Pub Date : 2024-10-20DOI: 10.1016/j.ajem.2024.10.034
Adem Az
{"title":"From memory to mastery: Optimizing AI models for ECG diagnostics in clinical practice.","authors":"Adem Az","doi":"10.1016/j.ajem.2024.10.034","DOIUrl":"https://doi.org/10.1016/j.ajem.2024.10.034","url":null,"abstract":"","PeriodicalId":55536,"journal":{"name":"American Journal of Emergency Medicine","volume":" ","pages":""},"PeriodicalIF":2.7,"publicationDate":"2024-10-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142513493","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-10-20DOI: 10.1016/j.ajem.2024.10.037
Arjun K Venkatesh, Jessica Duke, Silas Wong, Aman Shah, Craig Rothenberg, Amitkumar Patel, Wendy W Sun, Marc Shapiro, Andrew Ulrich, Vivek Parwani
{"title":"Quality improvement interventions to reduce coagulation testing overuse in the emergency department.","authors":"Arjun K Venkatesh, Jessica Duke, Silas Wong, Aman Shah, Craig Rothenberg, Amitkumar Patel, Wendy W Sun, Marc Shapiro, Andrew Ulrich, Vivek Parwani","doi":"10.1016/j.ajem.2024.10.037","DOIUrl":"https://doi.org/10.1016/j.ajem.2024.10.037","url":null,"abstract":"","PeriodicalId":55536,"journal":{"name":"American Journal of Emergency Medicine","volume":" ","pages":""},"PeriodicalIF":2.7,"publicationDate":"2024-10-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142570498","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-10-19DOI: 10.1016/j.ajem.2024.10.032
Peter B. Smulowitz MD, MPH
{"title":"Balancing admissions from the ED and from elective surgeries: Caring for the entire community","authors":"Peter B. Smulowitz MD, MPH","doi":"10.1016/j.ajem.2024.10.032","DOIUrl":"10.1016/j.ajem.2024.10.032","url":null,"abstract":"","PeriodicalId":55536,"journal":{"name":"American Journal of Emergency Medicine","volume":"87 ","pages":"Pages 63-64"},"PeriodicalIF":2.7,"publicationDate":"2024-10-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142586132","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-10-18DOI: 10.1016/j.ajem.2024.10.031
Mustafa Comoglu, Fatih Acehan, Osman Inan, Burak Furkan Demir, Yusufcan Yılmaz, Enes Seyda Sahiner
Background
It is important to predict which patients may require renal replacement therapy (RRT) at the time of initial presentation after crush injuries. There is limited data in the literature examining the predictors of RRT.
Methods
This study was conducted by evaluating 2232 patients who presented to our hospital following two major earthquakes of magnitudes 7.6 and 7.7 Mw that occurred in Kahramanmaras, Turkey, on February 6, 2023. A total of 314 patients who were hospitalized upon being rescued from the rubble and had a creatine kinase (CK) level above 1000 U/L were included in the final analysis. Factors predicting the need for RRT were investigated, and a dialysis score was developed for this prediction.
Results
Of the 314 patients included in the study, 95 (30.2 %) developed acute kidney injury (AKI). RRT was performed on 68 (21.6 %) patients. The optimal cut-off value of CK for the prediction of AKI was 23,000 U/L. Multivariate analysis revealed that factors predicting RRT were the number of traumatized sides (odds ratio [OR]: 2.2, 95 % confidence interval [CI]: 1.09–4.39, p = 0.026), albumin (OR:0.11, 95 % CI: 0.04–0.32, p < 0.001), and CK (OR: 1.00, 95 % CI 1.00–1.00, p < 0.001). A dialysis score was developed ranging from 0 to 7 based on the number of traumatized sides, albumin, and CK. The area under the curve (AUC) of the dialysis score in receiver operating characteristic analysis was 0.974. A dialysis score of 4 or higher had a sensitivity of 97.1 % and a specificity of 89.4 % for predicting the need for RRT.
