The important contribution of psychological factors to mortality and morbidity associated with ischaemic heart disease (IHD) is now well recognised. In a review of evidence-based cardiology, Hemingway & Marmot (1999) concluded that 'prospective cohort studies provide strong evidence that psychosocial factors, particularly depression and social support, are independent aetiological and prognostic factors for coronary heart disease.' The prevalence rates of major depressive illness in patients with IHD are three times that in the normal population. Cardiac mortality rates are increased by between three and four times for patients who are depressed post myocardial infarction (MI) compared with patients who have had a MI but are not depressed. Indeed the prognostic impact of depressive illness is comparable to that of degrees of ventricular dysfunction or coronary atherosclerosis in patients who have had a heart attack. Half of all cases of depression post-MI will remit spontaneously, with the other half either persisting or remitting followed by relapse within the year (Glassman, 2002).
Carney et al. (2002) has usefully summarised the literature with respect to the potential pathophysiological factors implicated. These can be separated into indirect (alterations in health-related behaviours such as compliance with treatments and smoking cessation) and direct mechanisms (including evidence of reduced heart rate variability in patients with IHD who are depressed, enhanced platelet reactivity and inflammatory processes). This suggests a plausible role for psychotherapy in helping to reduce the mortality and morbidity associated with cardiac disease, by focusing on behavioural interventions and/or the reduction of psychological distress.
Prior to the eagerly awaited outcome of the Enhancing Recovery in Coronary Heart Disease Patients (ENRICHD) trial, detailed below, much of the literature had relied on cardiac nurse-led psychosocial interventions (often termed psychological interventions or counselling), with variable outcomes. This led to conflicting views about the efficacy of addressing psychological issues in cardiac patients. Linden's (2000) review of psychological treatments in cardiac rehabilitation programmes commented on the differing treatment approaches ranging from relaxation/breathing retraining, unstructured support or psychoeducation to improve compliance, to psychological interventions to reduce emotional distress. He concluded that critical differences in study outcomes can be explained by the finding that studies which failed to have an impact on levels of psychological distress also fail to have an impact on mortality or event recurrence.
Most of the reliable evidence for psychological interventions involves cognitive behavioural therapy (CBT) or behavioural therapy. However, even the CBT used was described as 'broadly defined' by Bennett & Carroll (1994) in a review of cognitive-behavioural interventions in cardiac rehabilitation. They commented on the difficulty of reaching definite conclusions in an area so beset by methodological weaknesses. These include problems with differing measures of psychological distress and variable outcomes, as well as the application of treatments irrespective of baseline measures of psychological distress - the so-called floor effect. The random assignment of all cardiac patients to psychological treatment without an assessment of actual psychological need is unlikely to be cost effective. It is also known that individually tailored treatment is more effective than standardised treatment packages.
Many studies include very small subject numbers, which are particularly unimpressive to cardiologists who are used to large international trials of thrombolysis incorporating 50 000 subjects. In an attempt to address this, there have been four notable meta-analytic reviews of efficacy of psychological interventions additional to usual treatment in patients with IHD.
• The focus of earlier studies in the 1980s was on Type A behaviour (TAB), a personality trait characterised by competitive, driven behaviour, hostility and impatience, as a risk factor for IHD. Nunes, Frank & Kornfeld (1987) carried out a meta-analysis of modification of TAB in 18 studies and noted that longer, more complex treatments produced more significant impact on mortality and morbidity rates up to three years after a MI. The most significant of these was the Recurrent Coronary Prevention Project (Friedman et al., 1986), which randomly assigned patients to an educationally based cardiac counselling group (n = 270) or a Type A counselling group (n = 592). The Type A counselling, which combined didactic, CBT and existential humanistic components, resulted in a reduced coronary recurrence rate of 7 % compared to 13 % in the control group over a three-year follow-up. Of particular note was the finding that those who showed a reduction of TAB, irrespective of group allocation, had one-fifth of the cardiac recurrence rate of those who had no significant reduction in TAB. Subsequent studies have, however, failed to replicate this association, tending to focus instead on hostility as the pathogenic component ofTAB.
