A new algorithm for an old problem: Reducing false alarms and alarm fatigue for ventricular tachycardia detection in the hospital setting

Continuous cardiac telemetry monitoring is the gold standard of care for detecting ventricular tachycardia in hospitalized patients. The most recent practice standards for electrocardiographic (ECG) monitoring in the hospital setting were updated in 2017 and largely based on the seminal work done by Drew and colleagues, including ECG advances such as ST-segment and QTc monitoring. 1Sandau K.E. Funk M. Auerbach A. et al.Update to Practice Standards for Electrocardiographic Monitoring in Hospital Settings: A Scientific Statement From the American Heart Association.Circulation. 2017; 136: e273-e344https://doi.org/10.1161/CIR.0000000000000527Crossref PubMed Scopus (140) Google Scholar,2Drew B.J. Califf R.M. Funk M. et al.Practice standards for electrocardiographic monitoring in hospital settings: an American Heart Association scientific statement from the Councils on Cardiovascular Nursing, Clinical Cardiology, and Cardiovascular Disease in the Young: endorsed by the International Society of Computerized Electrocardiology and the American Association of Critical-Care Nurses [published correction appears in Circulation. 2005 Jan 25;111(3):378].Circulation. 2004; 110: 2721-2746https://doi.org/10.1161/01.CIR.0000145144.56673.59Crossref PubMed Scopus (466) Google Scholar The updated practice guidelines address several concerns that unfortunately still exist today, such as the frequency of false alarms and alarm fatigue. False alarms may occur when electrodes are not properly placed on the patient and estimates indicate 68%- 99% of alarms are false or nonactionable.1Sandau K.E. Funk M. Auerbach A. et al.Update to Practice Standards for Electrocardiographic Monitoring in Hospital Settings: A Scientific Statement From the American Heart Association.Circulation. 2017; 136: e273-e344https://doi.org/10.1161/CIR.0000000000000527Crossref PubMed Scopus (140) Google Scholar One common example is motion artifact that may mimic ventricular tachycardia (VT) resulting in misdiagnosis and unnecessary interventions such as antiarrhythmics agents, diagnostic catheterizations, and cardiac implantable electronic device implantations.3Knight B.P. Pelosi F. Michaud G.F. Strickberger S.A. Morady F. Clinical consequences of electrocardiographic artifact mimicking ventricular tachycardia.N Engl J Med. 1999; 341: 1270-1274https://doi.org/10.1056/NEJM199910213411704Crossref PubMed Scopus (163) Google Scholar Alarm fatigue occurs when nurses become desensitized due to the high frequency of false alarms, leading to action on nonactionable alerts such as unnecessary treatment or missed true events such as lethal arrhythmias.,4 The updated guidelines stress the need for proper skin preparation and lead placement, tailoring alarm settings, and nursing education, noting the trend for continued overuse of ECG monitoring in patients when it is not indicated.1Sandau K.E. Funk M. Auerbach A. et al.Update to Practice Standards for Electrocardiographic Monitoring in Hospital Settings: A Scientific Statement From the American Heart Association.Circulation. 2017; 136: e273-e344https://doi.org/10.1161/CIR.0000000000000527Crossref PubMed Scopus (140) Google Scholar Moreover, there is a need for research focusing on improved signal quality that reduces false or nonactionable alarms and increases specificity without an unacceptable loss of sensitivity.5Cvach M. Monitor alarm fatigue: an integrative review.Biomed Instrum Technol. 2012; 46: 268-277https://doi.org/10.2345/0899-8205-46.4.268Crossref PubMed Scopus (375) Google Scholar In efforts to provide enhanced ECG signal quality and reduced false alarms for VT detection, Prasad, and colleagues sought to compare current bedside monitor VT alerts to VT alerts using a new algorithm developed by their research group and to explore the association between VT alerts and 30-day in-hospital mortality.6Prasad PA, Isaksen JL, Abe-Jones Y, Zègre-Hemsey JK, Sommargren CE, Al-Zaiti SS, Carey MG, Badilini F, Mortara D, Kanters JK, Pelter MM, Ventricular tachycardia and in-hospital mortality in the intensive care unit, Heart Rhythm O2 (2023), doi: https://doi.org/10.1016/ j.hroo.2023.09.008.Google Scholar,7Pelter M.M. Carey M.G. Al-Zaiti S. et al.An annotated ventricular tachycardia (VT) alarm database: Toward a uniform standard for optimizing automated VT identification in hospitalized patients.Ann Noninvasive Electrocardiol. 