Conclusions
The dialysis score predicts the need for RRT quite well. The simplicity of use and high sensitivity and specificity of this score in earthquake-related crush injuries will greatly facilitate clinicians in patient triage and follow-up.
{"title":"A new score predicting renal replacement therapy in patients with crush injuries: Analysis of a major earthquake","authors":"Mustafa Comoglu, Fatih Acehan, Osman Inan, Burak Furkan Demir, Yusufcan Yılmaz, Enes Seyda Sahiner","doi":"10.1016/j.ajem.2024.10.031","DOIUrl":"10.1016/j.ajem.2024.10.031","url":null,"abstract":"<div><h3>Background</h3><div>It is important to predict which patients may require renal replacement therapy (RRT) at the time of initial presentation after crush injuries. There is limited data in the literature examining the predictors of RRT.</div></div><div><h3>Methods</h3><div>This study was conducted by evaluating 2232 patients who presented to our hospital following two major earthquakes of magnitudes 7.6 and 7.7 Mw that occurred in Kahramanmaras, Turkey, on February 6, 2023. A total of 314 patients who were hospitalized upon being rescued from the rubble and had a creatine kinase (CK) level above 1000 U/L were included in the final analysis. Factors predicting the need for RRT were investigated, and a dialysis score was developed for this prediction.</div></div><div><h3>Results</h3><div>Of the 314 patients included in the study, 95 (30.2 %) developed acute kidney injury (AKI). RRT was performed on 68 (21.6 %) patients. The optimal cut-off value of CK for the prediction of AKI was 23,000 U/L. Multivariate analysis revealed that factors predicting RRT were the number of traumatized sides (odds ratio [OR]: 2.2, 95 % confidence interval [CI]: 1.09–4.39, <em>p</em> = 0.026), albumin (OR:0.11, 95 % CI: 0.04–0.32, <em>p</em> < 0.001), and CK (OR: 1.00, 95 % CI 1.00–1.00, p < 0.001). A dialysis score was developed ranging from 0 to 7 based on the number of traumatized sides, albumin, and CK. The area under the curve (AUC) of the dialysis score in receiver operating characteristic analysis was 0.974. A dialysis score of 4 or higher had a sensitivity of 97.1 % and a specificity of 89.4 % for predicting the need for RRT.</div></div><div><h3>Conclusions</h3><div>The dialysis score predicts the need for RRT quite well. The simplicity of use and high sensitivity and specificity of this score in earthquake-related crush injuries will greatly facilitate clinicians in patient triage and follow-up.</div></div>","PeriodicalId":55536,"journal":{"name":"American Journal of Emergency Medicine","volume":"87 ","pages":"Pages 1-7"},"PeriodicalIF":2.7,"publicationDate":"2024-10-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142513497","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-10-18DOI: 10.1016/j.ajem.2024.10.016
Justin A. Seltzer MD , Garret A. Winkler MD , Jeremy Hardin MD , Henrik Galust MD , Timothy E. Albertson MD, PhD, MPH , Rais Vohra MD , Craig Smollin MD , Edward Castillo PhD , Daniel Lasoff MD , Richard F. Clark MD
Rattlesnake envenomations account for many of the Crotalid envenomations in the United States annually. Two antivenoms are currently available to treat Crotalid envenomation in this country: Crotalidae-polyvalent ovine immune Fab antivenom (CroFab®; FabAV) and Crotalidae equine immune F(ab’)₂ antivenom (ANAVIP®; F(ab’)₂AV). Few studies have compared the adverse effect rates for each.
We performed a retrospective chart review of rattlesnake envenomations called to the California Poison Control System from October 2018 to August 2022. Those treated at healthcare facilities with either antivenom were included. Those treated with both antivenoms were excluded.
Records were obtained from the poison center electronic medical records system. Demographic and clinical data were abstracted. “Severe” adverse events were defined as multi-organ system involvement, swelling of the patient's airway, and/or hemodynamic instability. All others were categorized as “non-severe.”
A total of 481 cases were included with 360 treated with FabAV and 121 with F(ab’)₂AV. The median age was 47 and 46 years, and 72 % and 73 % were male, respectively. Clinical signs and symptoms of envenomation were similar in each group.
The FabAV group received a median of six vials. The F(ab’)₂AV group received a median of 10 vials, based on the recommended loading doses of FabAV and F(ab’)2AV. Following antivenom administration, 18 individual acute non-severe AEs were reported in 12 FabAV-treated patients. Two acute non-severe AEs were reported in two F(ab’)₂AV-treated patients. Rash or urticaria was the most commonly reported adverse effect in both groups after antivenom administration. Five patients (1.5 %) had severe adverse events reported in the poison center records following FabAV administration, and none were reported following F(ab’)₂AV administration (p = 0.025).
Overall, our poison center data suggests the rate of adverse events is low following the use of either antivenom. Our findings are limited by the lack of consistent timing data, a smaller F(ab’)₂AV cohort, retrospective format, and use of poison center data.
{"title":"Acute adverse effects of F(ab’)₂AV and FabAV use for rattlesnake Envenomations: A four-year poison center study","authors":"Justin A. Seltzer MD , Garret A. Winkler MD , Jeremy Hardin MD , Henrik Galust MD , Timothy E. Albertson MD, PhD, MPH , Rais Vohra MD , Craig Smollin MD , Edward Castillo PhD , Daniel Lasoff MD , Richard F. Clark MD","doi":"10.1016/j.ajem.2024.10.016","DOIUrl":"10.1016/j.ajem.2024.10.016","url":null,"abstract":"<div><div>Rattlesnake envenomations account for many of the Crotalid envenomations in the United States annually. Two antivenoms are currently available to treat Crotalid envenomation in this country: Crotalidae-polyvalent ovine immune Fab antivenom (CroFab®; FabAV) and Crotalidae equine immune F(ab’)₂ antivenom (ANAVIP®; F(ab’)₂AV). Few studies have compared the adverse effect rates for each.</div><div>We performed a retrospective chart review of rattlesnake envenomations called to the California Poison Control System from October 2018 to August 2022. Those treated at healthcare facilities with either antivenom were included. Those treated with both antivenoms were excluded.</div><div>Records were obtained from the poison center electronic medical records system. Demographic and clinical data were abstracted. “Severe” adverse events were defined as multi-organ system involvement, swelling of the patient's airway, and/or hemodynamic instability. All others were categorized as “non-severe.”</div><div>A total of 481 cases were included with 360 treated with FabAV and 121 with F(ab’)₂AV. The median age was 47 and 46 years, and 72 % and 73 % were male, respectively. Clinical signs and symptoms of envenomation were similar in each group.</div><div>The FabAV group received a median of six vials. The F(ab’)₂AV group received a median of 10 vials, based on the recommended loading doses of FabAV and F(ab’)<sub>2</sub>AV. Following antivenom administration, 18 individual acute non-severe AEs were reported in 12 FabAV-treated patients. Two acute non-severe AEs were reported in two F(ab’)₂AV-treated patients. Rash or urticaria was the most commonly reported adverse effect in both groups after antivenom administration. Five patients (1.5 %) had severe adverse events reported in the poison center records following FabAV administration, and none were reported following F(ab’)₂AV administration (<em>p</em> = 0.025).</div><div>Overall, our poison center data suggests the rate of adverse events is low following the use of either antivenom. Our findings are limited by the lack of consistent timing data, a smaller F(ab’)₂AV cohort, retrospective format, and use of poison center data.</div></div>","PeriodicalId":55536,"journal":{"name":"American Journal of Emergency Medicine","volume":"87 ","pages":"Pages 28-31"},"PeriodicalIF":2.7,"publicationDate":"2024-10-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142513498","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-10-17DOI: 10.1016/j.ajem.2024.10.027
Kevin A Padrez, Benjamin L Stix, Cody A Cunningham, Abdelmohaymin Abdalla, Marisa Oishi, Vanessa Cardy, Sean Patrick Nordt
Background: Unintentional plant ingestions and poisoning are common. Generally, these ingestions are asymptomatic or minimally symptomatic. Increased toxicity is often associated with the foraging for plants incorrectly identified as edible plants. We present a case series of seven family members poisoned by suspected Death Camas, also known as Meadow Death Camas, (Toxicoscordion venenosum), when the plant was misidentified as edible while foraging for Wild Onion (Allium canadense).
