RESUMEN
OBJECTIVES: This study assesses the feasibility, inter-rater reliability, and accuracy of using OpenAI's ChatGPT-4 and Google's Gemini Ultra large language models (LLMs), for Emergency Medical Services (EMS) quality assurance. The implementation of these LLMs for EMS quality assurance has the potential to significantly reduce the workload on medical directors and quality assurance staff by automating aspects of the processing and review of patient care reports. This offers the potential for more efficient and accurate identification of areas requiring improvement, thereby potentially enhancing patient care outcomes. METHODS: Two expert human reviewers, ChatGPT GPT-4, and Gemini Ultra assessed and rated 150 consecutively sampled and anonymized prehospital records from 2 large urban EMS agencies for adherence to 2020 National Association of State EMS metrics for cardiac care. We evaluated the accuracy of scoring, inter-rater reliability, and review efficiency. The inter-rater reliability for the dichotomous outcome of each EMS metric was measured using the kappa statistic. RESULTS: Human reviewers showed high interrater reliability, with 91.2% agreement and a kappa coefficient 0.782 (0.654-0.910). ChatGPT-4 achieved substantial agreement with human reviewers in EKG documentation and aspirin administration (76.2% agreement, kappa coefficient 0.401 (0.334-0.468), but performance varied across other metrics. Gemini Ultra's evaluation was discontinued due to poor performance. No significant differences were observed in median review times: 01:28 min (IQR 1:12 - 1:51 min) per human chart review, 01:24 min (IQR 01:09 - 01:53 min) per ChatGPT-4 chart review (p = 0.46), and 01:50 min (IQR 01:10-03:34 min) per Gemini Ultra review (p = 0.06). CONCLUSIONS: Large language models demonstrate potential in supporting quality assurance by effectively and objectively extracting data elements. However, their accuracy in interpreting non-standardized and time-sensitive details remains inferior to human evaluators. Our findings suggest that current LLMs may best offer supplemental support to the human review processes, but their current value remains limited. Enhancements in LLM training and integration are recommended for improved and more reliable performance in the quality assurance processes.
RESUMEN
BACKGROUND: Despite significant improvements in cardiopulmonary resuscitation, sudden cardiac arrest is one of the leading causes of mortality in the United States. Ultrasound is a widely available tool that can be used to evaluate the presence of cardiac wall motion during cardiac arrest. Several clinical studies have evaluated the use of ultrasound to visualize cardiac motion as a predictor of mortality in cardiac arrest patients. However, there are limited data summarizing the prognostic value of point of care ultrasound evaluation during resuscitation. We performed a systematic literature review of the existing evidence examining the clinical utility of point-of-care ultrasound evaluation of cardiac wall motion as a predictor of cardiac resuscitation outcomes. METHODS/RESULTS: We performed a systematic PubMed search of clinical studies up to July 23, 2019 evaluating point-of-care sonographic cardiac motion as a predictor of mortality following cardiac resuscitation. We included studies written in English that reviewed short-term outcomes and included adult populations. Fifteen clinical studies met inclusion criteria for assessing cardiac wall motion with point-of-care ultrasound and outcomes following cardiac resuscitation. Fourteen of the fifteen studies showed a statistically significant correlation between the presence of cardiac motion on ultrasound and short-term survival. This was most evident in patients with ventricular fibrillation or ventricular tachycardia as a presenting rhythm. Absence of cardiac motion non-survival. The data were pooled and the overall pooled odds ratio for return of spontaneous circulation in the presence of cardiac motion during CPR was 12.4 +/1 2.7 (p < 0.001). CONCLUSION: Evaluation of cardiac motion on transthoracic echocardiogram is a valuable tool in the prediction of short-term cardiac resuscitation outcomes. Given the safety and availability of ultrasound in the emergency department, it is reasonable to apply point-of-care ultrasound to cardiopulmonary resuscitation as long as its use does not interrupt resuscitation.
Asunto(s)
Reanimación Cardiopulmonar/métodos , Paro Cardíaco/diagnóstico , Sistemas de Atención de Punto , Ultrasonografía/métodos , Paro Cardíaco/terapia , Humanos , PronósticoRESUMEN
BACKGROUND: Early administration of adrenaline is associated with improved survival after out-of-hospital cardiac arrest (OHCA). Delays in vascular access may impact the timely delivery of adrenaline. Novel methods for administering adrenaline before vascular access may enhance survival. The objective of this study was to determine whether an initial intramuscular (IM) adrenaline dose followed by standard IV/IO adrenaline is associated with improved survival after OHCA. METHODS STUDY DESIGN: We conducted a before-and-after study of the implementation of an early, first-dose IM adrenaline EMS protocol for adult OHCAs. The pre-intervention period took place between January 2010 and October 2019. The post-intervention period was between November 2019 and May 2024. SETTING: Single-center urban, two-tiered EMS agency. PARTICIPANTS: Adult, nontraumatic OHCA meeting criteria for adrenaline use. INTERVENTION: Single dose (5 mg) IM adrenaline. All other care, including subsequent IV or IO adrenaline, followed international guidelines. MAIN OUTCOMES AND MEASURES: The primary outcome was survival to hospital discharge. Secondary outcomes were time from EMS arrival to the first dose of adrenaline, survival to hospital admission, and favorable neurologic function at discharge. RESULTS: Among 1405 OHCAs, 420 (29.9%) received IM adrenaline and 985 (70.1%) received usual care. Fifty-two patients received the first dose of adrenaline through the IV or IO route within the post-intervention period and were included in the standard care group analysis. Age was younger and bystander CPR was higher in the IM adrenaline group. All other characteristics were similar between IM and standard care cohorts. Time to adrenaline administration was faster for the IM cohort [(median 4.3 min (IQR 3.0-6.0) vs. 7.8 min (IQR 5.8-10.4)]. Compared with standard care, IM adrenaline was associated with improved survival to hospital admission (37.1% vs. 31.6%; aOR 1.37, 95% CI 1.06-1.77), hospital survival (11.0% vs 7.0%; aOR 1.73, 95% CI 1.10-2.71) and favorable neurologic status at hospital discharge (9.8% vs 6.2%; aOR 1.72, 95% CI 1.07-2.76). CONCLUSION: In this single-center before-and-after implementation study, an initial IM dose of adrenaline as an adjunct to standard care was associated with improved survival to hospital admission, survival to hospital discharge, and functional survival. A randomized controlled trial is needed to fully assess the potential benefit of IM adrenaline delivery in OHCA.
