Aims: Large language models (LLMs) have shown potential in clinical decision support, but the influence of prompt design on their performance, particularly in complex cardiology decision-making, is not well understood. Methods and results: We retrospectively reviewed 231 patients evaluated by our Heart Team for severe aortic stenosis, with treatment options including surgical aortic valve replacement, transcatheter aortic valve implantation, or medical therapy. We tested multiple prompt-design strategies using zero-shot (0-shot), Chain-of-Thought (CoT), and Tree-of-Thought (ToT) prompting, combined with few-shot prompting, free/guided-thinking, and self-consistency. Patient data were condensed into standardized vignettes and queried using GPT4-o (version 2024–05–13, OpenAI) 40 times per patient under each prompt (147 840 total queries). Primary endpoint was mean accuracy; secondary endpoints included sensitivity, specificity, area under the curve (AUC), and treatment invasiveness. Guided-thinking-ToT achieved the highest accuracy (94.04%, 95% CI 90.87–97.21), significantly outperforming few-shot-ToT (87.16%, 95% CI 82.68–91.63) and few-shot-CoT (85.32%, 95% CI 80.59–90.06; P < 0.0001). Zero-shot prompting showed the lowest accuracy (73.39%, 95% CI 67.48–79.31). Guided-thinking-ToT yielded the highest AUC values (up to 0.97) and was the only prompt whose invasiveness did not differ significantly from Heart Team decisions (P = 0.078). An inverted quadratic relationship emerged between few-shot examples and accuracy, with nine examples optimal (P < 0.0001). Self-consistency improved overall accuracy, particularly for ToT-derived prompts (P < 0.001). Conclusion: Prompt design significantly impacts LLM performance in clinical decision-making for severe aortic stenosis. Tree-of-Thought prompting markedly improved accuracy and aligned recommendations with expert decisions, though LLMs tended toward conservative treatment approaches.

Hepatic hydrothorax (HH) is a complication in patients with cirrhosis and portal hypertension. It predominantly presents in the right pleural cavity and usually associates with ascites. Few cases of HH occurring without detectable ascites have been reported. This case report comprehensively presents a case of a refractory left unilateral HH without ascites. The patient benefited from palliative care and the HH was managed using a semipermanent indwelling pleural catheter until she died 3 months after diagnosis.

Background: Accurate operative risk stratification is essential for treatment selection in severe aortic stenosis. We developed an automated workflow using large language models (LLMs) to replicate Heart Team risk assessment. Methods: We retrospectively analyzed 231 consecutive patients with severe aortic stenosis evaluated by multidisciplinary Heart Teams (January 2022–December 2024). An automated system using GPT-4o was developed, comprising the following: (1) structured data extraction from clinical dossiers; (2) EuroSCORE II calculation via two methods (algorithmic vs. LLM-based); (3) clinical vignette generation; and (4) risk stratification comparing EuroSCORE-based thresholds versus guideline-integrated LLM approaches with/without EuroSCORE values. The primary endpoint was the risk stratification accuracy of each method compared to Heart Team decisions. Results: Mean age was 79.5 ± 7.7 years, with 58.4% female. The automated workflow processed patients in 32.6 ± 6.4 s. The LLM-calculated EuroSCORE II showed a lower mean difference from Heart Team values (−1.42%, 95% CI −2.32 to −0.53) versus algorithmic calculation (−1.88%, 95% CI −2.38 to −1.38). For risk stratification, the guideline-integrated LLM without EuroSCORE achieved the highest accuracy (90.0%) and AUC (0.93), outperforming both the EuroSCORE-based (accuracy 50.2% for high-risk, AUC 0.63) and guideline-integrated LLM with EuroSCORE approaches (accuracy 82.4%, AUC 0.76). However, systematic hallucinations occurred for cardiovascular risk factors when data were missing. Conclusions: LLMs accurately calculated EuroSCORE II and achieved 90% concordance with multidisciplinary Heart Team decisions. However, hallucinations, reproducibility concerns, and the absence of clinical outcome validation preclude direct clinical application.

Aims: Patients with mental health conditions (MHCs) experience delayed diagnosis and management. We aimed to assess whether these patients had longer treatment delays and poorer outcomes despite a dedicated ST-elevation myocardial infarction (STEMI) fast-track protocol. Methods and results: We analysed data from EVALFAST prospective registry of confirmed STEMI patients admitted directly to the catheterization laboratory at Fribourg Hospital since June 2008. Patients were classified into MHC cohort if their electronic health record documented any MHC at STEMI diagnosis. The primary endpoint was time from the first medical contact (FMC) to diagnosis. Secondary endpoints included FMC-to-balloon time, infarct size (peak CK-MB), and 30-day and 5-year major adverse cardiac and cerebrovascular events (MACCE). We used generalized linear models adjusted for demographics, risk factors, and presentation mode, and Cox regression for survival analyses. Among 1208 patients (excluding cardiac arrests), 147 (12.1%) had documented MHC. The MHC cohort had more women (36.7 vs. 23.3%; P = 0.002) and hypertension (57.8 vs. 47.8%; P = 0.03). After adjustment, FMC-to-diagnosis time was significantly longer in MHC patients [+16.43 min; 95% confidence interval (CI) +4.19 to +28.68; P = 0.009], resulting in prolonged FMC-to-balloon time (+18.63 min; 95% CI +4.86 to +32.39; P = 0.008). Diagnosis-to-balloon time did not differ (P = 0.420). Delays were pronounced among patients self-presenting to emergency departments (P = 0.002). Mental health condition patients had larger infarct sizes (peak CK-MB +71.3 U/L; 95% CI +18.0 to +124.6; P = 0.009) and higher risk of 30-day MACCE [adjusted hazard ratio (HR) 1.82; 95% CI 1.05–3.17; P = 0.034] and 5-year cardiovascular death (adjusted HR 2.04; 95% CI 1.18–3.55; P = 0.011). Conclusion: Among patients presenting with STEMI, being labelled with an MHC was linked with increased delays in early management, leading to greater infarct size and higher risk of adverse cardiovascular events. Mental health conditions were associated with increased long-term years cardiovascular death.