Dataset Information

Machine learning prediction of refractory ventricular fibrillation in out-of-hospital cardiac arrest using features available to EMS.

ABSTRACT:

Background

Shock-refractory ventricular fibrillation (VF) or ventricular tachycardia (VT) is a treatment challenge in out-of-hospital cardiac arrest (OHCA). This study aimed to develop and validate machine learning models that could be implemented by emergency medical services (EMS) to predict refractory VF/VT in OHCA patients.

Methods

This was a retrospective study examining adult non-traumatic OHCA patients brought into the emergency department by Singapore EMS from the Pan-Asian Resuscitation Outcomes Study (PAROS) registry. Data from April 2010 to March 2020 were extracted for this study. Refractory VF/VT was defined as VF/VT persisting or recurring after at least one shock. Features were selected based on expert clinical opinion and availability to dispatch prior to arrival at scene. Multivariable logistic regression (MVR), LASSO and random forest (RF) models were investigated. Model performance was evaluated using receiver operator characteristic (ROC) area under curve (AUC) analysis and calibration plots.

Results

20,713 patients were included in this study, of which 860 (4.1%) fulfilled the criteria for refractory VF/VT. All models performed comparably and were moderately well-calibrated. ROC-AUC were 0.732 (95% CI, 0.695 - 0.769) for MVR, 0.738 (95% CI, 0.701 - 0.774) for LASSO, and 0.731 (95% CI, 0.690 - 0.773) for RF. The shared important predictors across all models included male gender and public location.

Conclusion

The machine learning models developed have potential clinical utility to improve outcomes in cases of refractory VF/VT OHCA. Prediction of refractory VF/VT prior to arrival at patient's side may allow for increased options for intervention both by EMS and tertiary care centres.

SUBMITTER: Rahadian RE

PROVIDER: S-EPMC10963854 | biostudies-literature | 2024 Jun

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Machine learning prediction of refractory ventricular fibrillation in out-of-hospital cardiac arrest using features available to EMS.

Rahadian Rayhan Erlangga RE Okada Yohei Y Shahidah Nur N Hong Dehan D Ng Yih Yng YY Chia Michael Y C MYC Gan Han Nee HN Leong Benjamin S H BSH Mao Desmond R DR Ng Wei Ming WM Doctor Nausheen Edwin NE Ong Marcus Eng Hock MEH

Resuscitation plus 20240318

<h4>Background</h4>Shock-refractory ventricular fibrillation (VF) or ventricular tachycardia (VT) is a treatment challenge in out-of-hospital cardiac arrest (OHCA). This study aimed to develop and validate machine learning models that could be implemented by emergency medical services (EMS) to predict refractory VF/VT in OHCA patients.<h4>Methods</h4>This was a retrospective study examining adult non-traumatic OHCA patients brought into the emergency department by Singapore EMS from the Pan-Asia ...[more]

