Towards a machine-learning assisted non-invasive classification of dengue severity using wearable PPG data: a prospective clinical study.

BACKGROUND: Dengue epidemics impose considerable strain on healthcare resources. Real-time continuous and non-invasive monitoring of patients admitted to the hospital could lead to improved care and outcomes. We evaluated the performance of a commercially available wearable (SmartCare) utilising photoplethysmography (PPG) to stratify clinical risk for a cohort of hospitalised patients with dengue in Vietnam. METHODS: We performed a prospective observational study for adult and paediatric patients with a clinical diagnosis of dengue at the Hospital for Tropical Disease, Ho Chi Minh City, Vietnam. Patients underwent PPG monitoring early during admission alongside standard clinical care. PPG waveforms were analysed using machine learning models. Adult patients were classified between 3 severity classes: i) uncomplicated (ward-based), ii) moderate-severe (emergency department-based), and iii) severe (ICU-based). Data from paediatric patients were split into 2 classes: i) severe (during ICU stay) and ii) follow-up (14-21 days after the illness onset). Model performances were evaluated using standard classification metrics and 5-fold stratified cross-validation. FINDINGS: We included PPG and clinical data from 132 adults and 15 paediatric patients with a median age of 28 (IQR, 21-35) and 12 (IQR, 9-13) years respectively. 1781 h of PPG data were available for analysis. The best performing convolutional neural network models (CNN) achieved a precision of 0.785 and recall of 0.771 in classifying adult patients according to severity class and a precision of 0.891 and recall of 0.891 in classifying between disease and post-disease state in paediatric patients. INTERPRETATION: We demonstrate that the use of a low-cost wearable provided clinically actionable data to differentiate between patients with dengue of varying severity. Continuous monitoring and connectivity to early warning systems could significantly benefit clinical care in dengue, particularly within an endemic setting. Work is currently underway to implement these models for dynamic risk predictions and assist in individualised patient care. FUNDING: EPSRC Centre for Doctoral Training in High-Performance Embedded and Distributed Systems (HiPEDS) (Grant: EP/L016796/1) and the Wellcome Trust (Grants: 215010/Z/18/Z and 215688/Z/19/Z).

Metformin as adjunctive therapy for dengue in overweight and obese patients: a protocol for an open-label clinical trial (MeDO).

Background:  Dengue is a disease of major global importance. While most symptomatic infections are mild, a small proportion of patients progress to severe disease with risk of hypovolaemic shock, organ dysfunction and death.  In the absence of effective antiviral or disease modifying drugs, clinical management is solely reliant on supportive measures. Obesity is a growing problem among young people in Vietnam and is increasingly recognised as an important risk factor for severe dengue, likely due to alterations in host immune and inflammatory pathways. Metformin, a widely used anti-hyperglycaemic agent with excellent safety profile, has demonstrated potential as a dengue therapeutic in vitro and in a retrospective observational study of adult dengue patients with type 2 diabetes.  This study aims to assess the safety and tolerability of metformin treatment in overweight and obese dengue patients, and investigate its effects on several clinical, immunological and virological markers of disease severity. Methods: This open label trial of 120 obese/overweight dengue patients will be performed in two phases, with a metformin dose escalation if no safety concerns arise in phase one. The primary endpoint is identification of clinical and laboratory adverse events.  Sixty overweight and obese dengue patients aged 10-30 years will be enrolled at the Hospital for Tropical Diseases in Ho Chi Minh City, Vietnam. Participants will complete a 5-day course of metformin therapy and be compared to a non-treated group of 60 age-matched overweight and obese dengue patients. Discussion:  Previously observed antiviral and immunomodulatory effects of metformin make it a promising dengue therapeutic candidate in appropriately selected patients. This study will assess the safety and tolerability of adjunctive metformin in the management of overweight and obese young dengue patients, as well as its effects on markers of viral replication, endothelial dysfunction and host immune responses.  Trial registration: ClinicalTrials.gov: NCT04377451 (May 6 th 2020).

Physicians, Primary Caregivers and Topical Repellent: All Under-Utilised Resources in Stopping Dengue Virus Transmission in Affected Households.

BACKGROUND: Primary health care facilities frequently manage dengue cases on an ambulatory basis for the duration of the patient's illness. There is a great opportunity for specific messaging, aimed to reduce dengue virus (DENV) transmission in and around the home, to be directly targeted toward this high-risk ambulatory patient group, as part of an integrated approach to dengue management. The extent however, to which physicians understand, and can themselves effectively communicate strategies to stop focal DENV transmission around an ambulatory dengue case is unknown; the matter of patient comprehension and recollection then ensues. In addition, the effectiveness of N,N-diethyl-3-methylbenzamide (DEET)-based insect repellent in protecting dengue patients from Aedes aegypti mosquitoes' bites has not been investigated. METHODOLOGY: A knowledge, attitude and practice (KAP) survey, focusing on the mechanisms of DENV transmission and prevention, was performed using semi-structured questionnaires. This survey was targeted towards the patients and family members providing supportive care, and physicians routinely involved in dengue patient management in Southern Vietnam. An additional clinical observational study was conducted to measure the efficacy of a widely-used 13% DEET-based insect repellent to repel Ae. aegypti mosquitoes from the forearms of dengue cases and matched healthy controls. PRINCIPAL FINDINGS: Among both the physician (n = 50) and patient (n = 49) groups there were several respondents lacking a coherent understanding of DENV transmission, leading to some inappropriate attitudes and inadequate acute preventive practices in the household. The application of insect repellent to protect patients and their relatives from mosquito bites was frequently recommended by majority of physicians (78%) participating in the survey. Nevertheless, our tested topical application of 13% DEET conferred only ~1hr median protection time from Ae. aegypti landing. This is notably shorter than that advertised on the manufacturer's label. No differences in landing time between febrile dengue cases or matched healthy controls (n = 19 experiments) were observed. CONCLUSION/SIGNIFICANCE: Our study identifies missed opportunities for primary care physicians to improve public health through communication of strategies that could prevent focal dengue transmission in and around a case household. We advocate better access to more efficient communication methods for physicians and auxilliary health workers, supporting to educate those at high risk of DENV transmission. Our empirical testing of a widely-available 13% DEET-based repellent was limited in its protective efficacy against Ae. aegypti mosquito bites, and therefore DENV transmission, suggesting more frequent application is necessary to be beneficial.