Case series: Five children and two adults presented to the emergency department after eating bulbs of suspected Death Camas (Toxicoscordion venenosum). Symptoms began within 30 to 60 min including nausea, vomiting, and abdominal pain. This was followed by bradycardia and hypotension consistent with Death Camas poisoning from alpha blockade, sodium channel activation, and increased vagal activity from the toxins. All of the patients required admission to the intensive care unit. Six patients were treated with supportive care alone and two patients required vasopressor support. An 89-year-old female developed a wide complex tachycardia and was treated with amiodarone. All patients had resolution of toxicity within 24 h and were discharged to home. WHY SHOULD AN EMERGENCY PHYSICIAN BE AWARE OF THIS?: Foodborne illness is a common presentation highlighted by signs and symptoms manifesting in multiple people eating the same meal. The misidentification of toxic plants as edible is associated with acute onset of signs and symptoms of toxicity. The toxins associated with Death Camas (Toxicoscordion venenosum), commonly thought to be edible Wild Onion (Allium canadense), are unique as these can cause gastrointestinal effects (e.g., nausea, vomiting, diarrhea), and cardiovascular effects (e.g., bradycardia, hypotension), which are often refractory to supportive therapy alone (e.g., crystalloid fluids), and can require atropine and vasopressors.
{"title":"Acute death Camas (Toxicoscordion venenosum) plant poisoning in seven family members following foraging for wild onions.","authors":"Kevin A Padrez, Benjamin L Stix, Cody A Cunningham, Abdelmohaymin Abdalla, Marisa Oishi, Vanessa Cardy, Sean Patrick Nordt","doi":"10.1016/j.ajem.2024.10.027","DOIUrl":"https://doi.org/10.1016/j.ajem.2024.10.027","url":null,"abstract":"<p><strong>Background: </strong>Unintentional plant ingestions and poisoning are common. Generally, these ingestions are asymptomatic or minimally symptomatic. Increased toxicity is often associated with the foraging for plants incorrectly identified as edible plants. We present a case series of seven family members poisoned by suspected Death Camas, also known as Meadow Death Camas, (Toxicoscordion venenosum), when the plant was misidentified as edible while foraging for Wild Onion (Allium canadense).</p><p><strong>Case series: </strong>Five children and two adults presented to the emergency department after eating bulbs of suspected Death Camas (Toxicoscordion venenosum). Symptoms began within 30 to 60 min including nausea, vomiting, and abdominal pain. This was followed by bradycardia and hypotension consistent with Death Camas poisoning from alpha blockade, sodium channel activation, and increased vagal activity from the toxins. All of the patients required admission to the intensive care unit. Six patients were treated with supportive care alone and two patients required vasopressor support. An 89-year-old female developed a wide complex tachycardia and was treated with amiodarone. All patients had resolution of toxicity within 24 h and were discharged to home. WHY SHOULD AN EMERGENCY PHYSICIAN BE AWARE OF THIS?: Foodborne illness is a common presentation highlighted by signs and symptoms manifesting in multiple people eating the same meal. The misidentification of toxic plants as edible is associated with acute onset of signs and symptoms of toxicity. The toxins associated with Death Camas (Toxicoscordion venenosum), commonly thought to be edible Wild Onion (Allium canadense), are unique as these can cause gastrointestinal effects (e.g., nausea, vomiting, diarrhea), and cardiovascular effects (e.g., bradycardia, hypotension), which are often refractory to supportive therapy alone (e.g., crystalloid fluids), and can require atropine and vasopressors.