Asunto(s)
Reanimación Cardiopulmonar , Servicios Médicos de Urgencia , Epinefrina , Paro Cardíaco Extrahospitalario , Paro Cardíaco Extrahospitalario/mortalidad , Paro Cardíaco Extrahospitalario/tratamiento farmacológico , Paro Cardíaco Extrahospitalario/terapia , Humanos , Epinefrina/administración & dosificación , Masculino , Femenino , Inyecciones Intramusculares , Persona de Mediana Edad , Anciano , Servicios Médicos de Urgencia/métodos , Reanimación Cardiopulmonar/métodos , Tiempo de Tratamiento , Estudios Controlados Antes y Después , Vasoconstrictores/administración & dosificaciónRESUMEN
Objectives: Prior research indicates sex disparities in the incidence and outcomes of out-of-hospital cardiac arrest (OHCA). This study investigates the presence of such differences in Salt Lake City, Utah. Methods: We analyzed data from the Salt Lake City Fire Department (2008â2023). We included adults with non-traumatic OHCA. We calculated the annual incidence of OHCA and examined sex-specific survival outcomes using multivariable logistic regression, adjusting for OHCA characteristics known to be associated with survival. Results: The annual incidence of OHCA was 76 per 100,000 person-years. Among the 894 OHCA cases included in the analysis, 67.5% were males, 37.3% achieved return of spontaneous circulation (ROSC), and 13.6% survived hospital discharge. Unadjusted analysis revealed that males had significantly higher OHCA in public locations (43.9% vs. 28.6%), witnessed arrests (54.5% vs. 47.8%), and shockable rhythms (33.3% vs. 22.9%). Males also showed higher rates of ROSC (37.5% vs. 36.9%), hospital discharge survival (14.5% vs. 11.7%), and neurologically intact survival. After adjusting for the OHCA characteristics, there was no significant differences between males and females in ROSC, survival to hospital discharge, and favorable neurological function with adjusted odds ratios (male vs. female) of 0.92 (95% confidence interval [CI] 0.73â1.16), 0.85 (95% CI 0.59â1.22), and 0.92 (95% CI 0.62â1.40), respectively. Conclusion: Approximately, 128 adults suffer OHCA in Salt Lake City annually. Males initially showed higher crude survival rates, but after adjusting for OHCA characteristics, no significant sex differences in survival outcomes were found. Enhancing OHCA characteristics could benefit both sexes. Investigations into the relationship between sex- and region-specific factors influencing OHCA outcomes are needed.
RESUMEN
Control of blood pressure, in terms of both absolute values and its variability, affects outcomes in ischemic stroke patients. However, it remains challenging to identify the mechanisms that lead to poor outcomes or evaluate measures by which these effects can be mitigated because of the prohibitive limitations inherent to human data. In such cases, animal models can be utilized to conduct rigorous and reproducible evaluations of diseases. Here we report refinement of a previously described model of ischemic stroke in rabbits that is augmented with continuous blood pressure recording to assess the impacts of modulation on blood pressure. Under general anesthesia, femoral arteries are exposed through surgical cutdowns to place arterial sheaths bilaterally. Under fluoroscopic visualization and roadmap guidance, a microcatheter is advanced into an artery of the posterior circulation of the brain. An angiogram is performed by injecting the contralateral vertebral artery to confirm occlusion of the target artery. With the occlusive catheter remaining in position for a fixed duration, blood pressure is continuously recorded to allow for tight titration of blood pressure manipulations, whether through mechanical or pharmacological means. At the completion of the occlusion interval, the microcatheter is removed, and the animal is maintained under general anesthesia for a prescribed length of reperfusion. For acute studies, the animal is then euthanized and decapitated. The brain is harvested and processed to measure the infarct volume under light microscopy and further assessed with various histopathological stains or spatial transcriptomic analysis. This protocol provides a reproducible model that can be utilized for more thorough preclinical studies on the effects of blood pressure parameters during ischemic stroke. It also facilitates effective preclinical evaluation of novel neuroprotective interventions that might improve care for ischemic stroke patients.