BACKGROUND
In hemodialysis, venous needle dislodgement (VND) is one of the major concerns in healthcare safety. Although VND is infrequent, it can be life-threatening.
OBJECTIVE
The main objective of this study was, combining a novel leakage detection device and IoT (Internet of Things) technology, to implement a real-time multi-bed monitoring system for VND.
METHODS
The core of the system was a novel leakage detection patch that consisted of multiple concentric rings to detect the blood leakage and to quantify the leaked volume. The Acusense IoT platform was adopted to enable multi-bed monitoring and the processed information and alarms were wirelessly sent to caring personnel. The performance of the leakage detection system was evaluated on a prosthetic arm and the reliability of the IoT system was tested in the multi-bed configuration under many simulated conditions. Finally, the whole system was investigated in a clinical study. Hemodialysis patients having a high risk of blood leakage were recruited as candidates. The system was set up in a hospital and the subjects were monitored in five shift periods for 2 months.
RESULTS
In the part of the pre-clinical simulation experiment, the results could be calibrated for blood leakage volumes from 0.3 to 0.9 mL before the measurement saturated. The response time of our proposed device was between 1.5 and 4.3 seconds. These data served as the basis for different clinical users to determine the blood leakage values. In the test of the IoT system, the overall success rate of tests reached 100% with no lost packets. In the part of the clinical experiment, a total of 52 subjects with an average age of 62.8 ± 1.3-year-old were recruited. From the IoT platform, a total of 701 dialysis sessions were collected, including 22 events of venous needle dislodgement or bleeding, of which 20 events were detected. The accuracy, sensitivity, precision, specificity, negative predicted value and F1-score was 99.7%, 90.9%, 100%, 100%, 99.9% and 99.7%, respectively. There were no reported adverse reactions. Twenty-two valid questionnaires were collected from the nursing staff in the pre-test period and 21 valid questionnaires in the post-test period. The reliability of the questionnaires was 0.853 (system experience), 0.7111(system operation), and 0.711 (nurse-patient interaction), respectively. It shows that the care has changed altitude and reduced worry of nursing staff after the usage and training.
CONCLUSIONS
A novel detector combined with an IoT system realized automatic multi-bed monitoring of blood leakage and provided support for real-time clinical decisions. The system reduced the load of medical staff and improved patient safety. In the future, it can also be applied to home hemodialysis for telemedicine. During the era of the Covid-19 pandemic, there is an imperative need for a remote care system of home hemodialysis and other pipeline safety.
CLINICALTRIAL
The study design was approved by the Institutional Review Board of TMANH (IRB number TMANH108-REC001).