PMID: 38533482

Similar Datasets

Project description:BackgroundAmong patients with out-of-hospital cardiac arrest (OHCA) and ventricular fibrillation, more than half present with refractory ventricular fibrillation unresponsive to initial standard advanced cardiac life support (ACLS) treatment. We did the first randomised clinical trial in the USA of extracorporeal membrane oxygenation (ECMO)-facilitated resuscitation versus standard ACLS treatment in patients with OHCA and refractory ventricular fibrillation.MethodsFor this phase 2, single centre, open-label, adaptive, safety and efficacy randomised clinical trial, we included adults aged 18-75 years presenting to the University of Minnesota Medical Center (MN, USA) with OHCA and refractory ventricular fibrillation, no return of spontaneous circulation after three shocks, automated cardiopulmonary resuscitation with a Lund University Cardiac Arrest System, and estimated transfer time shorter than 30 min. Patients were randomly assigned to early ECMO-facilitated resuscitation or standard ACLS treatment on hospital arrival by use of a secure schedule generated with permuted blocks of randomly varying block sizes. Allocation concealment was achieved by use of a randomisation schedule that required scratching off an opaque layer to reveal assignment. The primary outcome was survival to hospital discharge. Secondary outcomes were safety, survival, and functional assessment at hospital discharge and at 3 months and 6 months after discharge. All analyses were done on an intention-to-treat basis. The study qualified for exception from informed consent (21 Code of Federal Regulations 50.24). The ARREST trial is registered with ClinicalTrials.gov, NCT03880565.FindingsBetween Aug 8, 2019, and June 14, 2020, 36 patients were assessed for inclusion. After exclusion of six patients, 30 were randomly assigned to standard ACLS treatment (n=15) or to early ECMO-facilitated resuscitation (n=15). One patient in the ECMO-facilitated resuscitation group withdrew from the study before discharge. The mean age was 59 years (range 36-73), and 25 (83%) of 30 patients were men. Survival to hospital discharge was observed in one (7%) of 15 patients (95% credible interval 1·6-30·2) in the standard ACLS treatment group versus six (43%) of 14 patients (21·3-67·7) in the early ECMO-facilitated resuscitation group (risk difference 36·2%, 3·7-59·2; posterior probability of ECMO superiority 0·9861). The study was terminated at the first preplanned interim analysis by the National Heart, Lung, and Blood Institute after unanimous recommendation from the Data Safety Monitoring Board after enrolling 30 patients because the posterior probability of ECMO superiority exceeded the prespecified monitoring boundary. Cumulative 6-month survival was significantly better in the early ECMO group than in the standard ACLS group. No unanticipated serious adverse events were observed.InterpretationEarly ECMO-facilitated resuscitation for patients with OHCA and refractory ventricular fibrillation significantly improved survival to hospital discharge compared with standard ACLS treatment.FundingNational Heart, Lung, and Blood Institute.

Project description:BackgroundEvidence of the association between AMplitude Spectral Area (AMSA) of ventricular fibrillation and outcome after out-of-hospital cardiac arrest (OHCA) is limited to short-term follow-up. In this study, we assess whether AMSA can stratify the risk of death or poor neurological outcome at 30 days and 1 year after OHCA in patients with an initial shockable rhythm or with an initial non-shockable rhythm converted to a shockable one.MethodsThis is a multicentre retrospective study of prospectively collected data in two European Utstein-based OHCA registries. We included all cases of OHCAs with at least one manual defibrillation. AMSA values were calculated after data extraction from the monitors/defibrillators used in the field by using a 2-s pre-shock electrocardiogram interval. The first detected AMSA value, the maximum value, the average value, and the minimum value were computed, and their outcome prediction accuracy was compared. Multivariable Cox regression models were run for both 30-day and 1-year deaths or poor neurological outcomes. Neurological cerebral performance category 1-2 was considered a good neurological outcome.ResultsOut of the 578 patients included, 494 (85%) died and 10 (2%) had a poor neurological outcome at 30 days. All the AMSA values considered (first value, maximum, average, and minimum) were significantly higher in survivors with good neurological outcome at 30 days. The average AMSA showed the highest area under the receiver operating characteristic curve (0.778, 95% CI: 0.7-0.8, p < 0.001). After correction for confounders, the highest tertiles of average AMSA (T3 and T2) were significantly associated with a lower risk of death or poor neurological outcome compared with T1 both at 30 days (T2: HR 0.6, 95% CI: 0.4-0.9, p = 0.01; T3: HR 0.6, 95% CI: 0.4-0.9, p = 0.02) and at 1 year (T2: HR 0.6, 95% CI: 0.4-0.9, p = 0.01; T3: HR 0.6, 95% CI: 0.4-0.9, p = 0.01). Among survivors at 30 days, a higher AMSA was associated with a lower risk of mortality or poor neurological outcome at 1 year (T3: HR 0.03, 95% CI: 0-0.3, p = 0.02).DiscussionLower AMSA values were significantly and independently associated with the risk of death or poor neurological outcome at 30 days and at 1 year in OHCA patients with either an initial shockable rhythm or a conversion rhythm from non-shockable to shockable. The average AMSA value had the strongest association with prognosis.

