Using technology to improve reconnection to remote monitoring in cardiac implantable electronic device patients.

Background: Remote monitoring (RM) of cardiac implantable electronic device (CIED) patients is now considered standard of care. However, a fundamental requirement of RM is continuous connectivity between the patient's implanted device and the CIED manufacturer's central server. This study examined the rate of RM disconnections in CIED recipients and the impact of short message service (SMS) to facilitate reconnections. Methods: Using a platform that collects RM data from CIED manufacturers, we retrospectively examined the disconnection and reconnection events in 6085 patients from 20 medical centers. Each medical center reported their usual practice regarding RM disconnections, which consisted of either an automatic SMS from the platform to patients who were disconnected for 2 weeks or the standard of care (SC) of a phone call to patients. Results: During a 1-year period, 43% of patients had at least 1 disconnection. Half of these patients experienced multiple disconnections. The use of SMS reduced the time to reconnection by 43% in comparison to SC. The median time to reconnect a disconnected patient was 11.0 [3.2, 29.0] days for SC vs 6.3 [1.3, 22.0] days for SMS (P < .0001). Furthermore, there was a high rate of reconnections within the first 48 hours of the SMS message, which was nearly double that in the SC arm. Conclusion: This study demonstrates the feasibility of an automatic system to deliver an SMS to patients with a disconnected CIED to facilitate early reconnection to RM.

Filtering of remote monitoring alerts transmitted by cardiac implantable electronic devices and reclassification of atrial fibrillation events by a new algorithm.

Background: Cardiac implantable electronic devices (CIEDs) are an important means of atrial fibrillation (AF) detection. However, the AF burden measurements and notifications transmitted by CIEDs are not directly related to the clinical classification of paroxysmal, persistent, or permanent AF. Moreover, AF alerts are the most frequent form of notification, imposing a time-consuming review on caregivers. Objective: The purpose of this study was to compare the incidence of standard AF burden-related notifications in remotely monitored (RM) patients with the incidence of events detected after filtering by a new proprietary algorithm implementing the standard European Society of Cardiology classification of AF. Methods: Between 2017 and 2022, all RM patients with daily AF burden measurements available for ≥30 days and ≥1 AF burden-related alerts were enrolled at 68 medical centers. The incidence of CIED-transmitted alerts was compared to that of AF episodes detected by a new proprietary algorithm and classified as "first recorded episode of AF", "paroxysmal AF", "increased paroxysmal AF", "persistent AF", or "end of persistent AF back to paroxysmal AF or back to sinus rhythm." Results: Between January 2017 and September 2022, this retrospective study analyzed data from 4162 recipients of an Abbott, Biotronik, Boston Scientific, or Medtronic CIED, RM over mean follow-up of 605 ± 386 days. The algorithm broke down 67,883 AF burden-related alerts into 9728 (14.3%) clinically relevant AF events. Conclusion: A new AF alert algorithm successfully identified clinically significant AF events in RM CIED recipients and would markedly limit the total number of transmitted alerts that require review by caregivers.

Improved diagnostic performance of insertable cardiac monitors by an artificial intelligence-based algorithm.

AIMS: The increasing use of insertable cardiac monitors (ICM) produces a high rate of false positive (FP) diagnoses. Their verification results in a high workload for caregivers. We evaluated the performance of an artificial intelligence (AI)-based ILR-ECG Analyzer™ (ILR-ECG-A). This machine-learning algorithm reclassifies ICM-transmitted events to minimize the rate of FP diagnoses, while preserving device sensitivity. METHODS AND RESULTS: We selected 546 recipients of ICM followed by the Implicity™ monitoring platform. To avoid clusterization, a single episode per ICM abnormal diagnosis (e.g. asystole, bradycardia, atrial tachycardia (AT)/atrial fibrillation (AF), ventricular tachycardia, artefact) was selected per patient, and analyzed by the ILR-ECG-A, applying the same diagnoses as the ICM. All episodes were reviewed by an adjudication committee (AC) and the results were compared. Among 879 episodes classified as abnormal by the ICM, 80 (9.1%) were adjudicated as 'Artefacts', 283 (32.2%) as FP, and 516 (58.7%) as 'abnormal' by the AC. The algorithm reclassified 215 of the 283 FP as normal (76.0%), and confirmed 509 of the 516 episodes as abnormal (98.6%). Seven undiagnosed false negatives were adjudicated as AT or non-specific abnormality. The overall diagnostic specificity was 76.0% and the sensitivity was 98.6%. CONCLUSION: The new AI-based ILR-ECG-A lowered the rate of FP ICM diagnoses significantly while retaining a > 98% sensitivity. This will likely alleviate considerably the clinical burden represented by the review of ICM events.

