Background: and Purpose Whether perfusion-diffusion mismatch modifies treatment effect of mechanical thrombectomy (MT) for large strokes is poorly known. To address this question, we conducted a post hoc secondary analysis of the Large Stroke Therapy Evaluation (LASTE) randomized controlled trial (RCT). Methods: The LASTE RCT compared MT plus best medical treatment (BMT) to BMT alone in patients with large infarct (Alberta Stroke Program Early CT Score 0–5) in the 0–7-hour timewindow. This secondary analysis was restricted to patients with available baseline MR perfusionweighted imaging. We investigated the potential heterogeneity of MT treatment according to the presence of perfusion-diffusion mismatch, defined as a mismatch ratio ≥1.2, calculated as the time-to-maximum >6 seconds cerebral volume divided by ischemic core volume. The primary outcome was better functional outcome (favorable shift in the distribution of modified Rankin Scale [mRS] at 90 days, analyzed using generalized odds ratio [GenOR]). Results: A total of 102/324 patients were included, among whom 55 (54%) had a perfusiondiffusion mismatch. No significant treatment effect heterogeneity by diffusion-perfusion mismatch was observed for the primary outcome (GenOR for better functional outcome: 1.70 [95% confidence interval, CI, 0.95 to 3.05] and 1.04 [95% CI, 0.57 to 1.87] in patients with and without mismatch, respectively; ratio of GenORs: 1.63 [95% CI, 0.71 to 3.74]; P for heterogeneity=0.24) or for the secondary efficacy and safety outcomes. Conclusions Our study found no evidence of heterogeneity of treatment effect with respect to benefit or safety in patients with unrestricted infarct size at baseline treated with MT by presence of perfusion-diffusion mismatch. An individual participant-data meta-analysis of RCT is needed for definitive conclusions.

Prior research suggests metformin has anti-cancer effects, yet data are limited. We examined the association between diabetes treatment (metformin versus sulfonylurea) and risk of incident diabetes-related and non- diabetes-related cancers in US veterans.This retrospective cohort study included US veterans, without cancer, aged ≥ 55 years, who were new users of metformin or sulfo-nylureas for diabetes between 2001 to 2012. Cox proportional hazards models, with propensity score-matched inverse probability of treatment weighting (IPTW) were constructed. A total of 88,713 veterans (mean age 68.6 ± 7.8 years; 97.7% male; 84.1% White, 12.6% Black, 3.3% other race) were followed for 4.2 ± 3.0 years. Among metformin users (n = 60,476), there were 858 incident diabetes-related cancers (crude incidence rate [IR; per 1,000 person-years] = 3.4) and 3,533 non-diabetes-related cancers (IR = 14.1). Among sulfonylurea users (n = 28,237), there were 675 incident diabetes-related cancers (IR = 5.5) and 2,316 non-diabetes-related cancers (IR = 18.9). After IPTW adjustment, metformin use was associated with a lower risk of incident diabetes-related cancer (hazard ratio [HR] = 0.66, 95% CI 0.58-0.75) compared to sulfonylurea use. There was no association between treatment group (metformin versus sulfonylurea) and non-diabetes-related cancer (HR = 0.96, 95% CI 0.89-1.02). Of diabetes-related cancers, metformin users had lower incidence of liver (HR = 0.39, 95% CI 0.28-0.53), colorectal (HR = 0.75, 95% CI 0.62-0.92), and esophageal cancers (HR = 0.54, 95% CI 0.36-0.81). Among US veterans, metformin users had lower incidence of diabetes-related cancer, particularly liver, colorectal, and esophageal cancers, as compared to sulfonylurea users. Use of metformin was not associated with non-diabetes-related cancer. Further studies are needed to understand how metformin use impacts cancer incidence in different patient populations.