• Linden, Stossel & Maurice (1996) then looked at psychological interventions (mainly 'broadly defined CBT') in a meta-analysis of 23 randomised controlled trials (RCTs) of cardiac rehabilitation and reported a positive impact on mortality and morbidity rates as well as a reduction in depression and anxiety levels.
• Dusseldorp et al. (1999) were less convinced by their further meta-analysis, based on 37 RCTs. However this included two large RCTs the first of which, by Jones & West (1996), involved 2 328 patients, most of whom were neither anxious nor depressed. The second, by Frasure-Smith et al. (1997) 1 376 patients were treated by cardiac nurses with no psychotherapy training using less sophisticated forms of psychological interventions with, not surprisingly, much less impact. Of particular relevance is the finding that mortality and MI recurrence rates were only reduced in those studies that also reduced levels of psychological distress.
• A Cochrane systematic review of the effectiveness of psychological interventions for patients with coronary heart disease has suggested that the reviews to date had not been fully systematic (Rees etal., 2004). Some had included non-randomised trials and failed to differentiate sufficiently between types of psychosocial interventions. The Cochrane group included 36 trials of psychological interventions in adults with coronary heart disease that had a minimum follow-up of six months, as well as identifying a subgroup of studies of 'stress management'. They noted extreme heterogeneity of the trials but were unable to carry out a planned stratified analysis of the data, due to a combination of insufficient trial information and numbers. Overall the outcomes of the 'stress management' subgroup were similar to all psychological interventions. They reported no evidence of reduced total mortality, but found some evidence of a reduction in non-fatal reinfarction in the intervention group. They caution that this apparent effect may be a result of ascertainment or publication bias, pointing out that the two largest studies (Berkman et al., 2003; Jones & West, 1996) had shown no effect. Psychological outcomes were reported less often in the analysed trials. They concluded that there were small but significant effects on anxiety and a significant reduction in depression with psychological intervention.
The recently completed Enhancing Recovery in Coronary Heart Disease Patients (ENRICHD) trial is, however, at the forefront of current research in this area (Berkman etal., 2003). The randomisation of 2481 patients, enrolled within 28 days of acute myocardial infarction from eight clinical centres, into CBT versus usual care over six months makes it the largest controlled trial of psychotherapy ever completed. Patients (1 084 women, 1 397 men) were included if they met criteria for a DSM-IV diagnosis of depression or dysthymia (~ 75 %) or had low perceived social support (LPSS) (~ 25 %). To maximise cost-effectiveness, CBT and social learning interventions were given as a combination of individual (one hour weekly up to a maximum of six months) and group therapy (two hours weekly for 12 weeks). Treatment continued until the patient completed a minimum of six sessions, had developed self-therapy skills and had scored below seven on the BDI on two consecutive occasions. The ENRICHD counsellors were intensively trained in CBT and supervised weekly by experienced CB therapists. Although the therapy was manualised and standardised, a flexible approach was used to optimise recruitment and participation. This included 'taking the treatment to the patient', offering transport, conducting sessions over the telephone and motivational interviewing techniques. Adjunctive pharmacotherapy (generally sertraline) was used to treat patients with severe depression or slow to responders after five weeks of treatment.
The primary aim was to determine the impact of treating depression and low perceived social support on all-cause mortality and reinfarction rates. Secondary end-points included depression, LPSS and cardiac morbidity, and they achieved 93 % follow-up rates. Disappointingly, the intervention had no significant impact on recurrent MI or death at 29 months follow-up. There was an improvement on psychosocial outcomes at 6 months but this was not maintained at longer term follow-up.