Project description:PurposeOut-of-hospital cardiac arrest (OHCA) has a poor prognosis, with an overall survival rate of about 5% at discharge. Shockable rhythm cardiac arrests (ventricular fibrillation (VF) and pulseless ventricular tachycardia (VT)) have a better prognosis. In case of shockable rhythm, treatment is based on defibrillation, and thereafter, in case of failure of 3 external electric shocks (EES), on direct intravenous administration of 300 mg amiodarone, or lidocaine when amiodarone is unavailable or inefficient. During surgical procedures under extracorporeal circulation, a high potassium cardioplegia solution is administered to interrupt cardiac activity and facilitate surgical procedure. By extension, direct intravenous administration of potassium chloride (KCl) has been shown to convert VF, resulting in return to a hemodynamically efficient organized heart rate within a few minutes. The aim of this study is to provide clinical evidence that direct intravenous injection of KCl, into a patient presenting with OHCA due to refractory VF although 3 EES, should interrupt this VF and then allow rapid restauration of an organized heart rhythm, and thus return of spontaneous circulation (ROSC).MethodsA multicenter, prospective, single group, phase 2 study will be conducted on 81 patients presenting with refractory VF. After failure of 3 EES, each patient will receive direct intravenous injection of 20 mmol KCl instead of amiodarone. The primary outcome will be survival rate at hospital admission. Major secondary outcomes will include ROSC and time to ROSC in the prehospital setting, number of VF recidivism after KCl injection, survival rate at hospital discharge with a good neurologic prognostic, and survival rate 3 months after hospital discharge with a good neurologic prognostic.ResultsNo patient is currently included in the study.DiscussionConventional guideline strategy based on antiarrhythmic drug administration, i.e. amiodarone or lidocaine, for OHCA due to shockable rhythm, has not yet demonstrated an increase in survival at hospital admission or at hospital discharge. This may be related to the major cardiodepressant effect of those drugs.Trial registrationClinicalTrials.gov Identifier: NCT04316611. Registered on March 2020. AP-HP180577 / N° EUDRACT: 2019-002544-24. Funded by the French Health Ministry. https://clinicaltrials.gov/ct2/show/NCT04316611.

Project description:BACKGROUND:Post-resuscitation hemodynamic instability following out-of-hospital cardiac arrest (OHCA) may occur from myocardial dysfunction underlying cardiogenic shock and/or inflammation-mediated distributive shock. Distinguishing the predominant shock subtype with widely available clinical metrics may have prognostic and therapeutic value. METHODS:A two-hospital cohort was assembled of patients in shock following OHCA. Left ventricular ejection fraction (LVEF) was assessed via echocardiography or cardiac ventriculography within 1 day post arrest and used to delineate shock physiology. The study evaluated whether higher LVEF, indicating distributive-predominant shock physiology, was associated with neurocognitive outcome (primary endpoint), survival, and duration of multiple organ failures. The study also investigated whether volume resuscitation exhibited a subtype-specific association with outcome. RESULTS:Of 162 patients with post-resuscitation shock, 48% had normal LVEF (> 40%), consistent with distributive shock physiology. Higher LVEF was associated with less favorable neurocognitive outcome (OR 0.74, 95% CI 0.58-0.94 per 10% increase in LVEF; p = 0.01). Higher LVEF also was associated with worse survival (OR 0.81, 95% CI 0.67-0.97; p = 0.02) and fewer organ failure-free days (β = - 0.67, 95% CI - 1.28 to - 0.06; p = 0.03). Only 51% of patients received a volume challenge of at least 30 ml/kg body weight in the first 6 h post arrest, and the volume received did not differ by LVEF. Greater volume resuscitation in the first 6 h post arrest was associated with favorable neurocognitive outcome (OR 1.59, 95% CI 0.99-2.55 per liter; p = 0.03) and survival (OR 1.44, 95% CI 1.02-2.04; p = 0.02) among patients with normal LVEF but not low LVEF. CONCLUSIONS:In post-resuscitation shock, higher LVEF-indicating distributive shock physiology-was associated with less favorable neurocognitive outcome, fewer days without organ failure, and higher mortality. Greater early volume resuscitation was associated with more favorable neurocognitive outcome and survival in patients with this shock subtype. Additional studies with repeated measures of complementary hemodynamic parameters are warranted to validate the clinical utility for subtyping post-resuscitation shock.

Dataset Information

Machine learning prediction of refractory ventricular fibrillation in out-of-hospital cardiac arrest using features available to EMS.

Background

Methods

Results

Conclusion

Publications

Machine learning prediction of refractory ventricular fibrillation in out-of-hospital cardiac arrest using features available to EMS.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets