This review looked at the use of ventricular assist devices in people with end-stage heart failure. The authors found that the level of evidence available was limited but, for adults waiting for heart transplant, the devices appeared to have a beneficial effect on survival rates. The evidence regarding children was more limited, but the results appeared similar to those for adults.

To assess the evidence on the effectiveness and cost-effectiveness of using implantable ventricular assist devices (VADs) in the treatment of end-stage heart failure.

The Cochrane Library and the INAHTA database were searched. The search of a prior published systematic review (see Other Publications of Related Interest) was updated by searching MEDLINE and EMBASE from January 2000 to January 2004. In addition, the websites of Health Canada, FDA, Organ Donation Ontario, United Network for Organ Sharing, the Agency for Healthcare Research and Quality and manufacturers of left VADs were checked, and the reference lists of identified reviews were scanned. Health Canada and Organ Donation Ontario were also contacted for further information. The authors stated that studies had to be published in English language journals, although some studies in French also appear to have been included.

Study designs of evaluations included in the review

Studies were eligible for inclusion if the study design and methods were clearly described and they were systematic reviews, randomised controlled trials (RCTs), non-randomised controlled studies, or cohort studies with at least 20 participants. Any paediatric or Canadian studies, regardless of sample size, were also eligible. The studies in the review also included case-series (i.e. non-controlled studies).

Specific interventions included in the review

The interventions of interest were the use of Canada-licensed left VADs or biventricular assist devices as: a temporary bridge to heart transplant; a short-term bridge to recovery; or as an alternative to transplant. The comparator group had optimal medical management, including inotropes, but no VAD implantation. The devices used in the included studies were Novarcor LVAS, Heartmate (implantable pneumatic and vented electric) and Thoratec VAD. Specifically excluded, because they were not licensed in Canada, were studies on other named devices (ABIOMED, LionHeart, Jarvik-7, Cardiowest, Abiocor, HeartSaver VAD and Berlin Heart VAD systems).

Participants included in the review

The inclusion criteria were:

(1) people accepted for heart transplant who were refractory to aggressive medical treatment, had severe heart failure New York Heart Association (NYHA) class III/IV and a highly expected impending mortality;

(2) people with post-cardiotomy or other cardiogenic shock, hibernating myocardium, acute heart failure, or other conditions where the patient is expected to recover sufficiently to be weaned off the device;

(3) people with heart failure similar to (1) above, who were expected to receive the device but not go on to transplant.

Studies on children or adults were eligible. In the included studies on children, their mean ages ranged from 4.9 to 16 years.

Outcomes assessed in the review

The primary outcome of interest was the survival rate on VAD support, the survival rate to transplant, or survival to recovery. The secondary outcomes were impact on NYHA functional classification, quality of life (QOL) including ambulatory status, device-related adverse events and impact on transplant survival rates.

How were decisions on the relevance of primary studies made?

One reviewer read the identified abstracts to assess eligibility. The full text of eligible articles were then reviewed to confirm eligibility.

The studies were classified according to levels of evidence, based on Goodman. The scale ranged from level 1 for large RCTs or systematic reviews of RCTs, to level 4g for unpublished case series. The authors also discussed the quality of studies within the narrative.

The authors did not state how the data were extracted for the review, or how many reviewers performed the data extraction. Data were extracted on study methodology and outcomes, and on inclusion and exclusion criteria for the comparative studies.

How were the studies combined?

The studies were described in tables and discussed in the narrative, grouped by aim and by study methodology.

How were differences between studies investigated?

Differences between the studies were described in tables and discussed in the narrative.

It appears that 1 RCT (129 participants), 10 comparative studies (2,099 participants) 42 case series (2,004+ participants), and 6 case series on children and adolescents (186 participants) were included.

The level of evidence identified was limited as, with one exception, all evidence came from observational studies.

Bridge-to-transplant (10 comparative studies and 33 case series).

