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 for medtech innovation briefings. 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 6 studies summarised in this briefing. This includes 3 randomised controlled trials, 1 prospective study and 2 retrospective studies. A total of 878 people are included in these studies.

The clinical evidence and its strengths and limitations is summarised in the overall assessment of the evidence.

Overall assessment of the evidence

The studies were in relevant populations, including those having major surgery with intra-arterial blood pressure measurement. However, there was limited evidence on using the device in intensive care. There was also limited published evidence relating to reduced hospital stay, need for further treatment or unplanned readmissions. One study was partially done in the UK, but other studies were done in the US or Europe which limits generalisability to the NHS. The evidence shown is on different versions of the monitor and sensor. The company states that the EV1000 monitor is the previous generation of monitoring platform and that the HemoSphere monitor has an improved interface and can have additional sensors attached. The FloTrac sensor is a different brand name that has subsequently been updated to the Acumen IQ sensor. It measures the same parameters as the Acumen IQ sensor, with the addition of hypotension prediction index (HPI), contractility and afterload. One study included in this evidence summary showed no changes to the incidence of hypotensive events. The company states that this is a result of the protocol design and the authors for this study note that over half of the hypotension alerts in the intervention groups were not treated.

Davies et al. (2020)

Intervention and comparator

EV1000 monitoring system with HPI software.

Key outcomes

This study included people having major surgery (major abdominal, vascular, or off-pump coronary artery bypass surgery) who had arterial cannulation for blood pressure and or cardiac output measuring. This included 78 women and 177 men with a mean age of 68 years. Of the 255 people in the study, 221 (86.7%) had 1 or more hypotensive events (defined as a mean arterial pressure [MAP] of less than 65 mmHg for more than 1 minute). The HPI software predicted hypotension with a sensitivity and specificity of 85.8% (95% confidence interval [CI], 85.8% to 85.9%) and 85.8% (95% CI, 85.8% to 85.9%) respectively, 5 minutes before a hypotensive event. At 10 minutes before a hypotensive event, the sensitivity and specificity were 81.7% (95% CI, 81.6% to 81.8%) and 81.7% (95% CI, 81.6% to 81.8%). At 15 minutes before a hypotensive event the sensitivity and specificity were 80.6% (95% CI, 80.5% to 80.7%) and 80.6% (95% CI, 80.5% to 80.7%).

Strengths and limitations

Because this study was retrospective, there was no data available on the treatment of hypotensive events. Some of the clinicians involved in the study had access to the HPI software at the time of the surgery and so could have treated hypotensive events before they happened. The study used an EV1000 monitor. It is not clear what device was used to measure invasive blood pressure.

Hatib et al. (2018)

Intervention and comparator

FloTrac algorithm analysis of collected data.

Key outcomes

The algorithm was developed using a retrospective training cohort of 1,334 records of those in intensive care or having surgery. External validation was then done using the software in people having surgery. The study showed that the software predicted arterial hypotension with a sensitivity and specificity of 88% and 87% 15 minutes before a hypotensive event, respectively. It had sensitivity and specificity of 89% and 90% 10 minutes before, and 92% and 92% 5 minutes before.

Strengths and limitations

This study used intensive care and surgical cases to train the algorithm, but only used surgical cases for the external validation. The study used a FloTrac algorithm. This study was funded by the company.

Schneck et al. (2019)

Intervention and comparator

EV1000 HPI monitoring system with FloTrac IQ sensor compared with routine anaesthetic care.

Key outcomes

There were 3 groups included in this study, 1 using the HPI software (n=25), a control group of routine anaesthetic care (n=24) and a historic control group (n=50). The study showed that significantly fewer people in the HPI group had at least 1 episode of hypotension during anaesthesia (p<0.001). The HPI group had significantly reduced relative duration of hypotensive episodes (minutes below MAP 65 mmHg in percent of total anaesthesia time) compared with both control groups (p<0.001).

Strengths and limitations

This feasibility study was underpowered to prove the outcome benefits of HPI use, as stated by the authors. This study was funded by the company.

Wijnberge et al. (2020)

Intervention and comparator

FloTrac IQ sensor and HemoSphere monitor with HPI software or standard anaesthetic care.

Key outcomes

People were randomly assigned to have either the intervention (n=34) or standard care control (n=34), with 60 people (88%) completing the trial. The median time of hypotension per person was 8 minutes in the intervention group compared with 32.7 minutes in the control group. The median difference was 16.7 minutes (95% CI, 7.7 to 31.0 minutes; p<0.001).

Strengths and limitations

Because this was a preliminary single-centre randomised controlled trial the samples size was small. The study was unblinded. It only included people who needed invasive blood pressure monitoring, who were generally more severely ill and were at higher risk of hypotension than the general population. This study was funded by the company.

Maheshwari et al. (2020)

Intervention and comparator

FloTrac IQ sensor and EV1000 monitoring system with HPI software with the HPI outputs only being unblinded to the clinicians in the intervention group.

Key outcomes

The HPI and advanced haemodynamic information was recorded in both groups but only available to the intervention group. The time-weighted MAP less than 65 mmHg was the same in the intervention and comparator groups (0.14, p=0.757). In the intervention group, more than half of all the alerts did not lead to interventions. However, when clinicians did intervene, hypotension was reduced by 57%.

Strengths and limitations

This pilot randomised controlled trial included a large number of people, with 80% power for detecting about 20% relative reduction in the mean of the area under the curve MAP less than 65 mmHg. However, the study excluded those with certain cardiac conditions and half of the alerts were not followed by treatment. The study was done in the US and so may not be generalisable to the NHS. The study was funded by the company.

Shin et al. (2020)

Intervention and comparator

Acumen IQ sensor with EV1000 monitor and HPI software.

Key outcomes

The study collected continuous waveform data from the EV1000 monitor for analysis at the end of the surgical procedure. The anaesthetic care team were blinded to the monitor outputs. The area under the curve, sensitivity, and specificity 5 minutes before a hypotensive event was 0.90 (95% CI, 0.853 to 0.949), 84% (95% CI, 77.7 to 90.5), and 84% (95% CI, 70.9 to 96.8), respectively. At 10 minutes before the event the area under the curve, sensitivity, and specificity was 0.83 (95% CI, 0.750 to 0.905), 79% (95% CI, 69.8 to 88.1), and 74% (95% CI, 58.8 to 89.6), respectively. At 15 minutes before the hypotensive event the area under the curve, sensitivity, and specificity was 0.83 (95% CI, 0.746 to 0.911), 79% (95% CI, 68.4 to 89.0), and 74% (95% CI, 58.8 to 89.6), respectively.

Strengths and limitations

This study was single-armed and retrospective. It was limited by having a small sample size, in which the procedural characteristics varied because multiple types of coronary procedures were included.

Sustainability

The Acumen IQ sensor is single use and the HemoSphere monitor is reclaimed after the life of the contract. There is no published evidence to support these claims.

Recent and ongoing studies