Hypertensive myocardial disease screening in low-income settings using smartphone-based technologies: a prospective multicenter trial

Abstract Background Mobile health technologies are revolutionizing cardiovascular (CV) medicine. However, more than 80% of heart disease related deaths occur in low-income settings, that are unable to take advantage of such technologies for healthcare awareness and prevention. Purpose To assess the feasibility of a pilot telemedicine-based CV screening campaign with D-Heart, a validated low-cost 8-lead smartphone portable ECG, coupled with blood-pressure (BP) monitoring in three regions of Africa. Methods A total of 950 patients (60%) men, age 29±21) were enrolled at 5 African rural dispensaries, 2 in Kenya, 2 in Senegal and at a hospital (Uganda). ECGs recording were obtained by community health workers with D- Heart smartphone ECG; BP was measured with a validated smartphone BP recorder. Global burden of ECG abnormalities was defined by a semi-quantitative score based on the sum of 9 criteria, identifying four classes of increasing severity. ECGs that were moderately and severely abnormal and from patients with severe hypertension, were sent for consultation via smartphone to D-Heart tele-cardiology centre: an advice on how to manage the patient was sent back to the community health worker’s smartphone within 15 minutes. If referral was needed, patients were sent to the local hospital. Results Mean BMI was 26.2 ± 8.5 (men 22.3±1.5 vs women 27.2±1.1, p<0.01). Active/recent history of smoking and alcohol intake were 327 (45%, 298 men) and 232/590 (40%, 198 men) respectively. A total of 142 (15%) patients had been previously diagnosed with diabetes, 199(25%) had known hypertension of whom 95/199 on treatment), whereas 43(4%) had a previous myocardial infarction. A total of 487 (51%) patients never had their blood pressure measured, whereas 798(91%) never received an ECG. Mean SBP/DBP were 138±31/71±12 mmHg respectively. Patients with normal BP were 304(32%), whereas 380(40%) were mildly-moderately hypertensive (SBP/DBP 148±7/91±5 mmHg) and 130 (29%) with severe hypertension (SBP/DBP 162±8/99±6 mmHg). D-Heart ECGs tracings were respectively classified as: normal: 361(38%); mildly abnormal: 332(35%); moderately abnormal: 199 (21%) and severely abnormal 58(6%). Most common ECG abnormality was a positive Romhilt-Estes index in 361, 38%. For 174(18%) patients smartphone tele-report advised referral to the local hospital, but only 121 69%) presented to the visit: 9 (8%) were found to have normal hearts, 57(47%) with hypertensive heart disease, 35 (29%) with rheumatic heart disease, 20 (16%) with HFrEF. Visit time was 8±2 minutes; cost of screening per patient was 1.10€: 0,30€ for consumable electrodes, 0.80€ for the community health worker visit time. Conclusions D-Heart ECG screening combined with smartphone BP measurement proved feasible and cost-effective. This should encourage to develop and extend low-cost/high-technology community-based CV screening programs in low-income settings.