July 2011
Reviewer: Jeff C. Huffman, MD
A randomized controlled trial of cognitive behavioral therapy tailored to psychological adaptation to an implantable cardioverter defibrillator
Irvine J, Firestone J, Ong L, et al
Psychosom Med 2011;73(3):226-233
Background: Implantable cardioverter defibrillators (ICDs) are life-saving devices but can also lead to complicated psychological outcomes. Receiving shocks from an ICD can be startling, painful, and even traumatic, with patients who receive multiple shocks having elevated rates of depression, anxiety, and PTSD. Such psychological symptoms can be difficult to treat, since it is true that at any moment, a person with an ICD could experience a sudden and painful shock or series of shocks. Cognitive behavioral therapy (CBT) approaches in patients with ICDs to prevent or address these psychological symptoms have been attempted, but the studies of these interventions have been limited by sample size or other factors.
Methods: The authors recruited subjects from two hospitals between 2003-2006 who had a first ICD placement for secondary prevention of a sudden cardiac death event or — in those patients with hypertrophic cardiomyopathy — primary prevention from such an event. Subjects who received an ICD for primary prevention related to heart failure, ischemic cardiomyopathy, or ischemic heart disease were excluded because their “poor and declining heart failure might confound evaluation of the CBT.” Subjects with limited English fluency and cognitive impairment/psychosis were also excluded. Subjects were recruited in the hospital or at their first posthospital visit. Enrolling subjects were randomized to a CBT condition (consisting of a CBT manual, psychoeducational booklet, CD with muscle relaxation exercises, and 8 telephone counseling sessions, with additional ones for patients who received a shock in the intervention period) specifically designed for this population or to treatment as usual. Outcome measures, assessed at baseline and 6/12 months, included the Hospital Anxiety and Depression Scale, a phobic anxiety scale, a measure of PTSD symptoms, and health-related quality of life (SF-36).
Results: 680 ICD patients were screened; 292 met study eligibility criteria, and 193 consented (96 randomized to intervention, 97 to usual care). Overall, subjects in both conditions had improvements over the 12 month period on essentially all outcome measures. Subjects in the CBT arm had greater improvements of depressive symptoms and PTSD symptoms (total symptoms and avoidance subscale) and marginally significant effects on mental health-related quality of life (HRQoL); when examined by gender, the effects on depression and mental HRQoL appeared to be limited to women, for whom they resulted in significant improvements. The intervention had no effects on physical HRQoL or arrhythmia events.
Discussion: Overall, there is good news here. A CBT treatment has now been created for this clinically-challenging population, and when used, it does appear to reduce PTSD symptomatology and, at least in women, lead to less depression and better mental HRQoL. On the other hand, there were no robust effects on depression/mental HRQoL in the overall sample, and like nearly every psychosomatic intervention, there were no effects on physical HRQoL or hard medical outcomes.
One important aspect of this study is that it was used up-front in all ICD patients, rather than selecting those with depression/anxiety/PTSD symptoms for treatment. Though this use of a preventive intervention is interesting, it seems unlikely that in the real world resources would be provided to provide all ICD subjects with a comprehensive CBT intervention. More likely, those patients who had depression/anxiety (before or after getting shocks), and those with PTSD symptomatology following shocks, would be referred for treatment. Indeed, the impact of CBT vs. usual care may be far greater in this highly-symptomatic population, and the approach of taking “all comers” may have washed out some of the treatment’s effects.
Another important consideration is that, although the investigators took all patients regardless of psychological status, they were rather selective in taking patients based on reason for ICD placement, excluding those getting ICDs for many common indications, and indeed, only ~40% of ICD patients were eligible for the study, limiting generalizability. Finally, there was no attentional control—it is not clear whether CBT was better because it was a superior treatment, or because there was more attention/contact. It would have been very helpful to have had more information on the CBT intervention itself within this paper, though there appears to be a separate paper describing the intervention.
In sum, this is an important line of work for one of our most challenging populations. Further study specifically in clinically-distressed patients—the patients who would get the intervention in the real world—and in patients with a wide range of ICD indications, will tell us even more about the efficacy of this CBT intervention tailored to ICD patients and may get it more ready for real-world implementation. This well-designed trial is a good start.