Using artificial intelligence to generate medical literature for urology patients: a comparison of three different large language models.

PURPOSE: Large language models (LLMs) are a form of artificial intelligence (AI) that uses deep learning techniques to understand, summarize and generate content. The potential benefits of LLMs in healthcare is predicted to be immense. The objective of this study was to examine the quality of patient information leaflets (PILs) produced by 3 LLMs on urological topics. METHODS: Prompts were created to generate PILs from 3 LLMs: ChatGPT-4, PaLM 2 (Google Bard) and Llama 2 (Meta) across four urology topics (circumcision, nephrectomy, overactive bladder syndrome, and transurethral resection of the prostate). PILs were evaluated using a quality assessment checklist. PIL readability was assessed by the Average Reading Level Consensus Calculator. RESULTS: PILs generated by PaLM 2 had the highest overall average quality score (3.58), followed by Llama 2 (3.34) and ChatGPT-4 (3.08). PaLM 2 generated PILs were of the highest quality in all topics except TURP and was the only LLM to include images. Medical inaccuracies were present in all generated content including instances of significant error. Readability analysis identified PaLM 2 generated PILs as the simplest (age 14-15 average reading level). Llama 2 PILs were the most difficult (age 16-17 average). CONCLUSION: While LLMs can generate PILs that may help reduce healthcare professional workload, generated content requires clinician input for accuracy and inclusion of health literacy aids, such as images. LLM-generated PILs were above the average reading level for adults, necessitating improvement in LLM algorithms and/or prompt design. How satisfied patients are to LLM-generated PILs remains to be evaluated.

Can we predict the risk of esophageal stricture after caustic injury?

Nonoperative management of severe caustic injuries has demonstrated its feasibility, avoiding the need for emergency esogastric resection and resulting in low mortality rates. However, leaving superficial necrosis in place could increase the risk of esophageal stricture development. Data on the risk factors of esophageal stricture secondary to caustic ingestion are scarce. The aim of our study was to identify the risk factors for esophageal strictures after caustic ingestion at admission. From February 2015 to March 2021, all consecutive patients with esophageal or gastric caustic injury score ≥ II according to the Zargar classification were retrospectively analyzed. For each patient, we collected over 50 criteria at admission to the emergency room and then selected among them 20 criteria with the best clinical relevance and limited missing data for risk factor analyses. Among the 184 patients included in this study, 37 developed esophageal strictures (cumulative rate 29.4%). All esophageal strictures occurred within 3 months. In multivariate analyses, the risk factors for esophageal strictures were voluntary ingestion (cause-specific hazard ratio 5.92; 95% confidence interval 1.76-19.95, P = 0.004), Zargar's esophageal score ≥ III (cause-specific hazard ratio 14.30; 95% confidence interval 6.07-33.67, P < 0.001), and severe ear, nose, and throat lesions (cause-specific hazard ratio 2.15; 95% confidence interval 1.09-4.22, P = 0.027). Intentional ingestion, severe endoscopic grade, and severe ENT lesions were identified as risk factors for esophageal stricture following caustic ingestion. Preventive measures for this population require further evaluation.

Impact on survival of sarcopenia, systemic inflammatory response and anthropometric factors after pancreatectomy for resectable pancreatic adenocarcinoma.

INTRODUCTION: Pancreatic adenocarcinoma (PDAC) is becoming a public health issue with a 5-years survival rate around 10%. Patients with PDAC are often sarcopenic, which impacts postoperative outcome. At the same time, overweight population is increasing and adipose tissue promotes tumor related-inflammation. With several studies supporting independently these data, we aimed to assess if they held an impact on survival when combined. METHODS: We included 232 patients from two university hospitals (CHU de Lille, Institut Paoli Calmette), from January 2011 to December 2018, who underwent Pancreaticoduodenectomy (PD) for resectable PDAC. Preoperative CT scan was used to measure sarcopenia and visceral fat according to international cut-offs. Neutrophil to lymphocyte (NLR) and platelet to lymphocyte ratios (PLR) were used to measure inflammation. For univariate and multivariate analyses, the Cox proportional-hazard model was used. P-values below 0.05 were considered significant. RESULTS: Sarcopenic patients with visceral obesity were less likely to survive than the others in multivariate analysis (OS, HR 1.65, p= 0.043). Cutaneous obesity did not influence survival. We also observed an influence on survival when we studied sarcopenia with visceral obesity (OS, p= 0.056; PFS, p = 0.014), sarcopenia with cutaneous obesity (PFS, p= 0.005) and sarcopenia with PLR (PFS, p= 0.043). This poor prognosis was also found in sarcopenic obese patients with high PLR (OS, p= 0.05; PFS, p= 0.01). CONCLUSION: Sarcopenic obesity was associated with poor prognosis after PD for PDAC, especially in patients with systemic inflammation. Pre operative management of these factors should be addressed in pancreatic cancer patients.

The Effect of Tranexamic Acid Administration During Liposuction on Bleeding Complications and Ecchymosis: A Systematic Review.

Liposuction is the most frequently performed cosmetic procedure. Tranexamic acid (TXA) has emerged as a promising blood loss reducing agent in plastic surgery, but its value in liposuction is still being studied. This systematic review investigates the safety and efficacy of TXA in reducing blood loss during liposuction procedures. A systematic review of PubMed (US National Library of Medicine, Bethesda, MD), MEDLINE (US National Library of Medicine), EMBASE (Elsevier, Amsterdam, the Netherlands), and Cochrane databases (Wiley, Hoboken, NJ) from inception to June 2023 was performed. The primary objective was to compare blood loss, hematoma rate, and ecchymosis from liposuction procedures in patients who received TXA versus those who did not. The secondary objective was to assess the incidence of TXA-related complications. A total of nine studies were included, published between 2018 and 2023 of which eight were prospective and one was retrospective. A total of 345 intervention versus 268 control arms were compared. Follow-up time ranged from 1-14 days. Mean age, and mean BMI ranged from 33 - 50 years, 23 to 30 kg/m2, respectively. Blood loss in aspirate was significantly less with TXA administration assessed by five studies (p < 0.05). Of the five studies that assessed the incidence of ecchymosis, all reported less bruising with TXA use. Among all the studies, only one found post-operative complications with five patients requiring transfusion in the control group (without TXA). The evidence provided in the literature suggests that TXA use in liposuction is safe and effective for reducing blood loss and ecchymosis with IV and local administration.