</p>","PeriodicalId":55536,"journal":{"name":"American Journal of Emergency Medicine","volume":" ","pages":""},"PeriodicalIF":2.7,"publicationDate":"2024-10-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142549142","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-10-16DOI: 10.1016/j.ajem.2024.10.029
Paulo Eduardo Souza Castelo Branco, Isabela Maurício Costa Carneiro, Adriane Helena Silva Franco, Amanda Prates de Oliveira
{"title":"Effect of intravenous ondansetron on QTc interval in children with gastroenteritis.","authors":"Paulo Eduardo Souza Castelo Branco, Isabela Maurício Costa Carneiro, Adriane Helena Silva Franco, Amanda Prates de Oliveira","doi":"10.1016/j.ajem.2024.10.029","DOIUrl":"https://doi.org/10.1016/j.ajem.2024.10.029","url":null,"abstract":"","PeriodicalId":55536,"journal":{"name":"American Journal of Emergency Medicine","volume":" ","pages":""},"PeriodicalIF":2.7,"publicationDate":"2024-10-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142481685","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-10-16DOI: 10.1016/j.ajem.2024.10.030
Erhan Arıkan, Faik Özel, İsa Ardahanlı
{"title":"Ondansetron and cardiac safety: Call for a comprehensive assessment.","authors":"Erhan Arıkan, Faik Özel, İsa Ardahanlı","doi":"10.1016/j.ajem.2024.10.030","DOIUrl":"https://doi.org/10.1016/j.ajem.2024.10.030","url":null,"abstract":"","PeriodicalId":55536,"journal":{"name":"American Journal of Emergency Medicine","volume":" ","pages":""},"PeriodicalIF":2.7,"publicationDate":"2024-10-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142481686","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-10-15DOI: 10.1016/j.ajem.2024.10.020
Fatma Tortum , Erdal Tekin , Bugra Kerget , Alperen Aksakal , Orhan Enes Tuncez
Introduction
This study investigated the feasibility of using the Roth score in the emergency setting to make hospitalization or discharge decisions for patients with acute exacerbations of chronic obstructive pulmonary disease (AECOPD).
Materials and methods
This study was conducted prospectively between March 1, 2023 and January 1, 2024 and included 101 patients with Group E chronic obstructive pulmonary disease who were treated in the emergency department of a tertiary hospital. The patients were categorized into two groups: those who were hospitalized and those who were discharged. The Roth score, determined by measuring patients' breath-holding times after forced inspiration and counting rhythmically, was measured in seconds and counted. Changes in Roth scores, arterial blood gas parameters, and transcutaneous oxygen saturation levels measured during AECOPD presentation and after appropriate treatment were examined.
Results
The study included 101 patients (57 males, 44 females) with a mean age of 61.4 years. After AECOPD treatment, the area under the curve for the Roth score was 0.937 s for the duration and 0.969 for the count. At a cut-off value of 9.5 s, the Roth score in seconds had a sensitivity of 92 % and a specificity of 75 %. At a cut-off value of 10.5, the Roth score had a sensitivity of 97 % and a specificity of 70 %.
Conclusion
The Roth score (only counts) increased in discharged patients after AECOPD treatment. It appears to be a viable method for predicting hospitalization or discharge decisions in patients with AECOPD who present to the emergency department.
