A Novel Device for Surveillance of Vascular Access Sites for Bleeding

Background

Post-procedural wound hemorrhage is a potential complication following any invasive procedure. In the setting of hemodialysis, with multiple repeated vascular access procedures performed on uremic patients prone to bleeding, post-procedural bleeding would not be infrequent.

Serious adverse events following central venous catheter removal, while rare, were not isolated incidents. The British National Reporting and Learning System, reported 6 incidents of late bleeding following femoral line removal over 3 years. Of these, 3 resulted in deaths and 2 suffered more than one litre of blood loss. The UK Renal Association, British Renal Society and Intensive Care Society subsequently released a recommendation for the safe removal of a temporary femoral dialysis line in 2019. Protocols implemented in some units required an intense period of monitoring every 5-15 minutes in the first 2 hours.

However, frequent monitoring protocols places additional demands on already scarce nursing manpower. As catastrophic bleeding is rare, it is difficult to expect healthcare staff to maintain a consistent high level of vigilance for such a low probability event. The possibility of missed catastrophic bleeding is always present, particularly if the patient is incapacitated and unable to call for assistance.

A technological solution to safely monitor patients for bleeding following catheter removal is needed. Ideally, this technology should also be applicable for other wounds at risk of severe bleeding, for example, after arterial access for coronary angiogram or vascular interventions. There were no commercially available devices which fully met the user requirements, which included ease of use, wireless, compact form factor and specific detection of bleeding with no direct contact with the wound or blood. A compact device was developed

Through a research collaboration between a hospital and a technology and design university, supported by a national innovation grant, a device was developed to monitor wound for bleeding. By the end of 2018, a working prototype - Blood Warning Technology with Continuous Hemoglobin sensor (BWATCH) - was available for first-in-human clinical trials. The technological concept has been presented in electrical engineering conferences. This is a stand-alone disc-shaped device (approximately 5 cm in diameter and <1 cm thick) which can be placed over dressings and continuous monitoring for bleeding will take place. The presence of fresh blood triggers a loud alarm, alerting medical staff to attend to the patient.

Objectives

The primary aim of this clinical trial was to evaluate the sensitivity of the device in detecting bleeding in a clinical environment. The secondary aim is to determine the specificity of the device for actual bleeding (false positive rates).

Materials and Methods

This study has been reviewed and approved by the institutional review board (Singhealth cIRB 2014-2036).

This a prospective, observational study performed in an acute care hospital. Inpatients planned for dialysis catheter insertion or removal based on standard clinical care and indications will be identified fro recruitment.

Consent will be obtained from the patient or their legal representative if cognitively impaired. As part of standard care, the procedurist will secure haemostasis following the procedure and apply a bandage. The investigators will then place the device over the dressing and secure it in place. The device will then be left in-situ for at least 6 hours. Any alarms triggered and bleeding episodes will be documented in a case report form.

For safety, throughout the observation period, routine monitoring and inspection for bleeding by the nurse-in-charge would continue as per hospital clinical protocol.

Baseline demographic data of the patients will be collected. Laboratory test results collected will include the platelet count, international normalized ratio (INR) and partial thromboplastin time (PTT). The use of any anti-coagulation or anti-platelet medication will be recorded.

From the bleeding incidences and device alarm activation rates, the sensitivity and specificity of the device for detection of true bleeding will be calculated.