Objective: Large language models like chat generative pre-trained transformer (ChatGPT) have found success in various sectors, but their application in the medical field remains limited. This study aimed to assess the feasibility of using ChatGPT to provide accurate medical information to patients, specifically evaluating how well ChatGPT versions 3.5 and 4 aligned with the 2012 North American Spine Society (NASS) guidelines for lumbar disk herniation with radiculopathy. Methods: ChatGPT's responses to questions based on the NASS guidelines were analyzed for accuracy. Three new categories—overconclusiveness, supplementary information, and incompleteness—were introduced to deepen the analysis. Overconclusiveness referred to recommendations not mentioned in the NASS guidelines, supplementary information denoted additional relevant details, and incompleteness indicated omitted crucial information from the NASS guidelines. Results: Out of 29 clinical guidelines evaluated, ChatGPT-3.5 demonstrated accuracy in 15 responses (52%), while ChatGPT-4 achieved accuracy in 17 responses (59%). ChatGPT-3.5 was overconclusive in 14 responses (48%), while ChatGPT-4 exhibited overconclusiveness in 13 responses (45%). Additionally, ChatGPT-3.5 provided supplementary information in 24 responses (83%), and ChatGPT-4 provided supplemental information in 27 responses (93%). In terms of incompleteness, ChatGPT-3.5 displayed this in 11 responses (38%), while ChatGPT-4 showed incompleteness in 8 responses (23%). Conclusion ChatGPT shows promise for clinical decision-making, but both patients and healthcare providers should exercise caution to ensure safety and quality of care. While these results are encouraging, further research is necessary to validate the use of large language models in clinical settings.

Objective: Large language models like chat generative pre-trained transformer (ChatGPT) have found success in various sectors, but their application in the medical field remains limited. This study aimed to assess the feasibility of using ChatGPT to provide accurate medical information to patients, specifically evaluating how well ChatGPT versions 3.5 and 4 aligned with the 2012 North American Spine Society (NASS) guidelines for lumbar disk herniation with radiculopathy. Methods: ChatGPT's responses to questions based on the NASS guidelines were analyzed for accuracy. Three new categories—overconclusiveness, supplementary information, and incompleteness—were introduced to deepen the analysis. Overconclusiveness referred to recommendations not mentioned in the NASS guidelines, supplementary information denoted additional relevant details, and incompleteness indicated omitted crucial information from the NASS guidelines. Results: Out of 29 clinical guidelines evaluated, ChatGPT-3.5 demonstrated accuracy in 15 responses (52%), while ChatGPT-4 achieved accuracy in 17 responses (59%). ChatGPT-3.5 was overconclusive in 14 responses (48%), while ChatGPT-4 exhibited overconclusiveness in 13 responses (45%). Additionally, ChatGPT-3.5 provided supplementary information in 24 responses (83%), and ChatGPT-4 provided supplemental information in 27 responses (93%). In terms of incompleteness, ChatGPT-3.5 displayed this in 11 responses (38%), while ChatGPT-4 showed incompleteness in 8 responses (23%). Conclusion ChatGPT shows promise for clinical decision-making, but both patients and healthcare providers should exercise caution to ensure safety and quality of care. While these results are encouraging, further research is necessary to validate the use of large language models in clinical settings.

