Advice
Clinical and technical evidence
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 mibs@nice.org.uk.
Published evidence
There are over 20 peer-reviewed studies evaluating HemaClear or S‑MART, including more than 2,000 patients. Of these, there are 4 UK studies with over 300 patients and 50 volunteers. This medtech innovation briefing contains a selection of studies based on quality and relevance of outcomes. Six studies, including 1 UK study, are summarised in this briefing: 3 prospective randomised comparative studies, 2 retrospective comparative studies and 1 mixed retrospective/prospective comparative study. Out of a total of 539 patients, 328 patients had the device (named as S‑MART or HemaClear depending on the date of the study).
Clinically relevant outcomes that were reported include: efficacy of limb exsanguination, patient-reported level of pain, need for a blood transfusion, complications and infection rates.
Table 3 summarises the clinical evidence as well as its strengths and limitations.
Overall assessment of the evidence
All of the included studies evaluating HemaClear were comparative studies comparing HemaClear or S‑MART with pneumatic tourniquets. Blinding was not possible because the visual differences between the 2 devices. Overall, HemaClear appears to be as good as a pneumatic tourniquet at achieving a bloodless field (Calif and Stahl 2008, Pereira et al. 2015). Brin et al. (2015) also showed that HemaClear may reduce the amount of post-operative blood collected and reduce the haemoglobin drop compared with pneumatic tourniquets. However, both Brin et al. 2015 and Jenny et al. (2016) noted that HemaClear did not reduce the number of blood transfusions needed. HemaClear appears to be as painful or more painful than pneumatic tourniquets, with only 1 included study suggesting it may be less painful (Drosos et al. 2013).
Despite being contraindicated in patients with a known deep vein thrombosis, there are still concerns about use in patients who may be at risk of developing a venous thromboembolism. A case report by Feldman et al. (2015) showed 2 patients with traumatic injuries who developed pulmonary emboli after using HemaClear, 1 of which was fatal. They suggest caution in using the device and highlight a safety concern that HemaClear could dislodge a pre-existing deep vein thrombosis. Brin et al. 2015 also saw 2 cases of deep vein thrombosis with HemaClear.
There is limited further evidence in different specialities. Ladenheim et al. (2013) and Bourquelot and Levy (2016) reviewed HemaClear in haemodialysis access surgery and Tang et al. (2013) assessed its use in haemorrhagic shock and trauma. Further evidence in specialities such as plastic surgery in addition to its use outside of the operating theatre are needed to back up the company's claims about its potential use in these areas.
Table 3 Summary of selected studies
|
Study size, design and location |
A multicentre prospective randomised trial involving 70 patients having elective knee procedures. Location: UK. |
|
Intervention and comparator(s) |
Intervention: S-MART (n=36). Comparator: pneumatic tourniquet (n=34). |
|
Key outcomes |
23 out of 34 non-sterile pneumatic tourniquets were contaminated (68%). 0 out of 36 sterile tourniquets were contaminated. This difference was statistically significant (p<0.01). Coagulase-negative Staphylococcus spp grew in 11 out of 23 positive samples (48%), Bacillus spp in 8 samples (35%), Coliform spp in 3 samples (13%) and Sphingomonas paucimobilis in 1 sample (4%). Colony counts were described as low, ranging from 1 to over 61. |
|
Strengths and limitations |
This was a relatively good sample size. The study was done in the UK study so the results are generalisable to the NHS. Laboratory scientists were blinded to the source of tourniquet samples. The study only assessed sterility, other outcomes such as ability to exsanguinate the limb and ease of use were not evaluated. There was no patient follow up so the clinical importance of colonisation cannot be determined, for example, development of surgical site infections. |
|
Study size, design and location |
A prospective, randomised, comparative study involving 50 patients having carpal tunnel release surgery. Location: Greece. |
|
Intervention and comparator(s) |
Intervention: silicone ring tourniquet (S-MART or HemaClear; n=25). Comparator: pneumatic tourniquet (n=25). |
|
Key outcomes |
VAS pain scores were measured at 3 intervals: at application (T0), 5 minutes (T5) and at removal (Tfinal). Mean VAS pain scores were similar between SRT and PT groups at T0 (3.92±2.12 versus. 3.12±2.05, p=0.181) and T5 (4.44±1.80 versus. 3.88±1.92, p=0.294). The increase in pain from T0 to Tfinal was significantly greater in the PT group compared with SRT (88.5% versus 26.5%, p=0.002) and mean Tfinal was significantly higher in the PT than SRT group (5.88±1.48 versus 4.96±1.65, p=0.043). There were no complications related to the tourniquet in either group. |
|
Strengths and limitations |
Patients had similar baseline characteristics and were randomised using a stratified and block method, minimising the effect of potential confounders. A 30‑day follow-up period enabled observation for further complications. Validated VAS scale used for measuring pain. Only observed in 1 surgical procedure and 1 site (forearm). Strict inclusion criteria may limit generalisability to target population. No information on how 50 patients were identified and selected. |
|
Study size, design and location |
A multicentre prospective randomised controlled study involving 60 patients having elective carpal tunnel surgery. Location: Israel. |
|
Intervention and comparator(s) |
Intervention: S-MART (n=30). Comparator: pneumatic tourniquet (n=30). |
|
Key outcomes |