{"title":"The use of the Roth score in emergency department for patients with acute exacerbation of chronic obstructive pulmonary disease","authors":"Fatma Tortum , Erdal Tekin , Bugra Kerget , Alperen Aksakal , Orhan Enes Tuncez","doi":"10.1016/j.ajem.2024.10.020","DOIUrl":"10.1016/j.ajem.2024.10.020","url":null,"abstract":"<div><h3>Introduction</h3><div>This study investigated the feasibility of using the Roth score in the emergency setting to make hospitalization or discharge decisions for patients with acute exacerbations of chronic obstructive pulmonary disease (AECOPD).</div></div><div><h3>Materials and methods</h3><div>This study was conducted prospectively between March 1, 2023 and January 1, 2024 and included 101 patients with Group E chronic obstructive pulmonary disease who were treated in the emergency department of a tertiary hospital. The patients were categorized into two groups: those who were hospitalized and those who were discharged. The Roth score, determined by measuring patients' breath-holding times after forced inspiration and counting rhythmically, was measured in seconds and counted. Changes in Roth scores, arterial blood gas parameters, and transcutaneous oxygen saturation levels measured during AECOPD presentation and after appropriate treatment were examined.</div></div><div><h3>Results</h3><div>The study included 101 patients (57 males, 44 females) with a mean age of 61.4 years. After AECOPD treatment, the area under the curve for the Roth score was 0.937 s for the duration and 0.969 for the count. At a cut-off value of 9.5 s, the Roth score in seconds had a sensitivity of 92 % and a specificity of 75 %. At a cut-off value of 10.5, the Roth score had a sensitivity of 97 % and a specificity of 70 %.</div></div><div><h3>Conclusion</h3><div>The Roth score (only counts) increased in discharged patients after AECOPD treatment. It appears to be a viable method for predicting hospitalization or discharge decisions in patients with AECOPD who present to the emergency department.</div></div>","PeriodicalId":55536,"journal":{"name":"American Journal of Emergency Medicine","volume":"86 ","pages":"Pages 135-140"},"PeriodicalIF":2.7,"publicationDate":"2024-10-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142481695","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-10-15DOI: 10.1016/j.ajem.2024.10.019
Michael Gottlieb MD, Evelyn Schraft MD, James O'Brien MD, Daven Patel MD, MPH
Introduction
Cardiac point-of-care ultrasound (POCUS) can evaluate for systolic and diastolic dysfunction to inform care in the Emergency Department (ED). However, accurate assessment can be limited by user experience. Artificial intelligence (AI) has been proposed as a model to increase the accuracy of cardiac POCUS. However, there is limited evidence of the accuracy of AI in the clinical environment. The objective of this study was to determine the diagnostic accuracy of AI for identifying systolic and diastolic dysfunction compared with expert reviewers.
Methods
This was a prospective, observational study of adult ED patients aged ≥45 years with risk factors for systolic and diastolic dysfunction. Ultrasound fellowship-trained physicians used an ultrasound machine with existing AI software and obtained parasternal long axis, parasternal short axis, and apical 4-chamber views of the heart. Systolic dysfunction was defined as ejection fraction (EF) < 50 % in at least two views using visual assessment or E-point septal separation >10 mm. Diastolic dysfunction was defined as an E:A < 0.8, or ≥ 2 of the following: septal e' < 7 cm/s or lateral e' < 10 cm/s, E:e' > 14, or left atrial volume > 34 mL/m2. AI was subsequently used to measure EF, E, A, septal e', and lateral e' velocities. The gold standard was systolic or diastolic dysfunction as assessed by two independent physicians with discordance resolved via consensus. We performed descriptive statistics (mean ± standard deviation) and calculated the sensitivity, specificity, positive likelihood ratio (LR+), and negative likelihood ratio (LR-) of the AI in determining systolic and diastolic dysfunction with 95 % confidence interval (CI). Subgroup analyses were performed by body mass index (BMI).
Results
We enrolled 220 patients, with 11 being excluded due to inadequate images, resulting in 209 patients being included in the study. Mean age was 60 ± 9 years, 51.7 % were women, and the mean BMI was 31 ± 8.1 mg/kg2. For assessing systolic dysfunction, AI was 85.7 % (95 %CI 57.2 % to 98.2 %) sensitive and 94.8 % (95 %CI 90.6 % to 97.5 %) specific with a LR+ of 16.4 (95 %CI 8.6 to 31.1) and LR- of 0.15 (95 % CI 0.04 to 0.54). For assessing diastolic dysfunction, AI was 91.9 % (95 %CI 85.6 % to 96.0 %) sensitive and 94.2 % (95 %CI 87.0 % to 98.1 %) specific with a LR+ of 15.8 (95 %CI 6.7 to 37.1) and a LR- of 0.09 (0.05 to 0.16). When analyzed by BMI, results were similar except for lower sensitivity in the BMI ≥ 30 vs BMI < 30 (100 % vs 80 %).
