Clinical and technical evidence

A literature search was carried out for this briefing in accordance with the interim 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

Published evidence

This briefing summarises 4 published studies on PleuraFlow Active Clearance Technology (ACT): 3 retrospective cohort studies (St-Onge et al. 2017, Maltais et al. 2016, Sirch et al. 2016) and 1 user experience study (Perrault et al. 2012), including a total of 827 adults having cardiac surgery in the US, Canada and Germany.

All 3 retrospective cohort studies showed a reduction in the rates of a composite outcome, 'retained blood syndrome', with patients having PleuraFlow ACT compared with standard chest drains. Two studies also showed a reduction in the rates of postoperative atrial fibrillation. Table 1 summarises the included studies as well as their individual strengths and limitations.

Overall assessment of the evidence

The evidence base on PleuraFlow ACT is still developing and currently limited in both quantity and quality. Most of the studies were non-randomised retrospective analyses of cohort studies, which compared the outcomes of different patient groups during consecutive time periods. Although the patients in these studies were not randomised, propensity match scoring or logistic regression was used to attempt to match individuals in the intervention and control group, control for confounding and mimic randomisation. Changes in chest tube management protocols and other variables were not controlled for. Moreover, none of the studies was done in the UK, which may limit their generalisability to the NHS.

Multicentre, randomised controlled trials and evidence from a UK setting comparing PleuraFlow ACT with standard chest drains would be useful. These should include relevant technical and clinical outcomes such as incidence of tube occlusion, postoperative atrial fibrillation and reintervention rates.

Table 1 Published evidence on PleuraFlow ACT

St-Onge et al. (2017)

Study size, design and location

Non-randomised retrospective cohort study of 300 patients having cardiac surgery in 2 separate time periods. Used propensity match scoring between groups.

Primary end point: POAF

Single centre in Canada.

Intervention and comparator(s)

People with PleuraFlow ACT (n=142).

People with standard chest drains (n=158).

Using a 1:1 propensity score match, 214 patients were included in paired analysis (107 in each group).

Key outcomes

Unmatched patients having PleuraFlow ACT had a 34% reduction in POAF rate compared with those having standard chest drains (23% vs 35%, p=0.01).

In the matched cohort, PleuraFlow ACT was associated with a 31% reduction in POAF rate (24% vs 35%, p=0.09) and a trend toward shorter postoperative length of stay (5.0 [4.0; 7.0] vs 6.0 [5.0; 8.0], p=0.08).

In multivariate analysis, chest drainage with PleuraFlow ACT showed a protective effect on POAF with odds ratio of 0.5 (95% CI 0.1 to 0.9; p=0.02).

Strengths and limitations

This was a non-randomised, single-centre study with interventions used at different time points. One of the authors has served on a company scientific advisory board and 2 others have received honorarium for scientific presentations.

Maltais et al. (2016)

Study size, design and location

Non-randomised, retrospective cohort analysis of prospectively collected data from 252 patients having cardiac surgery.

Primary end point: composite end point of additional interventions for RBS defined as: any re-exploration for bleeding, delayed sternal closure, pericardial interventions for drainage (pericardial window or pericardiocentesis), and pleural interventions for drainage for haemothorax or bloody effusions.

Single centre in the US.

Intervention and comparator(s)

Patients were divided into 2 groups for analysis:

Group 1 was patients having standard chest drains (n=77, April 2009 to May 2013). Group 2 was patients having PleuraFlow ACT according to a defined ICU use protocol (n=175, June 2013 to July 2015).

Key outcomes

Univariate and multivariate analyses were done, adjusting for the use of conventional sternotomy and minimally invasive left thoracotomy.

Univariate analysis showed a 65% reduction in re-exploration (p<0.001) and an 82% reduction in delayed sternal closure (p<0.001). In a subanalysis of conventional sternotomy only, there continued to be a significant 53% reduction in re-exploration (45% vs 21%, p=0.0011) and a 77% reduction in delayed sternal closure (35% vs 8%, p<0.001) in group 2.

