Advanced heart failure (HF) is associated with severe sleep-disordered breathing (SDB). In addition, most patients with HF are treated with an implantable cardioverter-defibrillator (ICD) for primary prevention of sudden cardiac death. The incidence of ICD therapy in such a patient cohort with SDB has never been investigated. The present study sought to determine the effect of SDB on the incidence of appropriate and inappropriate ICD therapy in patients with a categorical primary prevention ICD indication. A total of 133 consecutive ICD patients with New York Heart Association class II-III HF and depressed left ventricular function (≤35%) with no history of ventricular arrhythmia underwent a sleep study before ICD implantation and were followed for 24 ± 8 months, prospectively. A relevant SDB was defined as an apnea-hypopnea index of ≥10 events/hour. Of these 133 patients, 82 (62%) had SDB. Overweight (body mass index >29.1 vs 24.7 kg/m 2 ; p <0.001) was identified as the only independent risk factor for SDB. Appropriate ICD therapy intervention was significantly greater among patients with SDB than among patients without SDB (54% vs 34%, p = 0.03). Inappropriate ICD therapy intervention was documented more often in patients with SDB (n = 24 [29%] vs 7 [14%]; p = 0.04). An apnea-hypopnea index >10 events/hour was an independent predictor of appropriate ICD therapy on multivariate analysis (odds ratio 2.5, 95% confidence interval 1.8 to 4.04; p = 0.01). In conclusion, the present study is the first trial exploring the effect of SDB on the incidence of appropriate and inappropriate ICD therapy in patients with HF with a primary prevention indication. These results indicate that a preimplantation sleep study will identify patients with HF prone to receive appropriate and inappropriate ICD therapy.
Previous studies have demonstrated that a low left ventricular ejection fraction combined with either coronary artery disease or nonischemic cardiomyopathy poses a relevant risk factor for sudden cardiac death. However, many of these endangered patients never experience life-threatening events in their lifetimes. This leads to an enormous economic problem owing to the costs generated by implantable cardioverter-defibrillator (ICD) implantation for primary prevention and follow-up examinations. In addition, inappropriate therapy can result in unpleasant and potentially life-threatening events that, despite all efforts, occur in a relevant proportion of ICD patients. Sleep-disordered breathing (SDB) is common in patients with heart failure (HF) and seems to be associated with poorer survival and an increased risk of cardiovascular events. In particular, obstructive sleep apnea plays an important role in the development and recurrence of atrial fibrillation. This arrhythmia, especially if rapidly conducted, poses an important risk factor for acute cardiac decompensation in patients with HF and can also be interpreted as a surrogate of increased morbidity and mortality in such endangered subjects. Some previous studies have investigated the association of SDB and the incidence of malignant ventricular arrhythmias in ICD collectives enrolling patients with ischemic and nonischemic cardiomyopathy with both primary and secondary prevention indications. Recently, Bitter et al reported that the presence of SDB in patients with cardiac resynchronization therapy and ICD poses an independent risk factor for malignant arrhythmias.
Methods
The present prospective observational study focused in its primary end point on the investigation of the effect of SBD on the occurrence of appropriate ICD therapy for any ventricular tachyarrhythmia (VA). As a secondary end point, we evaluated the influence of several co-morbidities on the incidence of SDB. Another secondary end point tried to discover the influence of SDB on the incidence of inappropriate ICD interventions. Finally, we compared the period until the occurrence of the first appropriate and inappropriate ICD intervention stratified by the presence of SDB.
Patients with pre-existing ventilation therapy for SDB were excluded, and no SDB treatment was initiated during the study period. A total of 133 consecutive patients were enrolled from 2006 to 2009 (median follow-up period 24 ± 8 months). Patients receiving cardiac resynchronization therapy were excluded because of the potential reverse remodeling after such therapy. Thus, the potential change in left ventricular function by cardiac resynchronization therapy can influence the presence of SDB. Each patient was followed quarterly in the outpatient clinic of our university hospital. All patients went through an overnight sleep study (Embletta, Embla, The Netherlands), which was performed within 1 week before ICD implantation. The results of the sleep study were analyzed and reviewed by 2 independent SDB disorder experts. In accordance with the studies by Serizawa et al and Zeidan-Shwiri et al, SDB was defined as an apnea-hypopnea index (AHI) of ≥10 events/hour. A 90% reduction from baseline to peak amplitude of the signal from nasal cannula that lasted ≥10 seconds was defined as apnea. Hypopnea was diagnosed when ≥30% reduction in flow and a ≥3% desaturation from pre-event baseline occurred for >10 seconds. In addition, the Epworth Sleepiness Scale was obtained from each patient. For the exploration of the primary and secondary end points, the following baseline variables were chosen for uni- and multivariate analyses: age, ejection fraction, underlying heart disease, New York Heart Association class, history of atrial fibrillation, former myocardial infarction, history of hypertension, presence of diabetes mellitus, serum creatinine, Epworth Sleepiness Scale, body mass index, and medication.
Appropriate ICD therapy events were defined as successful termination of a sustained ventricular tachycardia or ventricular fibrillation by antitachycardia pacing or shock therapy.
The device programming of the included patients concerning the detection of VA in accordance with documented AHI was not different in patients with and without SDB ( Table 1 ).
| Variable | AHI ≥10 Events/hr Threshold (n = 82) ∗ | AHI <10 Events/hr Threshold (n = 51) ∗ | p Value |
|---|---|---|---|
| Ventricular tachycardia detection (beats/min) | 160 ± 27 | 162 ± 18 | 0.45 |
| Fast ventricular tachycardia by ventricular fibrillation (beats/min) ∗ | 200 ± 9 | 200 ± 12 | 0.64 |
| Ventricular fibrillation detection (beats/min) ∗ | 250 ± 4 | 245 ± 6 | 0.77 |
The triggers for inappropriate therapy were defined as either atrial fibrillation or other supraventricular tachycardia such as sinus tachycardia. T-wave oversensing, lead failures, or other external causes of oversensing resulting in inappropriate ICD therapy were withheld from the analysis, consciously. These device-related triggers were not likely to be influenced by the physical condition of the ICD recipient. Thus, they were judged irrelevant for the analysis, because their appearance could not be influenced by the presence or absence of SDB. To enrich the quality of the analysis of each ICD therapy, every episode was reviewed by 2 ICD experts, independently. In the case of a disagreement, a third reviewer was consulted, and the majority vote of these 3 reviewers posed the final classification of the episode.
Statistical analyses were performed using the SPSS, version 12.0 system (SPSS, Chicago, Illinois). We applied univariate and multivariate analyses to identify the risk of receiving either an appropriate or an inappropriate ICD intervention according to the presence of a relevant SDB during follow-up. The same analyses were performed to discover the risk of experiencing a relevant SDB disorder. In addition, we conducted a Kaplan-Meier analysis displaying the first appearance of an appropriate and inappropriate ICD intervention in any enrolled patient in terms of the presence of an SDB disorder.
Results
During a median follow-up of 24 ± 8 months, 133 consecutive primary prevention ICD recipients with New York Heart Association class II-III HF participated in the present study. Their demographics are listed in Table 2 .
| Variable | AHI ≥10 Events/hr (n = 82) | AHI <10 Events/hr (n = 51) | p Value |
|---|---|---|---|
| Age (yrs) | 63 ± 11 | 61 ± 12 | 0.23 ∗ |
| Left ventricular ejection fraction | 26 ± 5 | 28 ± 6 | 0.71 ∗ |
| Hypertension | 58 (71%) | 27 (53%) | 0.03 † |
| Diabetes mellitus | 27 (33%) | 11 (21%) | 0.14 † |
| Atrial fibrillation | 24 (30%) | 14 (27%) | 0.84 † |
| New York Heart Association class | 0.9 † | ||
| II | 21 (26%) | 12 (24%) | |
| III | 61 (74%) | 39 (76%) | |
| Coronary artery disease | 56 (68%) | 39 (76%) | |
| Nonischemic cardiomyopathy | 26 (32%) | 12 (24%) | 0.56 † |
| Serum creatinine (mg/dl) | 1.26 ± 0.4 | 1.18 ± 0.3 | 0.41 ∗ |
| Myocardial infarction | 47 (34%) | 34 (25%) | 0.66 † |
| Body mass index (kg/m 2 ) | 29.1 ± 1.4 | 24.7 ± 2.1 | <0.001 ∗ |
| Apnea-hypopnea index | 23 ± 8 | 4.5 ± 2 | <0.001 ∗ |
| β Blocker | 80 (98%) | 50 (98%) | 0.63 † |
| Angiotensin-converting enzyme inhibitors/angiotensin receptor blockers | 80 (98%) | 49 (96%) | 0.85 † |
| Epworth sleepiness scale (points) | 6 ± 4.4 | 6 ± 4.5 | 0.97 ∗ |
We diagnosed SDB in 82 patients (62% of the whole collective). Appropriate ICD therapies for VA were significantly greater among patients with SDB than among patients without SDB (54% vs 34%, p = 0.03; Figure 1 ). The presence of SDB was the only significant univariate predictor of the occurrence of an appropriate ICD therapy during follow-up ( Table 3 ). Moreover, the multivariate analysis revealed that an AHI ≥10 events/hour was the only independent predictor of an appropriate ICD therapy (odds ratio 2.5, 95% confidence interval 1.8 to 4.04; p = 0.01).
| Variable | OR | 95% CI | p Value |
|---|---|---|---|
| Pathologic finding on apnea-hypopnea index (≥10) | 2.34 | 1.7–4.97 | 0.02 ∗ |
| Left ventricular ejection fraction (per 5% change) | 0.8 | 0.25–2.59 | 0.71 ∗ |
| Hypertension | 1.35 | 0.64–2.87 | 0.43 † |
| Diabetes mellitus | 1.1 | 0.51–2.43 | 0.8 † |
| Nonischemic cardiomyopathy | 1.37 | 0.62–3.0 | 0.43 † |
| Coronary artery disease | 1.44 | 0.72–2.89 | 0.54 † |
| Atrial fibrillation | 1.53 | 0.7–3.37 | 0.29 † |
| New York Heart Association class | 1.18 | 0.52–2.65 | 0.7 † |
| Serum creatinine (per 0.2 mg/dl change) | 0.92 | 0.79–1.07 | 0.28 ∗ |
| Myocardial infarction | 0.99 | 0.48–2.04 | 0.96 † |
| Body mass index (kg/m 2 ) | 1.13 | 0.85–1.51 | 0.41 ∗ |
| β Blocker | 0.87 | 0.12–6.4 | 0.9 † |
| Angiotensin-converting enzyme inhibitors/angiotensin receptor blockers | 0.43 | 0.04–4.87 | 0.5 † |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
