 18.0, was used for the analyses based on the data stored in the Microsoft Excel spreadsheets. A two-tailed p value of less than 0.05 was considered significant in the statistical analyses. Continuous variables with parametric distribution were expressed as mean ± standard deviation, while nonparametric variables were shown as median and interquartile range. The comparisons were made with Student's t -test for independent samples in variables with central tendency distribution and with Mann-Whitney test in those considered to be asymmetric. Categorical variables were expressed as absolute (n) and relative (%) frequencies and compared with the chi-square test. Probability estimates were calculated by the Kaplan-Meier method using long-rank. It was not possible to perform the multivariate analysis using Cox Regression due to the absence of events occurring in the ICD group, which prevented the calculation of the hazard ratio (HR). Results During the study period, 1,174 surgeries were performed for device implantation. Of these, 224 were for ICD/CRT, 25 of which were exclusively for generator change, and were excluded from the evaluation. Figure 1 shows patient inclusion flowchart. The final analysis was performed with 199 patients. Table 1 shows the characteristics of the assessed population. There was a higher prevalence of male individuals in both groups. The mean age was similar, as well as the ejection fraction. The non-ischemic cardiomyopathy was the most prevalent etiology in both groups. Most implants were performed through the Brazilian Unified Public Health System (SUS). There was a statistically significant difference between the functional classes, with a higher percentage of patients in class III in both groups. Of all procedures, 57% were carried out for primary prevention of sudden death. Outcomes Regarding the outcomes, the rate of unscheduled visits to the emergency unit related to the device was 4.8% in the ICD group and 10.6% in the CRT group (p = 0.20). Operative wound pain was the most prevalent complication related to the device (Table 2). Figure 2 shows the incidence of the primary outcome of emergency visits-free survival during a median follow-up of 285 days (p = 0.214). The incidence of unscheduled visits related to clinical conditions (not related to device implantation) did not differ between the groups, being 28.2% in the ICD group and 18.6% in the CRT group (p = 0.17), including readmissions due to HF, as shown in Table 3. Figure 3 shows the survival curve in both groups. There were 4 deaths in the CRT group and none in the ICD group. None of the deaths were related to the procedure itself. The causes of death were: 1 hemorrhagic stroke, 1 sudden death at home, 1 death due to multiple organ failure as a complication of infective endocarditis (secondary to a dental abscess, diagnosed 194 days after the implant) and 1 due to refractory HF. Discussion Artificial cardiac stimulation has shown significant benefits since its initial implantation in 1958, crossing generations in continuous technical evolution and extending its range from the atrioventricular conduction disorders to dyssynchrony reduction. However, it still shows a significant percentage of complications, despite almost sixty years of use. Currently, the volume of procedures for CIED implantation shows increasing annual rates, due to the technological evolution of the method, the increase in the indications and the higher number of eligible patients. 6 At the same time, the greater longevity of the affected populations is a non-modifiable risk factor for long-term complications. Such a change in the scenario limits 492