2023; 28e13054https://doi.org/10.1111/anec.13054Crossref PubMed Scopus (0) Google Scholar A closed network system was used for ECG and physiologic waveform data using GE Healthcare software and data were sent to a secure and hospital approved server maintained specifically for their research. To process new VT, they used multichannel ECG waveforms, the SpO2, and invasive arterial blood pressure (BP) waveforms. The new VT algorithm alerts are designed to decrease false VT by identifying factors such as motion/noise artifact, bundle branch blocks, and ventricular pacing as primary false sources of VT. Additional algorithm strategies include identifying P waves and correlating simultaneous drops in SpO2 and drops in invasive BP waveforms during a potential VT event. VT events identified using the new algorithm were then annotated as true or false based on a defined protocol conducted by PhD nurse scientists. Based on a total of 5,670 ICU admissions, 503 (8.9%) experienced 30-day in-hospital mortality. VT events were higher in the current bedside monitor alert group (30%) compared to new-unannotated algorithm VT alerts (14.3%) and true annotated VT (11.6%). The authors also found that while there was no association for an increased rate of 30-day mortality with current bedside monitoring for VT (HR 1.06, 95% CI 0.88-1.27), there was an increased association for 30-day hospital mortality using the new-unannotated algorithm (HR 1.38,95% CI 1.12-1.69) and true-annotated VT (HR 1.39, 95% CI 1.12-1.73). Since 30-day hospital mortality was associated with unannotated and annotated true VT, the authors conclude the new algorithm provides more accuracy in identifying high-risk VT than current bedside monitoring for VT. The work from Prasad and colleagues highlights the ongoing issue of false alarms and alarm fatigue in the hospital setting, particularly in the ICU and step-down units. The Joint Commission continues to call for alarm safety and is a 2023 National Patient Safety Goal with recommendations to ensure that alarms on medical equipment are heard and responded to on time.8 Paine and colleagues conducted a systematic review of physiologic alarm characteristics and pragmatic interventions to reduce alarm frequency. Their search yielded 8 studies that evaluated interventions such as widening alarm parameters, instituting alarm delays, and using disposable ECG wires or frequently changing ECG electrodes. While results are unproven and are at risk of bias, the authors conclude that the interventions are promising in reducing false alarms.9Paine C.W. Goel V.V. Ely E. et al.Systematic Review of Physiologic Monitor Alarm Characteristics and Pragmatic Interventions to Reduce Alarm Frequency.J Hosp Med. 2016; 11: 136-144https://doi.org/10.1002/jhm.2520Crossref PubMed Scopus (122) Google Scholar Other available solutions to reduce alarm fatigue include monitoring only patients with indications, daily electrode change and skin prep, optimizing alarm state recognition, and continued education.10Hravnak M. Pellathy T. Chen L. et al.A call to alarms: Current state and future directions in the battle against alarm fatigue.J Electrocardiol. 2018; 51: S44-S48https://doi.org/10.1016/j.jelectrocard.2018.07.024Crossref PubMed Scopus (47) Google Scholar Future solutions to improve algorithms include more sensitive/specific pattern recognition, integrating multichannels simultaneously, and including additional patient information.10Hravnak M. Pellathy T. Chen L. et al.A call to alarms: Current state and future directions in the battle against alarm fatigue.J Electrocardiol. 2018; 51: S44-S48https://doi.org/10.1016/j.jelectrocard.2018.07.024Crossref PubMed Scopus (47) Google Scholar To that end, the new algorithm to detect VT discussed in this study is promising and integrates multichannel ECG waveforms and uses other physiologic data such as simultaneous drops in invasive BP, and SpO2. Additionally, the study includes a large data set with 5,679 ICU admissions with 572,500 hours of continuous ECG monitoring over a time span of 19 months. While there are acknowledged limitations including the retrospective approach and need for prospective studies, the authors and their research team should be commended for their commitment to this important work and decades of research aimed at reducing false alarms and alarm fatigue. The new algorithm also holds promise for accurately identifying VT patients at greatest risk for mortality and we look forward to further refinements addressing the sensitivity and specificity for detection of true VT. Scientists in the biomedical engineering community, such as Zhu and colleagues should also be encouraged to continue their work on novel methods to reduce noise and false alarms.11Zhu W. Qiu L. Cai W. et al.A novel method to reduce false alarms in ECG diagnostic systems: capture and quantification of noisy signals.Physiol Meas. 2021; 42 (Published 2021 Jul 28. doi:10.1088/1361-6579/abf9f4)https://doi.org/10.1088/1361-6579/abf9f4Crossref Scopus (4) Google Scholar In summary, false alarms and alarm fatigue continue to be an ongoing issue in the hospital setting and there is a need for ongoing research, translation of current basic science, and data-driven interventions to ensure optimal patient safety. 4Pelter M.M. Suba S. Sandoval C. et al.Actionable Ventricular Tachycardia During In-Hospital ECG Monitoring and Its Impact on Alarm Fatigue.Crit Pathw Cardiol. 2020; 19: 79-86https://doi.org/10.1097/HPC.0000000000000216Crossref PubMed Scopus (4) Google Scholar, 8The Joint Commission. Org. 2023 National Safety Goals/ https://www.jointcommission.org/-/media/tjc/documents/standards/national-patient-safety-goals/2023/hap-npsg-simplified-2023-july.pdfGoogle Scholar. Ventricular tachycardia and in-hospital mortality in the intensive care unitHeart Rhythm O2PreviewContinuous electrocardiographic (ECG) monitoring is used to identify ventricular tachycardia (VT), but false alarms occur frequently.