What value do digital health solutions bring, what are the funding mechanisms and evaluations?

Digital health is currently booming, providing major innovations, particularly in terms of changing the practices of the stakeholders in the healthcare system as a whole. It allows our healthcare system to draw on new synergies between independent, hospital and medico-social professionals, as well as on high-performance digital tools for the benefit of all, users, patients and professionals. These tools, or digital solutions, have a strong potential to improve the healthcare system but also a strong potential for economic development. In this respect, the great diversity of existing and future digital solutions, as well as their vast fields of application, are prompting public and private stakeholders in the sector to question their integration into our healthcare system. The resulting challenges concern the identification of the targets they are intended for, the values they embody and, as a result, the methods of funding and evaluation. At a time when the first reimbursement terms for digital solutions are taking shape in the context of the Social Security Financing Bill for 2022, the roundtable wished to propose 8 recommendations to help structure exchanges between the various stakeholders and initiate avenues of work around the integration of digital solutions into the healthcare system. The main orientations are based on the proposal of a common and transparent reflection methodology around the technical scope of these solutions, the values they bring and the funding mechanisms. Other work will be necessary beyond the points addressed by the round table in order to go into greater depth on certain themes such as the adaptation of existing funding methods to the momentum and specificities of digital technology or the development of research work on the evaluation of the value claimed by these digital solutions.

Impact of COVID-19 on the incidence of cardiac arrhythmias in implantable cardioverter defibrillator recipients followed by remote monitoring.

BACKGROUND: The coronavirus disease 2019 (COVID-19) has been a fast-growing worldwide pandemic. AIMS: We aimed to investigate the incidence of cardiac arrhythmias among a large French cohort of implantable cardioverter defibrillator recipients over the first 5 months of 2020. METHODS: Five thousand nine hundred and fifty-four implantable cardioverter defibrillator recipients were followed by remote monitoring during the COVID-19 period (from 01 January to 31 May 2020). Data were obtained from automated remote follow-up of implantable cardioverter defibrillators utilizing the Implicity® platform. For all patients, the type of arrhythmia (atrial fibrillation, ventricular tachycardia or ventricular fibrillation), the number of ventricular arrhythmia episodes and the type of implantable cardioverter defibrillator-delivered therapy were recorded. RESULTS: A total of 472 (7.9%) patients presented 4917 ventricular arrhythmia events. An increase in ventricular arrhythmia incidence was observed after the first COVID-19 case in France, and especially during weeks #10 and #11, at the time of major governmental measures, with an increase in the incidence of antitachycardia pacing delivered therapy. During the 11 weeks before the lockdown order, the curve of the percentage of live-stream television coverage of COVID-19 information matched the ventricular arrhythmia incidence. During the lockdown, the incidence of ventricular arrhythmia decreased significantly compared with baseline (0.05±0.7 vs. 0.09±1.2 episodes per patient per week, respectively; P<0.001). Importantly, no correlation was observed between ventricular arrhythmia incidence and the curve of COVID-19 incidence. No changes were observed regarding atrial fibrillation/atrial tachycardia episodes over time. CONCLUSIONS: An increase in ventricular arrhythmia incidence was observed in the 2 weeks before the lockdown order, at the time of major governmental measures. Ventricular arrhythmia incidence decreased dramatically during the lockdown.

A deep neural network for 12-lead electrocardiogram interpretation outperforms a conventional algorithm, and its physician overread, in the diagnosis of atrial fibrillation.

