Advice
Clinical and technical evidence
Clinical and technical evidence
A literature search was carried out for this briefing in accordance with the published process and methods statement. This briefing includes the most relevant or best available published evidence relating to the clinical effectiveness of the technology. Further information about how the evidence for this briefing was selected is available on request by contacting mibs@nice.org.uk.
Published evidence
This briefing includes a summary of 1 systematic review and 5 studies that were published after the systematic review was done, including a total of 1,060 patients.
The systematic review (Borisenko, 2014a) included 38 studies (678 patients in total), most of which were retrospective case series. The 5 other studies (Hashmi, 2015; Loforte, 2014; Mohite, 2014; Sabashnikov, 2013; Takayama, 2014) were all retrospective case series and included a total of 382 patients. The evidence suggests that CentriMag or PediVAS can be used in different patient groups with heart failure, resulting in varying survival rates. These groups include patients with pre- or post-cardiotomy cardiogenic shock, graft failure or rejection after transplant, and right ventricular failure after LVAD placement. However, there is no evidence comparing CentriMag with other interventions. Adverse events were also seen, with a higher incidence of bleeding and thrombosis in children than adults. Device failure was rare.
Table 1 summarises the clinical evidence as well as its strengths and limitations.
Strengths and limitations of the evidence
Because the evidence is mostly retrospective and based on case series without a comparator, it is of low quality and inherently prone to bias. Some of the series include consecutive patients, but in others specific groups or types of patients were selected with little or no explicit criteria reported. For this reason it is difficult to ascertain how representative these populations are to an NHS setting, despite some of the studies being done in the UK. There may also be some overlap across studies in the patients included.
Few studies were done in children, and when they were separate analyses were not possible. Some patients were supported for longer than 30 days, and in most studies there was little description of the intervention or other supportive treatments used during the study. Results from studies that analysed data from over several years may not be applicable to current practice because of changes in patient management. Most of the studies included outcomes of success rates and survival, but there was little evidence on the levels of anticoagulant drug use or micro-emboli formation, which are outcomes stated by the manufacturer to be relevant to this device.
Table 1: Summary of clinical evidence
|
Study size, design and location |
Intervention |
Outcomes |
Strengths and limitations |
|
Systematic review and meta-analysis of 38 studies (case series/cohort studies) with 678 patients Argentina, Austria Croatia, Czech Republic, Germany, Italy, Netherlands, Poland, UK, US |
Magnetically levitated centrifugal pump; CentriMag or PediVAS (manufactured by Levitronix or Thoratec), for extracorporeal membrane oxygenation (excluded from briefing) or ventricular assistance. |
CentriMag can be used in different patient groups, including pre-cardiotomy, post-cardiac surgery cardiogenic shock, post-transplant graft failure or rejection, post-LVAD placement right ventricular failure. Survival rates and adverse event rates were reported; however comparative data were not identified. |
Limited by the quality of included studies, (majority were retrospective case series) and small number of studies for each indication. Includes 2 studies where the mean duration of support was more than 30 days. Presents an overall summary of study quality but not for individual studies. Statistical heterogeneity between studies not reported, but appears to be present on inspection of the forest plots. Sponsored by the CentriMag manufacturer. |
|
Retrospective case series with 50 patients Two centres Italy |
Levitronix CentriMag. Surgical placement reported to be traditional (reference provided) and anticoagulant protocols proposed by the manufacturer adopted. No further details. Mean support time 10.2 (SD 6.6) days (range: 3 to 43 days). LVAD: n=12 RVAD: n=24 BiVAD: n=14. No comparator. |
The CentriMag led to successful outcomes for approximately half of those supported with refractory heart failure. Adverse events during the period of support were low, although deaths occurred in approximately one third of patients. |
Retrospective case series, unclear if cases were consecutive or any missing. CentriMag used during period of 2004 to December 2012 (9 years) and there may have been changes in patient management over this time frame which will not be accounted for in the study results. Some inconsistencies in results between text and tables. Unclear if representative population, few characteristics reported. No conflicts of interest from authors. |
|
Retrospective case series with 31 patients (of 154 total implanted) Single centre UK |
CentriMag, mean duration of support 11.7 days (range 0-65). 17 (54.8%) supported for > 7 days, of which 3 (9.7%) supported for > 30 days. LVAD 64.5%, RVAS 9.7%, BiVAD 28.8% Anticoagulation protocols, target flow, and monitoring of patients described. No comparator. |
CentriMag can be used for patients with post-cardiotomy cardiogenic shock as a bridge to decision. Over half of patients died on support. |
Limited to post-cardiotomy cardiogenic shock indication. Retrospective study of prospectively collected data. Focus of paper is on comparison between survivors and non-survivors and risk factors for surviving. Study period from 2004 to 2011. There may have been changes in patient management over this time frame which will not be accounted for in the study results. No conflicts of interest from authors. |
|
Retrospective case series with 143 patients (158 device implants) Single centre US |
CentriMag. Median duration of support: 14 days (IQR, 8–26). 158 device runs in 143 patients: BiVAD 66.5%, RVAD 26.0%, LVAD 7.5% Anticoagulant protocols described. No comparator. |
Bridge to decision therapy with surgical CentriMag VAD is feasible in a variety of refractory cardiogenic shock settings, with a third to up to three quarters of patients surviving to discharge. Neurologic complications occurred in 7-17% of patients. |
Retrospective case series, although states that cases were consecutive. CentriMag used during period of 2007 to 2012 and there may have been changes in patient management over this time frame which will not be accounted for in the study results. Some patients received devices on more than one occasion. Rates presented for type of device (e.g. LVAD, BiVAD) were based on number of device runs rather than number of participants. 2 authors received consultant fees from Thoratec. |
|
Retrospective case series with 33 patients Single centre UK |
CentriMag. Median duration of support 26 days (2-110). No comparator. |
A third of patients died on support but over half of patients were alive at 10 years. |
Published as conference abstract only, limited details reported. Retrospective case series, although states consecutive cases. CentriMag used during period of 2005 to 2014 and there may have been changes in patient management over this time frame which will not be accounted for in the study results. Range of support duration was over the 30 days indication (up to 110 days). Includes 2 patients having ECMO. |
|
Retrospective case series with 125 patients Single centre UK |
CentriMag as bridge to recovery, long‑term VAD or transplantation. No comparator. |
Around a third to two thirds of the different patient groups were weaned from support; the remaining patients died on support. |
Published as a conference abstract only, limited details reported. Retrospective case series, although states that cases were consecutive. Updated data from a subgroup of these patients are reported in Mohite, 2014. CentriMag used during period of 2003 to 2011 and there may have been changes in patient management over this time frame which will not be accounted for in the study results. Some patients had devices more than once. Some proportions were based on number of devices rather than number of participants. |