Recording an ECG with a connected watch: gadget or revolution?

Focus

Published on 05 Dec 2022Lecture 5 min

Watches are gradually freeing themselves from the smartphone with which they are associated to become more or less independent connected objects, with their own functionalities: possibility of making calls and sending SMS, direct connection to the Internet, GPS chip, music storage , etc. A number of medical data can be monitored in real time: measurement of activity, evaluation of the amount of sleep, detection of falls, heart rate sensor and recording of an electrocardiographic trace (ECG). Tomorrow, new features are announced Blood pressure and blood sugar measurement, diagnosis of sleep apnea syndrome, transformation of headphones into hearing aids and thermometers, possibility of storing and consulting medical data on your smartphone (iCloud health), are the future features announced. The ECG trace An ECG trace can be acquired from 2 electrodes: the first located on the back of the watch, in contact with the skin, the second, positioned in the wheel to the right of the quadrant. To record a trace, the user must be proactive and place the index finger on the wheel continuously for 30 seconds. A tracing obtained with a watch is only partial and very different from an electrocardiogram carried out in a medical structure since only one derivation (DI if the watch is positioned on the left wrist) is recorded. A significant amount of information, relating essentially to the heart rate, can however be deduced from this single derivation. A tracing limited to DI cannot, however, constitute a substitute for the 12-lead ECG. In many pathologies, the occurrence of a localized anomaly in certain myocardial territories may not be demonstrated in DI and a tracing restricted to DI may turn out to be falsely normal, wrongly reassure a patient and delay treatment. To optimize diagnostic yield, it is however possible to obtain additional leads by positioning the watch at the level of the left ankle (2 additional peripheral leads: DII and DIII) or on the location corresponding to the 6 precordial leads. Free access to the ECG in real time, without prescription and without prescription It could allow users to play a more active role in their health care, facilitate prevention efforts and constitute a real revolution for our discipline. Most of the literature has focused on the interest of this type of device in the diagnosis of atrial fibrillation. Each brand offers an algorithm approved by the FDA analyzing the regularity of the ventricular rhythm and allowing the automatic diagnosis of atrial fibrillation (figure)(1-3). Progress in terms of artificial intelligence should make it possible in the future to reduce the number of false positives but also to considerably increase the automatic diagnostic capacities for the moment limited to the differentiation between sinus rhythm and atrial fibrillation. Figure. Example of recordings of an ECG tracing in a patient with atrial fibrillation with 3 watch models. An interest that remains to be demonstrated The potential interest of connected watches in the context of various clinical situations (unexplained palpitations, chest pain, syncope, prevention of sudden death, training of athletes, etc.) remains to be demonstrated(4). Positive results could allow approval by competent authorities for indications other than the diagnosis of atrial fibrillation. It is therefore an extremely promising field of research, the potential clinical applications being numerous and certainly still largely underestimated. It is not uncommon in our daily practice to arrive today at a diagnosis thanks to a tracing recorded with a watch and to measure the potential interest in the field of rhythmology. However, there is a mismatch between the initial marketing target of connected watches, namely a rather young population attracted by new technologies but a priori in good health, and the target population of connected medical devices, namely seniors who are often more reluctant and less consumers of this type of device but presenting a greater risk of cardiovascular disease. Initial feedback suggests that this type of device can be heaven or hell for hypochondriacs, among whom the health app is very popular. The risk of creating imaginary diseases is also high in a young, healthy and low-risk population. A false positive, an erroneous diagnosis of atrial fibrillation, can create anxiety and lead to the performance of unnecessary additional tests or even the introduction of unsuitable treatments. A controversial tool The irruption of new digital technologies in the health field has been accompanied by fiery and often contradictory debates in the medical community. It seems preferable to avoid peremptory and definitive judgments – “The connected watch? it’s just a gadget”, “Older patients from the depths of the Creuse will never wear a connected watch”, etc. – or conspiracy theories – “Apple already sold your health data to big pharma”. However, it is not a question of adopting blissful optimism and many questions, of a practical, organizational or more philosophical nature, arise concerning the integration of health applications into the daily lives of consumers. A first question concerns the respect of data confidentiality, a major problem in the field of connected health. Since the companies hold the raw data, certain legitimate questions relate to a possible use without the consent of the users. Regulators will have to be very vigilant about respecting the anonymity and informed consent of users before the possible use of personal data by the company. Similarly, in the absence of reimbursement by the various health systems, the high price may constitute an obstacle to the democratization of the use of watches in clinical practice. There is also the problem of the additional workload caused for healthcare professionals in the absence of specific billing codes. A wide distribution of these new devices in our health systems requires the support of professionals in the sector and probably a financial incentive for the time spent analyzing digital data.