BACKGROUND: Automated electrocardiogram (ECG) interpretations may be erroneous, and lead to erroneous overreads, including for atrial fibrillation (AF). We compared the accuracy of the first version of a new deep neural network 12-Lead ECG algorithm (Cardiologs®) to the conventional Veritas algorithm in interpretation of AF. METHODS: 24,123 consecutive 12-lead ECGs recorded over 6 months were interpreted by 1) the Veritas® algorithm, 2) physicians who overread Veritas® (Veritas®â€¯+ physician), and 3) Cardiologs® algorithm. We randomly selected 500 out of 858 ECGs with a diagnosis of AF according to either algorithm, then compared the algorithms' interpretations, and Veritas®â€¯+ physician, with expert interpretation. To assess sensitivity for AF, we analyzed a separate database of 1473 randomly selected ECGs interpreted by both algorithms and by blinded experts. RESULTS: Among the 500 ECGs selected, 399 had a final classification of AF; 101 (20.2%) had ≥1 false positive automated interpretation. Accuracy of Cardiologs® (91.2%; CI: 82.4-94.4) was higher than Veritas® (80.2%; CI: 76.5-83.5) (p < 0.0001), and equal to Veritas®â€¯+ physician (90.0%, CI:87.1-92.3) (p = 0.12). When Veritas® was incorrect, accuracy of Veritas®â€¯+ physician was only 62% (CI 52-71); among those ECGs, Cardiologs® accuracy was 90% (CI: 82-94; p < 0.0001). The second database had 39 AF cases; sensitivity was 92% vs. 87% (p = 0.46) and specificity was 99.5% vs. 98.7% (p = 0.03) for Cardiologs® and Veritas® respectively. CONCLUSION: Cardiologs® 12-lead ECG algorithm improves the interpretation of atrial fibrillation.

Role of medical reaction in management of inappropriate ventricular arrhythmia diagnosis: the inappropriate Therapy and HOme monitoRiNg (THORN) registry.

AIMS: Implantable cardioverter-defibrillators (ICDs) reduce sudden cardiac death in selected patients but inappropriate ICD shocks have been associated with increased mortality. The THORN registry aims to describe the rate of inappropriate ventricular arrhythmia diagnoses and therapies in patients followed by remote monitoring, as well as the following delay to next patient contact (DNPC). METHODS AND RESULTS: One thousand eight hundred and eighty-two patients issued from a large remote monitoring database first implanted with an ICD for primary or secondary prevention in 110 French hospitals from 2007 to 2014 constitute the THORN population. Among them, 504 patients were additionally followed prospectively for evaluation of the DNPC. Eight hundred and ninety-five out of 1551 (58%) patients had ischaemic heart disease and 358/771 (46%) were implanted for secondary prevention. During 13.7 ± 3.4 months of follow-up, the prevalence of first inappropriate diagnosis in a ventricular arrhythmia zone with enabled therapy was 162/1882 (9%). Among those patients, 122/162 (75%) suffered at least one inappropriate therapy and 58/162 (36%) at least one inappropriate shock. Eighty-three out of 162 (51%) of first inappropriate diagnosis occurred during the first 4 months following implantation. The median DNPC was 8 days (interquartile range 1-26). At least one other day with recording of an inappropriate diagnosis of the same cause occurred in 13/43 (30%) of available DNPC periods, with an inappropriate therapy in 7/13 (54%). CONCLUSION: Inappropriate diagnoses occurred in 9% of patients implanted with an ICD during the first 14 months. The DNPC after inadequate ventricular arrhythmia diagnoses remains long in daily practice and should be optimized. CLINICALTRIALS.GOV IDENTIFIER: NCT01594112.

Remote Monitoring of Cardiac Implantable Devices: Ontology Driven Classification of the Alerts.

The number of patients that benefit from remote monitoring of cardiac implantable electronic devices, such as pacemakers and defibrillators, is growing rapidly. Consequently, the huge number of alerts that are generated and transmitted to the physicians represents a challenge to handle. We have developed a system based on a formal ontology that integrates the alert information and the patient data extracted from the electronic health record in order to better classify the importance of alerts. A pilot study was conducted on atrial fibrillation alerts. We show some examples of alert processing. The results suggest that this approach has the potential to significantly reduce the alert burden in telecardiology. The methods may be extended to other types of connected devices.

Personalized and automated remote monitoring of atrial fibrillation.

