Statistical analysis
Categorical data are represented as proportions, continuous data as mean ± SD for normally distributed variables or median and interquartile range for non-normal distribution. Comparisons were made using Chi-Square Test, Fisher’s exact test, unpaired student T-test and Mann-Whitney test. Multivariable COX proportional hazard models were used to identify independent characteristics and medical treatment associated with VA or mortality. To assess the impact of VA on overall mortality a Cox model with time to first VA as a time dependent covariate was used. Unadjusted and adjusted Hazard ratios (HRs) with 95% confidence intervals (CIs) were displayed. All tests were two sided, p-values < 0.05 were considered statistically significant. Analyses were carried out using IBM SPSS Statistics for Windows, Version 25.0. Armonk, NY.
Results
There were 186 patients implanted with an ICD/CRTD between the years 2007-2017 that matched the study’s inclusion criteria (Figure 1). Their mean age was 66.4±12 years, 15.1% were female. ICD was implanted in 79 (42.5%) and a CRTD in 107 (57.5%). Median [IQR] follow-up time was 3.8 [2.1-6.7] years. Patient characteristics are shown in table 1. There were 52 (28%) patients with VA, including VT in 31/52 patients (59.5%), VF in 6/52 patients (11.5%) or both in 15/52 patients (29%). These VA cases were treated successfully by anti-tachycardia pacing (ATP) in 22 (42.4%) patients and by device shock in 30 patients (57.6%). There were 77 (41.4%) deaths during the study F/U period. The prevalence of HF medication treatment at index hospitalization discharge was: 155/186 (83.3%) BB, 162/186 (87.1%) AngA, and 110/186 (59.1%) MRA. AAD were prescribed in 81/186 (43.5%) patients. Doses (% target) of HF medication were: 32±25% for BB, 38.2±30% for AA and 31±30% for MRA. The median dose (% target dose) for all 3 guideline-based medication groups included in our study was 25% (Table 2). Few patients were prescribed with >50% of target dose: 18/155 (11.6%), 34/162 (21%), and 16/110 (14.5%) of patients taking >50% target dose of BB, AngA, and MRA, respectively (Table 2).
Only 18/186 (9.7%) of study patients were followed regularly in the hospital’s HF clinic by HF specialist (most patients were followed regularly by their general cardiologists and came to our hospital only for device clinic interrogations). There were more patients treated by BB among the group followed in HF clinic (100% vs 81.5%, p=0.046) and their dose (% target dose) was higher (61.1% vs. 33.9%, p=0.023). There was a non-significant trend for higher prevalence of AngA (88.9% vs. 86.9%, p=0.81) and MRA (72.2% vs. 57.7%, p=0.23) among those followed at HF clinic as well.