Demographics
There were no baseline differences between HD patients and controls in age, sex, prevalence of atrial fibrillation/flutter, prevalence of heart failure (HF), cardiac medication regimen (beta-blocker, calcium channel blocker, antiarrhythmic drug, or digoxin use), device type, most recent EF, most recent thyroid function testing, or levothyroxine use. However, HD patients were more likely to have OSA (55% vs 9.1%, p = 0.001,Table 1 ). Additionally, the control group had a higher number of white patients compared to the HD group (HD 10/22 vs control 18/22, p = 0.01, Table 1 ). Of patients on HD, 12 (54.5%) were maintained on a Monday-Wednesday-Friday schedule, 9 (40.9%) on a Tuesday-Thursday-Saturday schedule, and one patient (4.5%) was maintained on a specialized home HD schedule (five HD sessions per week, three hours per session). The mean dialysis vintage was 3.36 ± 2.5 years. There were no differences in baseline characteristics between HD patients who did and did not suffer a VA event. Finally, there were no differences between HD patients and controls in prevalence of device type. Among patients with ICDs, no differences existed between groups in device indication (i.e. equivalent rates of primary and secondary prevention ICDs, Table 2 ).
In total, HD patients experienced 206 events, while control patients experienced 164 (Figure 1 ). Of the 206 total events in the HD group, 168 were NSVT (81.6%), 2 were VT (0.97%), and 36 were VF (17.5%). Of the 164 total events in the control group, 142 were NSVT (86.6%), 1 was VT (0.6%), and 21 were VF (12.8%). There was no difference between groups in the proportion of patients experiencing any VA overall (HD 45.4% vs control 63.6%, p = 0.226). There were no differences between groups in the adjusted overall event rate (HD 9.81 x 10-5 ± 1.5 x 10-3events/patient-hours vs control 3.71 x 10-5 ± 9.1 x 10-4 events/patient-hours, p = 0.902), NSVT adjusted event rate (8.00 x 10-5 ± 9.7 x 10-4vs 3.58 x 10-5 ± 8.7 x 10-4, p = 0.910), VT event rate (5.35 x 10-5 ± 0 vs 4.29 x 10-5 ± 0, p = N/A), or VF event rate (9.48 x 10-4 ± 8.1 x 10-4 vs 1.29 x 10-4 ± 8.4 x 10-5, p = 0.154). There were no differences in device indication between groups (Table 2 ).
Though we found a similar rate of VA events in both groups, the control group had a higher incidence of VA requiring therapies. As stated, HD patients experienced a total of 38 VT/VF events, compared to 22 events in the control group. Despite experiencing fewer numbers of VT/VF events, control patients required significantly higher rates of device therapy in the form of both ATP episodes and ICD shocks (total ATP episodes 2/38 in HD vs 10/22 in controls, p = 0.0002, total ICD shocks 10/38 in HD vs 17/22 in controls, p = 0.0001, Table 2 ). The significance observed in these therapies was driven primarily by statistically-significant differences in VF episodes requiring device intervention (Table 2 ).