AIMS: Remote monitoring of cardiac implantable electronic devices is a growing standard; yet, remote follow-up and management of alerts represents a time-consuming task for physicians or trained staff. This study evaluates an automatic mechanism based on artificial intelligence tools to filter atrial fibrillation (AF) alerts based on their medical significance. METHODS AND RESULTS: We evaluated this method on alerts for AF episodes that occurred in 60 pacemaker recipients. AKENATON prototype workflow includes two steps: natural language-processing algorithms abstract the patient health record to a digital version, then a knowledge-based algorithm based on an applied formal ontology allows to calculate the CHA2DS2-VASc score and evaluate the anticoagulation status of the patient. Each alert is then automatically classified by importance from low to critical, by mimicking medical reasoning. Final classification was compared with human expert analysis by two physicians. A total of 1783 alerts about AF episode >5 min in 60 patients were processed. A 1749 of 1783 alerts (98%) were adequately classified and there were no underestimation of alert importance in the remaining 34 misclassified alerts. CONCLUSION: This work demonstrates the ability of a pilot system to classify alerts and improves personalized remote monitoring of patients. In particular, our method allows integration of patient medical history with device alert notifications, which is useful both from medical and resource-management perspectives. The system was able to automatically classify the importance of 1783 AF alerts in 60 patients, which resulted in an 84% reduction in notification workload, while preserving patient safety.

An ontology-based annotation of cardiac implantable electronic devices to detect therapy changes in a national registry.

The patient population benefitting from cardiac implantable electronic devices (CIEDs) is increasing. This study introduces a device annotation method that supports the consistent description of the functional attributes of cardiac devices and evaluates how this method can detect device changes from a CIED registry. We designed the Cardiac Device Ontology, an ontology of CIEDs and device functions. We annotated 146 cardiac devices with this ontology and used it to detect therapy changes with respect to atrioventricular pacing, cardiac resynchronization therapy, and defibrillation capability in a French national registry of patients with implants (STIDEFIX). We then analyzed a set of 6905 device replacements from the STIDEFIX registry. Ontology-based identification of therapy changes (upgraded, downgraded, or similar) was accurate (6905 cases) and performed better than straightforward analysis of the registry codes (F-measure 1.00 versus 0.75 to 0.97). This study demonstrates the feasibility and effectiveness of ontology-based functional annotation of devices in the cardiac domain. Such annotation allowed a better description and in-depth analysis of STIDEFIX. This method was useful for the automatic detection of therapy changes and may be reused for analyzing data from other device registries.

Active or passive pulmonary vein in atrial fibrillation: is pulmonary vein isolation always essential?

BACKGROUND: The role of pulmonary veins (PVs) in persistent atrial fibrillation (AF) perpetuation appears less important than in paroxysmal AF. Electrogram-based substrate ablation is not widely performed as a stand-alone strategy. OBJECTIVE: To evaluate PV activity in AF perpetuation and efficacy of our patient-tailored ablation strategy (electrogram-based substrate ablation with or without pulmonary vein isolation [PVI]). METHODS: One hundred twenty-one patients with paroxysmal (n = 19; 15.7%), persistent (n = 77; 63.6%), or long-standing persistent (n = 25; 20.7%) AF underwent electrogram-based substrate ablation with AF termination end point: sinus rhythm or atrial tachycardia conversion. Before ablation, we classified PVs as "passive" if silent PV or if PV cycle length is greater than left atrial appendage cycle length. No PVI was performed in such cases. RESULTS: Passive PVs were observed in 52 of 121 patients (paroxysmal AF = 0%, persistent AF = 40%, and long-standing persistent AF = 76%; P < .0001]). Substrate ablation terminated AF in 95.6% (sinus rhythm conversion in 80.2%). Compared with patients with active PVs, patients with passive PVs had longer AF sustained duration (19.1 ± 29.7 months vs 4.9 ± 11.1 months; P < .0001), larger left atrial diameter (46.9 ± 7.3 mm vs 41.9 ± 6.0 mm; P = .0014), lower left ventricular ejection fraction (45.4% ± 13.5% vs 55.1% ± 9.4%; P < .0001), and more often structural heart disease (57% vs 33%; P = .02). After a follow-up of 20.39 ± 11.23 months (1.6 procedures per patient), 82% were arrhythmia free with this strategy. CONCLUSIONS: PV activity during AF decreases with AF chronicity, left atrial dilatation, and left ventricular ejection fraction. Our patient-tailored ablation strategy without systematic PVI provides good results.