Using a logistic regression model adjusting for conventional sternotomy versus minimally invasive left thoracotomy, there was a significant reduction in re-exploration (OR=0.44, 95% CI 0.23 to 0.85; p=0.014) and delayed sternal closure (OR=0.20, 95% CI 0.08 to 0.46; p<0.001) in group 2.

Other postoperative outcomes, such as POAF, length of stay, time in ICU and mortality were similar between groups.

Strengths and limitations

This was a non-randomised study with interventions used at different time points. Changes in ICU protocols over time and other variables were not controlled for. Patients were grouped according to their treatment and no other matching was done, but differences between groups were controlled for in analyses.

The study was partly funded by the company. One of the authors was a consultant and equity holder in the company and another has received a grant from the company.

Sirch et al. (2016)

Study size, design and location

Non-randomised retrospective cohort analysis of prospectively collected data of 2,327 patients having cardiac surgery in 3 separate periods, using propensity match scoring between groups.

Primary end point: RBS composite outcome consisting of any of the following: take back for re-exploration for haemorrhage; pericardial interventions; and pleural interventions for haemothorax, pneumothorax and effusions.

Single centre in Germany.

Intervention and comparator(s)

3 phases, data were collected during consecutive time periods:

  • patients having cardiac surgery with conventional chest tubes (n=1,849)

  • using PleuraFlow ACT in all patients, after a period of training (n=256)

  • return to use of conventional tubes (n=222).

Key outcomes

51 of 256 matched patients (19.9%) had at least 1 intervention for RBS in phase 1, compared with 29 (11.3%) in phase 2 (a 43% reduction; p=0.0087). This was mainly because of a reduction in interventions to treat pleural effusions. There were non-significant reductions in re-explorations for bleeding, interventions for pericardial effusions, and interventions for pneumothorax.

There was a significant reduction in POAF in phase 2 compared with phase 1 (p=0.013), but not in hospital mortality, cardiac arrest or permanent stroke. There was a significant reduction in median chest drainage (p=0.0024) and ventilation hours (p=0.0047), but not in overall length of stay (p=0.24).

In phase 3, there was an increase in the outcomes from phase 2 and the differences with phase 1 were not statistically significant. POAF was the only outcome that neared statistical significance (phase 1 31.5% v 23.9% phase 3, p=0.079).

Strengths and limitations

This was a non-randomised study with interventions used at different time points. Changes in ICU protocols over time and other variables were not controlled for.

Treatment and control groups were propensity matched to limit differences in patient characteristics between groups. RBS was a composite outcome which did not specifically measure retained mediastinal blood. The study did not have an end point to directly image retained blood and quantify the reduction with imaging.

Phase 2 and 3 could not be compared because of small sample sizes. The study was also part funded by the company, and 1 of the authors is a founding shareholder and board member of the company.

Perrault et al. (2012)

Study size, design and location

Survey responses from 7 surgeons and 42 intensive care nurses caring for 19 adults having cardiac surgery using PleuraFlow ACT at a centre in Canada.

Intervention and comparator(s)

PleuraFlow ACT and standard chest tubes.

Key outcomes

41 (98%) of nurses considered the system was easy to use and all that it was easy to understand.

35% of nurses reported that they had to strip or milk conventional chest tubes when clotting formed.

77% of nurses considered the device more efficient than stripping, milking or tapping the chest tube to keep it open, and 86% that it was more effective than these methods.

ICU specialists inspected tubes on removal: standard chest tubes were visually noted to be obstructed 33% (5/15) of the time, correlating with observed respiratory variation. Some non-obstructive clot on the guidewire was noted in the PleuraFlow ACT chest tubes in only 13% (2/15).

Respiratory variation was intact in all the PleuraFlow ACT systems, suggesting that the tubes were open and functional.

All the PleuraFlow chest tubes were found to be functional on removal.

Strengths and limitations

This was a single-centre study with small patient numbers. One of the co‑authors is the inventor of PleuraFlow ACT.

Abbreviations: ACT, active clearance technology; CI, confidence interval; ICU, intensive care unit; OR, odds ratio; POAF, postoperative atrial fibrillation; RBS, retained blood syndrome.

Recent and ongoing studies