Conclusion
When compared with expert assessment, AI had high sensitivity and specificity for diagnosing both systolic and diastolic dysfunction.
{"title":"Diagnostic accuracy of artificial intelligence for identifying systolic and diastolic cardiac dysfunction in the emergency department","authors":"Michael Gottlieb MD, Evelyn Schraft MD, James O'Brien MD, Daven Patel MD, MPH","doi":"10.1016/j.ajem.2024.10.019","DOIUrl":"10.1016/j.ajem.2024.10.019","url":null,"abstract":"<div><h3>Introduction</h3><div>Cardiac point-of-care ultrasound (POCUS) can evaluate for systolic and diastolic dysfunction to inform care in the Emergency Department (ED). However, accurate assessment can be limited by user experience. Artificial intelligence (AI) has been proposed as a model to increase the accuracy of cardiac POCUS. However, there is limited evidence of the accuracy of AI in the clinical environment. The objective of this study was to determine the diagnostic accuracy of AI for identifying systolic and diastolic dysfunction compared with expert reviewers.</div></div><div><h3>Methods</h3><div>This was a prospective, observational study of adult ED patients aged ≥45 years with risk factors for systolic and diastolic dysfunction. Ultrasound fellowship-trained physicians used an ultrasound machine with existing AI software and obtained parasternal long axis, parasternal short axis, and apical 4-chamber views of the heart. Systolic dysfunction was defined as ejection fraction (EF) < 50 % in at least two views using visual assessment or E-point septal separation >10 mm. Diastolic dysfunction was defined as an E:A < 0.8, or ≥ 2 of the following: septal e' < 7 cm/s or lateral e' < 10 cm/s, E:e' > 14, or left atrial volume > 34 mL/m<sup>2</sup>. AI was subsequently used to measure EF, E, A, septal e', and lateral e' velocities. The gold standard was systolic or diastolic dysfunction as assessed by two independent physicians with discordance resolved via consensus. We performed descriptive statistics (mean ± standard deviation) and calculated the sensitivity, specificity, positive likelihood ratio (LR+), and negative likelihood ratio (LR-) of the AI in determining systolic and diastolic dysfunction with 95 % confidence interval (CI). Subgroup analyses were performed by body mass index (BMI).</div></div><div><h3>Results</h3><div>We enrolled 220 patients, with 11 being excluded due to inadequate images, resulting in 209 patients being included in the study. Mean age was 60 ± 9 years, 51.7 % were women, and the mean BMI was 31 ± 8.1 mg/kg<sup>2</sup>. For assessing systolic dysfunction, AI was 85.7 % (95 %CI 57.2 % to 98.2 %) sensitive and 94.8 % (95 %CI 90.6 % to 97.5 %) specific with a LR+ of 16.4 (95 %CI 8.6 to 31.1) and LR- of 0.15 (95 % CI 0.04 to 0.54). For assessing diastolic dysfunction, AI was 91.9 % (95 %CI 85.6 % to 96.0 %) sensitive and 94.2 % (95 %CI 87.0 % to 98.1 %) specific with a LR+ of 15.8 (95 %CI 6.7 to 37.1) and a LR- of 0.09 (0.05 to 0.16). When analyzed by BMI, results were similar except for lower sensitivity in the BMI ≥ 30 vs BMI < 30 (100 % vs 80 %).</div></div><div><h3>Conclusion</h3><div>When compared with expert assessment, AI had high sensitivity and specificity for diagnosing both systolic and diastolic dysfunction.</div></div>","PeriodicalId":55536,"journal":{"name":"American Journal of Emergency Medicine","volume":"86 ","pages":"Pages 115-119"},"PeriodicalIF":2.7,"publicationDate":"2024-10-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142481693","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
扫码关注我们
求助内容：
应助结果提醒方式：
京公网安备 11010802042870号