Much debate has ensued to explain this negative outcome. The trial was initially designed in 1994, when the study of social support received as much attention as the impact of depression in the cardiac population. The ENRICHD trial involved the development of a new screening tool and CBT intervention for LPSS. It has been suggested in hindsight that there is as yet insufficient understanding of the impact of social support on cardiac prognosis to have targeted this (Frasure-Smith & Lesperance, 2003). Study power may have been reduced by the explicit inclusion of women (44 %) and minorities (34 %), unique to this cardiac study, with later subgroup analysis suggesting that white men alone may have benefited from the intervention. The study effect size, differentiating the outcomes between the intervention and standard care groups, was based on the assumption of no treatment of depression in the controls. There was a greater improvement in depression scores in the control group than expected, in part surely related to the finding at 36 months that 21 % of the controls had been prescribed antidepressants, compared to 28 % of the intervention group. On reflection, the study power, based on finding a 30 % difference in mortality and reinfarction between the groups, is also believed to have been unrealistic.
Finally, the significant but disappointingly small impact on depression scores is worth considering in the context of the other recent study looking at the effect of (pharmacological) treatment of depression post MI or unstable angina (Glassman et al., 2002). The Sertraline Antidepressant Heart Attack Randomised trial (SADHART) of 369 depressed patients found no significant difference between 24 weeks double-blind treatment with sertraline and placebo. The authors recognised that screening patients soon after significant medical morbidity was likely to result in inclusion of milder, self-limiting depressive states, contributing to a significant placebo response. They therefore identified an a priori more severe depression group. Despite the smaller numbers, sertraline was found to be highly efficacious compared to placebo in this group. In an analysis of the ENRICHD study, Frasure-Smith & Lesperance (2003) suggest that a more aggressive, step-wise approach to treatment of patients who are depressed following MI may be necessary.
However this important study has proven that a large-scale multicentre trial of a standardised but individually tailored psychotherapy, requiring the close collaboration of psychologists, psychiatrists and cardiologists, is achievable. It has shown that most eligible patients can be successfully recruited with systematic screening and with little attrition on prolonged follow-up. Indeed, this trial is expected to remain a standard of comparison for many years (Frasure-Smith & Lesperance, 2003).
Although much of the research has focused on patients following MI, there is also some emerging evidence of benefit of CBT for patients with congestive cardiac failure, in reducing restenosis rates after angioplasty and in the treatment of recipients of implantable cardioverter defibrillators for ventricular dysrhythmias (Edelman & Lemon, 2003). A Cochrane systematic review attempting to assess the effects of psychological interventions for depression in patients with congenital heart disease failed to identify any RCTs in this area, despite evidence from cross-sectional studies of an association between congenital heart disease and depression (Lip et al., 2003).
There is now considerable agreement, supported by robust research evidence, that psychological factors (particularly depression and poor social support) are associated with increased mortality and morbidity in patients with coronary artery disease. However, the pathophysiological mechanisms underlying this remain unclear and the evidence for psy-chotherapeutic interventions in reducing adverse clinical outcomes is equivocal.
In clinical terms, a stepped approach to patient care is advisable. The Scottish National Clinical Guidelines Network (SIGN) publishes evidence-based recommendations that have been extrapolated in a standardised format from meta-analyses, systematic reviews and/or RCTs. Their 2002 national clinical guideline on cardiac rehabilitation recommends that a comprehensive cardiac rehabilitation programme, including psychological interventions, is offered to all patients with significant angina, congestive cardiac failure or following a MI. In addition they advise appropriate treatment of patients with significant psychological distress diagnosed on routine screening. Of particular note are their recommendations that 'Rehabilitation staff should identify and address health beliefs and cardiac misconceptions in patients with CHD. (Lincoln & Flannaghan 2003). Patients with moderate to severe psychological difficulties should be treated by staff with specialist training in techniques such as CBT.'
The American College of Cardiology/American Heart Association have updated their guidelines for the management of patients with MI and recommend that the long-term management includes an evaluation of the patient's psychosocial status, with a particular emphasis on depression, anxiety and levels of social support (Antman et al., 2004). They also recommend that treatment with CBT and selective serotonin reuptake inhibitors can be useful for patients who develop a depressive illness within a year of discharge from hospital.
Was this article helpful?
Do you suffer from a habit or a behavior or a repetitive thought pattern that keeps you from being who you want to be? Do you try to change this or that aspect of your life, but wind up right back where you started? You're not alone! Millions of Americans try to make changes, but the whopping majority fail exceptionally.