Summary of Clinical Trial and results

When and where was the study done?

The study was caried out at Thoracic Surgery Department in Lund, Sweden. The department have well documented expertise in the field of heart surgery and is one of two places in Sweden that carry out heart transplantation.

Which patients were included in the study

The patient included in the study were patient that were recovering after  heart surgery. This group/cohort was expected to have a reasonably high number of  ECG alarms, but an otherwise good prognosis for recovery.

Who sponsored this study?

The study was sponsored by the manufacturer of the CardioSenseSystem:
Novosense AB
Ideon Science Park
SE-223 70 Lund
Sweden

What was studied?

Study Design

The study was a prospective, controlled, comparative clinical trial of a new ECG monitoring system, CardioSenseSystem, prior to CE marking. The aim of the study was to demonstrate that the performance of the cable-free CardioSenseSystem ECG monitoring device (investigational device) is equivalent to, or better than, a traditional standard cable-based ECG monitoring system (control device). In this study the accepted industry standard was a Philips IntelliVue monitoring device.

ECG monitoring interruptions*, management time* and alarm performance* were measured. The required study population was estimated to be forty (40) to sixty (60) adult patients requiring ECG, and 49 fulfilled the eligibility criteria and were included in the study. Both the investigational device and the control device were used simultaneously on each patient to measure data loss*, management time* and alarm function* for periods of up to 24 hours.

Primary Outcome Measure

The primary outcome measure was the loss of monitoring data, as this is the most crucial parameter in any monitoring system. Any loss of data may lead to fatal arrythmias going unnoticed or a delay in alerts, which could lead to permanent injury, or even death of the patient.

Secondary Outcome Measures

The secondary outcome measures were: number of correct yellow and red alarms, number of false or technical alarms, time for handling and operating the monitoring systems and any incidence and severe adverse events.

Possible side effects where carefully monitored. Prior to the study, skin irritation resulting from a reaction to the adhesive on the sensors was identified as a potential side effect, as this has been reported previously. However, no side effects of any kind, or adverse events, were observed in the study.

Out of the potential maximum 49 days of observation (24 hours per patient), the 49 patients were monitored by both CSS and PS for a total time of 41.59 days.

Patient monitoring – data losses

Based on data adjusted for outliers, the mean loss in ECG monitoring per patient day was 145 seconds for CSS (2 minutes and 25 seconds) and 584 seconds for PS (9 minutes and 44 seconds). The superiority of CSS was established with a very high statistical significance (p-value < 0.0001). The data losses for the comparator PS, CSS and the difference can be seen in Table 1.

*Extrapolated data loss for a patient with a mean hospital stay of 7.3 days
**Data loss extrapolated for the whole department during a year 7378 patient monitoring days, 1 day correspond to 24 hour.

Alarm fatigue – Lead off alarm

Lead off alarms, i.e. electrodes or ECG cables becoming detached, have been identified as the main cause of alarm fatigue among health care staff. This is a frequent alarm that requires immediate attention as the patient is not monitored when this technical alarm is activated. The nurse must connect the cable or replace ECG sensor to allow patient monitoring to continue. The number of lead off alarms in this study, was 4 for the investigational device and 30 for the control device. The frequency is thus significantly lower for the CardioSenseSystem, reducing the risk of alarm fatigue, compared to the control system.

ECG alarms

The investigational device registered 28 correct red, high-priority ECG alarms, compared to 12 by the control system.

The investigational device registered 479 correct yellow, low-priority ECG alarms, compared to 399 by the control system.

The CardioSenseSystem thus gave significantly more correct alarms, especially in the case of the high-priority alarms. However, statistical significance was not achieved in the statistical analysis of all ECG alarms.  Thus, the 16 life-threatening ECG alarms given by the CardioSenseSystem, but not by the control system, could have been the result of chance.

Handling time

The handling includes four activities

• Application of electrodes and cables, or sensor
• Extra handling of loose cables / electrodes, or sensor
• Replacing the battery
• Sterilization of cables / box

The total handling time was considerably shorter for the CardioSenseSystem, as can be seen from Table 2.

Title of Study: CardioSenseSystem Compared Study Regarding Efficacy and Safety in the Monitoring of ECG

Protocol Number: NovoECG. 5.1-2018-18343 ( Registry Identifier: Swedish Medical Products Agency ) 2018/9 ( Registry Identifier: Regional Ethical Review Board in Lund, Sweden )

Clinicaltrials.gov: NCT03610529

Study Sponsor: Novosense AB, Ideon Science Park, SE-223 70 Lund, Sweden