Comparing Drools and ontology reasoning approaches for telecardiology decision support.

Implantable cardioverter defibrillators can generate numerous alerts. Automatically classifying these alerts according to their severity hinges on the CHA2DS2VASc score. It requires some reasoning capabilities for interpreting the patient's data. We compared two approaches for implementing the reasoning module. One is based on the Drools engine, and the other is based on semantic web formalisms. Both were valid approaches with correct performances. For a broader domain, their limitations are the number and complexity of Drools rules and the performances of ontology-based reasoning, which suggests using the ontology for automatically generating a part of the Drools rules.

Automatic computation of CHA2DS2-VASc score: information extraction from clinical texts for thromboembolism risk assessment.

The CHA2DS2-VASc score is a 10-point scale which allows cardiologists to easily identify potential stroke risk for patients with non-valvular fibrillation. In this article, we present a system based on natural language processing (lexicon and linguistic modules), including negation and speculation handling, which extracts medical concepts from French clinical records and uses them as criteria to compute the CHA2DS2-VASc score. We evaluate this system by comparing its computed criteria with those obtained by human reading of the same clinical texts, and by assessing the impact of the observed differences on the resulting CHA2DS2-VASc scores. Given 21 patient records, 168 instances of criteria were computed, with an accuracy of 97.6%, and the accuracy of the 21 CHA2DS2-VASc scores was 85.7%. All differences in scores trigger the same alert, which means that system performance on this test set yields similar results to human reading of the texts.

Mapping BFO and DOLCE.

Upper level ontologies are key technology for integrating heterogeneous information coming from different sources. DOLCE and BFO, are the favorite candidates which propose rigorous foundational principles to model any domain. The objective of the AKENATON project is to improve alert management and to support patient-centered medical decision in telecardiology. This requires to integrate information transmitted by implantable cardiac devices with clinical data extracted from patient health records. To achieve this goal, we have designed an ontology of telecardiology based on DOLCE. In order to integrate ontologies based on BFO such as FMA, we have developed a framework for mapping BFO and DOLCE categories in terms of equivalence and subsumption between categories.

Resumption of sinus rhythm in patients with heart failure and permanent atrial fibrillation undergoing cardiac resynchronization therapy: a longitudinal observational study.

AIMS: To investigate the temporal patterns, predictors, and prognostic impact of spontaneous sinus rhythm resumption (SRR) of heart failure (HF) patients with permanent atrial fibrillation (AF) treated with cardiac resynchronization therapy (CRT). METHODS AND RESULTS: This multicentre, retrospective, longitudinal study analysed 330 consecutive HF patients with permanent AF treated with a CRT device (mean age 70 +/- 9 years, male 83%, ischaemic aetiology 44%, NYHA class III-IV 93%, mean QRS duration 167 +/- 40 ms, and mean ejection fraction 26 +/- 7%). Clinical, echocardiographic, and outcome data were collected during follow-up. Thirty-four patients experienced SRR after CRT (10.3%) at a median 4-month follow-up. The strongest independent predictors were end-diastolic diameter (EDD) [hazard ratios (HR) 4.03, 95% confidence intervals (95% CI) 1.43-11.36, P = 0.008], post-CRT QRS

Testing tactics to localize de-identification.

Recent renewed interest in de-identification (also known as "anonymisation") has led to the development of a series of systems in the United States with very good performance on challenge test sets. De-identification needs however to be tuned to the local documents and their specificities. We address here two issues raised in this context. First, tuning is generally performed by language engineers who should not have to work on identified text. We therefore perform a first gross de-identification step in the hospital. Second, to set up a de-identification system for new documents in a language different from English, here French patient reports, we tested two methods: the first attempts to adapt an existing US de-identifier for English, the second re-develops a new system which applies the same methods. The first method involved localizing patterns designed for English, which proved cumbersome and did not quickly obtain good performance. With a similar effort, the latter method obtained much better results. Evaluated on a set of 23 randomly selected texts from a corpus of 21,749 clinical texts, it obtained 83% recall and 92% precision.