Survival rates after a left VAD implant ranged from 58 to 90%. The percentage going on to have a transplant ranged from 39 to 90%. The evidence suggested that post-transplant survival rates after the use of a left VAD were similar or better than those treated with medical management including inotropes. The left VAD appeared to improve NYHA functional classification and quality of life.

Serious adverse events included infection (median 53%, range: 6 to 72), bleeding (median 35%, range: 8.6 to 48%), thromboembolism (range: 5 to 37%), neurological disorders (range: 7 to 28%), right ventricular failure (range: 11 to 26%), and haemolysis (range: 6 to 20%). Bleeding tended to occur in the first few post-implant days, but infection and thromboembolism occurred at any time. The malfunction rate of the devices was approximately 1.55 per patient in the first year and required back-up components in 9% of cases.

Bridge-to-recovery (9 case series, 328 participants).

The percentage of patients weaned from left VAD and surviving discharge ranged from 0 to 36% (median 26). Fourteen to 50% went on to transplant directly from left VAD or required transplant after weaning from the device.

The adverse events were similar to those associated with bridge-to-transplant patients.

Destination therapy (1 RCT, 129 participants).

Long-term left VAD use was associated with a 48% relative reduction and 27% absolute reduction in mortality at 1 year. The morbidity associated with the left VAD was considerable, with adverse events occurring 2.35 times more than in the controls (6.45 per patient-year for left VAD and 2.75 per patient-year for medical treatment). One-year Kaplan-Meier (KM) estimated survival was 52% for left VAD and 25% for medical therapy; KM survival at 2 years was 23% and 8%, respectively (P=0.09). Longer term (more than 2 years) outcomes were not available.

Left VAD in children and adolescents (6 studies, 186 participants).

Small devices designed for adults have been successfully used in paediatric patients awaiting transplant, but their use in smaller patients presents problems with fitting such a device. The survival rate was comparable to those reported for adults (68.9% survived to transplant or recovery).

The authors summarised literature on the costs and cost-effectiveness of left VADs. They concluded that reports showed that left VADs are costly, mainly because of the high cost of the devices (Can$90,000 to Can$98,000s each) and the costs associated with initial hospitalisation and rehospitalisation for complications or device malfunction. The estimated cost-effectiveness ratio for the elective use of left VADs as a bridge to transplant in Canada is Can$91,000 to Can$117,000 per quality-adjusted life-year.

For adults waiting for heart transplant, left VAD support has been shown to improve the survival rates, functional status and quality of life. Only limited evidence was available for children, but the survival rates appeared to be similar to those for adults.

The aims of this review were clearly stated. The limitation of the search to studies published in English and French might have led to publication bias. The authors considered the quality of the studies and used a narrative synthesis to describe the overall results. This was appropriate given the variation in the studies included. While the inclusion criteria stated that only controlled studies would be included, much of the evidence presented came from case series. Information on the included studies was tabulated, although there was some discrepancy between the number of studies described in the narrative and that presented in the tables. This meant that it was difficult to calculate how many people were involved in the reviewed studies. In addition, one included study was not referenced. Given that the authors acknowledged that the evidence they found, from observational studies, is not as reliable as that from RCTs, their conclusions appear to reflect the data presented.

Practice: The authors implied that although the use of left VADs appears to improve survival in practice, because of a shortage of donors, this would result in longer waiting lists for heart transplants. They also stated that the introduction of left VADs in the paediatric population would be more cost- effective and might not have negative effects on transplant waiting lists.

Research: The authors did not state any implications for further research.

Agence Nationale d'Accreditation et d'Evaluation en Sante (ANAES). Evaluation de l'assistance ventriculaire en attente ou en alternative a la transplantation cardiaque. Paris: ANAES; 2001.

This is a critical abstract of a systematic review that meets the criteria for inclusion on DARE. Each critical abstract contains a brief summary of the review methods, results and conclusions followed by a detailed critical assessment on the reliability of the review and the conclusions drawn.