ABC | Volume 112, Nº5, May 2019

Original Article Warpechowski Neto et al Hospital readmission – implantable devices Arq Bras Cardiol. 2019; 112(5):491-498 Figure 2 – Kaplan-Meier estimate of survival probability according to device-related unscheduled emergency visits. Note: p = 0.214. 1.0 0.8 0.6 0.4 0.2 0.0 0 100 200 300 400 500 600 Number at risk 61 34 25 13 4 0 p = 0.214 Time (days) Survival (%) Device ICD CRT Group (ICD) Group (CRT) 110 82 60 39 23 0 Table 2 – Outcomes of the study population. Device-related unscheduled emergency visit Variable Total n = 199 ICD n = 124 CRT-P/D n = 75 p value Device-related unscheduled emergency visit * 14(7%) 6(4.8%) 8(10.6) 0.20 Device-related complications * 0.45 Surgical wound infection 5(2.5%) 2(1.6%) 3(4%) Surgical wound pain 6(3%) 2(1.6%) 4(5.3%) Lead change 1(0.5%) 0 1(1.3%) Upper limb venous thrombosis 1(0.5%) 1(0.8) 0 Pocket hematoma 1(0.5%) 1(0.8) 0 Inappropriate shocks 2(1%) 2(1.6%) 0 Mortality 0.008 Related to device implantation 0 Other causes 4(2%) 0 4(5.3%) * Data shown as absolute and relative frequency; ICD: implantable cardioverter-defibrillator; CRT-P/D: cardiac resynchronization therapy. Van Rees et al., 6 in a systematic review of 18 clinical trials involving ICD/CRT implantation, found a mortality rate of 2.7% after ICD implantation (0.6% if considering only those without thoracotomy) and 0.7% after CRT. The incidence of lead dislodgement was 1.8% in the ICD group (without thoracotomy) and 5.9% in the CRT studies. Device pocket hematomas occurred in 2.4% of those with ICD and 2.2% with CRT; however, these percentages represented only cases that required surgical intervention. In our study, there was no difference between devices regarding the incidence of cable dislodgement, which is probably due to the low incidence of this complication in our sample. The incidence of CIED pocket hematoma found in our center was 0.8% in the ICD group and 0% in the CRT group. When compared to other cohorts, we also found similar incidences of CIED-related complications. In a cohort of 1,929 patients, the incidence of surgical reintervention due to stimulation cable dislodgement, infection and mortality was, respectively, 4.4%, 1.5% and 3.2%. 9 Our cohort showed lower mortality and cable dislodgement rates, but 1%more incidence of bleeding. Among the patients with CRT, the incidence of cable dislodgement was 5%, compared to 1.3% in our cohort. A retrospective record of 30,984 Medicare users submitted to device implantation found an incidence of major complications (cable dislodgement, cardiac tamponade, hemothorax and pneumothorax) of 4.26%, with no difference between CRT and ICD. 12 In the same analysis, ICD implantation showed a higher incidence of mechanical 494