Relationship between HIV protease inhibitors and QTc interval duration in HIV-infected patients: a cross-sectional study.

AIMS: QTc interval prolongation and torsades de pointes have been reported in HIV-infected patients. Protease inhibitors (PIs) are suspected to contribute to this adverse reaction. However, many factors can prolong QTc interval. We examined factors influencing QTc duration in HIV-infected patients. METHODS: Unselected HIV-infected patients (n = 978) were enrolled in this prospective, single-centre cross-sectional study. Variables related to infection and treatments were collected. A digital electrocardiographic record was recorded in each patient and QT interval duration was measured and corrected using both Bazett's (QTcB) and Fridericia's (QTcF) formula. Results were analysed with a multivariable linear model. RESULTS: After excluding arrhythmias and complete bundle branch blocks, QT interval was measured in 956 patients. The mean (SD) QTcB was 418 ms (23) and QTcF was 405 ms (20). QTc was found prolonged (>450 ms in women and >440 ms in men) in 129 [13.5%; 95% confidence interval (CI) 11.5, 15.8] and 38 (4%; 95% CI 2.9, 5.4) patients using Bazett and Fridericia corrections, respectively. On multivariable analysis, incomplete bundle branch block, ventricular hypertrophy, signs of ischaemic cardiopathy, female gender, White ethnic origin and age were significantly associated with QTc prolongation. The only HIV variable independently associated with QTc prolongation was the duration of infection (P = 0.023). After adjustment, anti-HIV treatment, in particular PI (P = 0.99), was not associated with QTc prolongation. CONCLUSIONS: Although PIs block in vitro hERG current, they are not independently associated with QTc interval prolongation. Prolonged QTc interval in HIV-infected patients is primarily associated with factors commonly known to prolong QT and with the duration of HIV infection.

Using regular expressions to extract information on pacemaker implantation procedures from clinical reports.

OBJECTIVE: This study evaluated natural language processing methods to extract clinical data from free text in surgical reports related to cardiac pacing and defibrillation in order to populate a registry. METHODS: The information extraction system that we have developed is a name entity recognition system based on GATE using regular expressions. 232 reports were analyzed. For each report, we performed manual abstraction, we collected the information stored in the registry, and we performed information extraction with our system. Sensitivity,positive predictive value and accuracy were used to evaluate our method. RESULTS: Our system extracted information, including numeric data, text and combination of numbers and strings, with a high sensitivity (>90%) and very high predictive positive value (>95%). It featured a precision that was higher than the precision of the original registry database populated by manual input.Conclusion This tool based on GATE open source components provides a robust approach to extracting information from documents related to a specific narrow domain such as pacemaker reports. Further evaluation is needed for application to broader domains.

Resting echocardiographic assessments of left atrial function and filling pressure interest in the understanding of exercise capacity in patients with chronic congestive heart failure.

OBJECTIVE: Doppler echocardiography is well recognized as the primary noninvasive imaging technique to assess global and regional cardiac function. Cardiopulmonary exercise testing has become a powerful tool to predict outcome in chronic heart failure (CHF). We looked for cardiac determinants, using resting echocardiographic parameters, of exercise tolerance in patients with CHF. METHODS: Transthoracic echocardiography and standardized cardiopulmonary exercise testing were performed in 75 patients (59 +/- 11 years) with CHF, New York Heart Association functional class II to III. Systolic and diastolic function, filling pressures, and left ventricular, right ventricular, and left atrial (LA) regional function were assessed using Doppler tissue velocities and strain values. RESULTS: Maximal workload (86 +/- 41 W) and peak oxygen (14.6 +/- 3.1 mL/min/kg) correlated with left ventricular filling pressure estimates (E/Ea and E/Vp) but also with end-diastolic pulsed Doppler tissue velocity at the mitral annulus (Aa), LA volume, and regional LA function assessed by strain analysis. In multivariate analysis, maximum workload and peak oxygen were shown to correlate with right ventricular peak strain, although Aa and E/Ea were the best predictors of exercise capacity. CONCLUSION: Capacity to exercise in patients with CHF is understandable by resting echocardiography. Filling pressures, and LA and right ventricular functions, are its cardiac best determinants. Adding Aa peak velocity in resting echocardiographic evaluation of patients with CHF is found useful.