Implantable cardioverter-defibrillators (ICDs) do not improve overall long-term survival compared to medical treatment in patients with non-ischaemic systolic heart failure, according to results of the DANISH clinical trial. These data prompted researchers and other physicians at the European Society of Cardiology Congress (ESC; 27‒31 August, Rome, Italy) to suggest the need for more accurate selection of non-ischaemic patients for ICD implantation that takes into consideration age and comorbidities.
The trial, presented at a Hot Line session of ESC by study investigator Lars Kober (Rigshospitalet, Copenhagen University Hospital in Copenhagen, Denmark), however indicated that the risk of sudden cardiac death is reduced by half with ICD placement and observed in a subgroup of young patients (>68 years) a reduction of total mortality by 36% in the ICD arm compared to the control arm.
Current American Heart Association guidelines include ICD implantation as a class 1A recommendation for all-cause mortality in ischaemic and non-ischaemic patients with symptomatic systolic heart failure, while European guidelines have a class 1B recommendation specifically for patients with non-ischaemic heart failure. These recommendations have been based on small to medium-sized trials with neutral outcomes and subgroup analyses on larger trials. “The evidence is much weaker for patients with non-ischaemic aetiology,” said Kober; therefore, “our trial fills that gap by suggesting ICDs should not be routinely implanted in all patients with systolic heart failure.”
The DANISH (The Danish study to assess the efficacy of ICDs in patients with non-ischaemic systolic heart failure on mortality) study, which was simultaneously published in the New England Journal of Medicine (NEJM), was an investigator-initiated, multicentre, randomised, unblinded, controlled clinical trial conducted at five centres in Denmark.
From February 2008 to June 2014, the trial enrolled 1,116 stable patients with chronic, non-ischaemic, symptomatic heart failure (556 patients were randomly assigned to ICD implantation and 560 to the control group, which included guideline-recommended medication including beta blockers, renin-angiotensin inhibitors, and mineralocorticoid-receptor antagonists). In total, 58% of patients in both groups received cardiac resynchronisation therapy (CRT). This was delivered via biventricular pacemakers in the control arm, and via a device combining CRT and ICD in the ICD arm. Patients were followed-up for a median time of 67.6 months.
Kober reported that the primary outcome of death from all causes occurred in 21.6% of the ICD patients and 23.4% of the controls ‒ a non-significant difference (hazard ratio [HR] 0.87; 95% confidence interval 0.68 to 1.12; p=0.28). The secondary outcome of sudden cardiac death occurred in 4.3% of the ICD patients and in 8.2% in the control group (HR 0.50; 95% CI 0.31 to 0.82; P=0.01).
These results were independent of whether or not a patient received a CRT device, noted Kober, and added that there was an important interaction with age. “Patients younger than 68 years of age had a significant reduction in all-cause mortality if they received an ICD (HR 0.64; 95% CI 0.45 to 0.90, P=0.01), suggesting that younger patients may have a survival benefit with ICD implantation,” he said.
John McMurray (University of Glasgow, Glasgow, UK) interpreted the results of the study in an accompanying editorial. He explained that unlike previous ICD trials, the combination of medical treatment—using angiotensin-converting-enzyme inhibitors or angiotensin-receptor blockers and beta-blockers in almost all patients and mineralcorticoid-receptor antagonist in nearly 60% of patients—plus CRT therapy in 58% of patients contributed to the risk reduction of premature cardiovascular death, including the risk of sudden cardiac death “very substantially”.
Best non-ischaemic candidates to receive ICDs
In NEJM, Kober et al commented that 31% of deaths in the study were attributed to non-cardiovascular causes. “This is not surprising in the elderly population, but it highlights the importance of selecting patients for ICD implantation carefully,” they wrote. “The rate of death among patients we included in our trial was relatively low, and patients at higher risk may be more likely to benefit from ICD implantation.” Additionally, “the subgroup data that suggest a lower likelihood of benefit in older patients might be used as an argument for not implanting ICDs in frail patients.”
The authors also noted that given the benefit of ICD implantation with respect to sudden cardiac death, “patients who are not expected to die from other causes may be good candidates for ICD implantation.”
Is it time to revisit the guidelines?
In light of the DANISH trial results, physicians at ESC had various reactions regarding modifications of the current guidelines:
Kober commented: “Guidelines are based on multiple studies and ICD treatment should still have a class 1A recommendation for prevention of sudden cardiac death in non-ischaemic heart failure also. However, patients with a high risk of non-sudden death may not benefit, and age could be an important factor in the decision to give an ICD, along with comorbidities.”
Discussant of the trial, Cristoph Leclerq (Centre Cardio-Pneumologique, Rennes, France), said: “The DANISH trial is a very important study which raises the question of revisiting the guidelines for primary prevention of sudden cardiac death in non-ischaemic cardiomyopathy. Clearly, we need a more accurate selection of non-ischaemic patients before implanting an ICD, as suggested by the DANISH trial including age (<70 years), probably comorbidities and the use of imaging techniques.”
Michel Komajda (Pitié-Salpétrière Hospital, Paris, France), who is involved in the European guidelines development said: “I would add an amendment to the guidelines. Of course the primary outcome (of DANISH trial) was neutral, but I think that the reduction in sudden cardiac death, which is the primary objective of implanting an ICD, is quite significant so from an ethical point of view this should be taken into consideration.”
Cecilia Linde (Karolinska University Hospital Solna, Stockholm, Sweden) who gave a round-up of the key highlights on the arrhythmias subject at ESC, had a more neutral consideration, she told Cardiac Rhythm News: “We have to carefully analyse the details of previous studies (such as DEFINITE and SCD-HeFT) and the DANISH study first to answer the question if guidelines revisions are necessary”.
The European Heart Rhythm Association (EHRA) will launch a similar trial in ischaemic cardiomyopathy, the RESET SCD trial, which will include 2,000 patients with a follow-up of five years and the primary endpoint is total mortality.
The DANISH trial was supported by unrestricted grants from Medtronic, St Jude Medical, TrygFonden and the Danish Heart Foundation.