A bloodless field was obtained in 29 out of 30 cases in both S-MART and PT group. The S-MART device failed in 2 out of 30 patients compared with 0 patients in PT group. Failure constituted bleeding with inadequate visual quality in 1 case and intolerable pain in another. Higher VAS pain scores were reported in S‑MART group compared with PT group, which neared significance (5.7±2.5 versus 4.53±1.99, p=0.05). S‑MART was rated more difficult to apply compared with PT group (15 cases 'simple' compared with 15 cases 'moderate' difficulty in S‑MART group, compared with all 30 cases rated 'simple' to apply in PT group). Local skin redness was seen after surgery in 100% of patients. No post-operative neurovascular complications were noted. |
|
Strengths and limitations |
Intention to treat analysis used. Similar baseline characteristics between the 2 groups. Mean duration of surgery significantly higher in control group (p<0.001), which may have affected results. One week follow up for observation of complications. Higher proportion of patients having analgesics before surgery in S‑MART group neared significance (p=0.085) and will likely have reduced pain scores. Strict inclusion criteria, for example, excluding patients with congestive heart failure or chronic vascular disorders may limit generalisability to the target population. Surgeon completed intra-operative measurements, such as ease of device application, creating ascertainment bias. No random allocation to the 2 groups may also have introduced bias. Oneg HaKarmel Ltd supplied the S‑MART devices free of charge to be used in the study. |
|
Study size, design and location |
A single-centre mixed retrospective and prospective comparative study of 72 patients having total knee arthroplasty between May 2014 and June 2015. Location: France. |
|
Intervention and comparator(s) |
Intervention: HemaClear (n=33). Comparator: pneumatic tourniquet (n=39). |
|
Key outcomes |
Comparing the HemaClear group with the PT group, there was no significant difference in mean blood loss (901±488 ml compared with 989±505 ml), mean haemoglobin drop (1.8±0.8 g/dl compared with 2.2±1.0 g/dl), blood transfusion requirements (4 compared with 10 units) or mean VAS day 3 pain scores (3.2±1.3 compared with 2.9±1.7). The reduction in discharge delay in the HemaClear group compared with PT neared significance (mean 3.8 compared with 5.4 days, p=0.05). There were significantly fewer complications (1 compared with 9 cases, p=0.02) in the HemaClear group compared with PT group. The complication in the HemaClear group was wound dehiscence. |
|
Strengths and limitations |
3-month follow-up period. A power calculation was used for sample size to detect a difference of 200 ml of blood loss. Consecutive patient inclusion with no loss to follow up. Potential biases from lack of randomisation and the use of only 1 surgeon for all procedures. There was a significant difference in tourniquet time comparing HemaClear with PT group (86±18 minutes compared with 95±18 minutes, p=0.04), which limits comparability of certain outcomes such as pain score. The study was not controlled, because the control group used retrospective data, whereas intervention group used prospective data with at most an 11-month gap between procedures. Although the authors state no change in operating technique or post-operative care, the lack of controlled conditions does limit comparability. |
|
Study size, design and location |
A single-centre retrospective comparative study involving 211 patients having elective unilateral total knee arthroplasty between 2006 and 2011. Location: Israel. |
|
Intervention and comparator(s) |
Intervention: HemaClear (n=166). Comparator: pneumatic tourniquet (n=145). |
|
Key outcomes |
Comparing HemaClear with PT group, there was significantly less reduction in haemoglobin post-operatively on days 1 (2.53±0.95 compared with 2.78±0.98 g/dl, p<0.028) and 3 (3.00±1.14 compared with 3.28±1.18 g/dl p<0.045) and significantly less mean blood collected in drains at 24 hours (252.8 cc compared with 346.1 cc, p<0.001). There was no significant difference in the number of blood transfusions (no reported data) or incidence of wound complications (4.2% compared with 7.7%, p=0.189) between HemaClear and PT groups. There were 2 cases of deep vein thrombosis in HemaClear group compared with 1 case in PT group. |
|
Strengths and limitations |
Power calculation used to determine sample size. Similar baseline characteristics between the 2 groups. Use of historical control group operated on between 2006 and 2007 compared with HemaClear group operated on between 2010 and 2011. Long time between 2 groups does not assure groups were treated in the same way. No blinding when collecting study data from patient files. Same operating team and same procedure for all subjects may limit generalisability. |
|
Study size, design and location |
A single-centre retrospective study involving 76 patients having carpal tunnel surgery during October 2013. Location: France. |
|
Intervention and comparator(s) |
Intervention: HemaClear Model-F (n=38). Comparator: pneumatic tourniquet (n=38). |
|
Key outcomes |
Limb exsanguination quality was measured by the operating surgeon as 0 (no bleeding) to 10 (uncontrollable bleeding). Tourniquet-related pain was measured from 0 (no pain) to 10 (worst pain). There was a significant difference (4.5 minutes compared with 5 minutes, p=0.03) in mean operative time comparing HemaClear with PT. Comparing HemaClear with PT group at the time of tourniquet release, the quality of limb exsanguination was not significantly different (0.5 compared with 0.3, p=0.7) and there was no significant difference in pain (0.3 compared with 0.9, p=0.1). No complications were noted in either group. Although the surgery was significantly faster in the HemaClear group compared with PT group, HemaClear was more expensive by 30€. |
|
Strengths and limitations |
Only 1 surgeon, 1 operative procedure and 1 model of HemaClear assessed, limiting generalisability of results. Surgeon-rated quality of limb exsanguination may introduce ascertainment bias. Difference in operative time between 2 groups limits comparability of outcomes such as pain. Provin Medical supplied HemaClear for the purpose of the study. |
|
Abbreviations: SRT, silicone ring tourniquet; VAS, visual analogue scale; PT, pneumatic tourniquet. |
|