ABSTRACT
Background: We assessed 2 versions of the large language model (LLM) ChatGPT-versions 3.5 and 4.0-in generating appropriate, consistent, and readable recommendations on core critical care topics. Research Question: How do successive large language models compare in terms of generating appropriate, consistent, and readable recommendations on core critical care topics? Design and Methods: A set of 50 LLM-generated responses to clinical questions were evaluated by 2 independent intensivists based on a 5-point Likert scale for appropriateness, consistency, and readability. Results: ChatGPT 4.0 showed significantly higher median appropriateness scores compared to ChatGPT 3.5 (4.0 vs 3.0, P < .001). However, there was no significant difference in consistency between the 2 versions (40% vs 28%, P = 0.291). Readability, assessed by the Flesch-Kincaid Grade Level, was also not significantly different between the 2 models (14.3 vs 14.4, P = 0.93). Interpretation: Both models produced "hallucinations"-misinformation delivered with high confidence-which highlights the risk of relying on these tools without domain expertise. Despite potential for clinical application, both models lacked consistency producing different results when asked the same question multiple times. The study underscores the need for clinicians to understand the strengths and limitations of LLMs for safe and effective implementation in critical care settings. Registration: https://osf.io/8chj7/.
ABSTRACT
BACKGROUND: Resuscitative transesophageal echocardiography (TEE) is an emerging POCUS modality that can be used to guide trauma resuscitation. METHODS: Trauma patients who underwent TEE within 24 h of admission from 2013 to 2022 were prospectively identified. We retrospectively analyzed resuscitative TEE reports and patient charts in duplicate. RESULTS: 29 providers performed TEE for 54 acute trauma patients. 28 (52%) died in hospital; 33 (61%) required operative intervention (<24 h). Median injury severity score was 29 [IQR 22-43]. The most common indications for TEE were hemodynamic instability (34, 63%), inadequate windows for transthoracic echocardiography (14, 26%) and cardiac arrest (11, 20%). There were no identified complications. A new diagnosis was made in 31 (57%) cases: most commonly right ventricular dysfunction (10, 19%), pericardial effusion (9, 17%), and hypovolemia (6, 11%). TEE ruled out major cardiac injury in 83% of cases. TEE changed resuscitative strategy, in 17 (32%) patients, diagnostic imaging approach in 6 (11%) patients, procedural or operative approach in 5 (9%) patients and disposition from the trauma bay in 4 (7%) patients. CONCLUSION: Resuscitative TEE during acute trauma care has an additional diagnostic yield to existing diagnostic pathways and may impact definitive management for some patients in the trauma bay.