Prior research suggests metformin has anti-cancer effects, yet data are limited. We examined the association between diabetes treatment (metformin versus sulfonylurea) and risk of incident diabetes-related and non- diabetes-related cancers in US veterans.This retrospective cohort study included US veterans, without cancer, aged ≥ 55 years, who were new users of metformin or sulfo-nylureas for diabetes between 2001 to 2012. Cox proportional hazards models, with propensity score-matched inverse probability of treatment weighting (IPTW) were constructed. A total of 88,713 veterans (mean age 68.6 ± 7.8 years; 97.7% male; 84.1% White, 12.6% Black, 3.3% other race) were followed for 4.2 ± 3.0 years. Among metformin users (n = 60,476), there were 858 incident diabetes-related cancers (crude incidence rate [IR; per 1,000 person-years] = 3.4) and 3,533 non-diabetes-related cancers (IR = 14.1). Among sulfonylurea users (n = 28,237), there were 675 incident diabetes-related cancers (IR = 5.5) and 2,316 non-diabetes-related cancers (IR = 18.9). After IPTW adjustment, metformin use was associated with a lower risk of incident diabetes-related cancer (hazard ratio [HR] = 0.66, 95% CI 0.58-0.75) compared to sulfonylurea use. There was no association between treatment group (metformin versus sulfonylurea) and non-diabetes-related cancer (HR = 0.96, 95% CI 0.89-1.02). Of diabetes-related cancers, metformin users had lower incidence of liver (HR = 0.39, 95% CI 0.28-0.53), colorectal (HR = 0.75, 95% CI 0.62-0.92), and esophageal cancers (HR = 0.54, 95% CI 0.36-0.81). Among US veterans, metformin users had lower incidence of diabetes-related cancer, particularly liver, colorectal, and esophageal cancers, as compared to sulfonylurea users. Use of metformin was not associated with non-diabetes-related cancer. Further studies are needed to understand how metformin use impacts cancer incidence in different patient populations.

Objective: Large language models like chat generative pre-trained transformer (ChatGPT) have found success in various sectors, but their application in the medical field remains limited. This study aimed to assess the feasibility of using ChatGPT to provide accurate medical information to patients, specifically evaluating how well ChatGPT versions 3.5 and 4 aligned with the 2012 North American Spine Society (NASS) guidelines for lumbar disk herniation with radiculopathy. Methods: ChatGPT's responses to questions based on the NASS guidelines were analyzed for accuracy. Three new categories—overconclusiveness, supplementary information, and incompleteness—were introduced to deepen the analysis. Overconclusiveness referred to recommendations not mentioned in the NASS guidelines, supplementary information denoted additional relevant details, and incompleteness indicated omitted crucial information from the NASS guidelines. Results: Out of 29 clinical guidelines evaluated, ChatGPT-3.5 demonstrated accuracy in 15 responses (52%), while ChatGPT-4 achieved accuracy in 17 responses (59%). ChatGPT-3.5 was overconclusive in 14 responses (48%), while ChatGPT-4 exhibited overconclusiveness in 13 responses (45%). Additionally, ChatGPT-3.5 provided supplementary information in 24 responses (83%), and ChatGPT-4 provided supplemental information in 27 responses (93%). In terms of incompleteness, ChatGPT-3.5 displayed this in 11 responses (38%), while ChatGPT-4 showed incompleteness in 8 responses (23%). Conclusion ChatGPT shows promise for clinical decision-making, but both patients and healthcare providers should exercise caution to ensure safety and quality of care. While these results are encouraging, further research is necessary to validate the use of large language models in clinical settings.

Objective: Large language models like chat generative pre-trained transformer (ChatGPT) have found success in various sectors, but their application in the medical field remains limited. This study aimed to assess the feasibility of using ChatGPT to provide accurate medical information to patients, specifically evaluating how well ChatGPT versions 3.5 and 4 aligned with the 2012 North American Spine Society (NASS) guidelines for lumbar disk herniation with radiculopathy. Methods: ChatGPT's responses to questions based on the NASS guidelines were analyzed for accuracy. Three new categories—overconclusiveness, supplementary information, and incompleteness—were introduced to deepen the analysis. Overconclusiveness referred to recommendations not mentioned in the NASS guidelines, supplementary information denoted additional relevant details, and incompleteness indicated omitted crucial information from the NASS guidelines. Results: Out of 29 clinical guidelines evaluated, ChatGPT-3.5 demonstrated accuracy in 15 responses (52%), while ChatGPT-4 achieved accuracy in 17 responses (59%). ChatGPT-3.5 was overconclusive in 14 responses (48%), while ChatGPT-4 exhibited overconclusiveness in 13 responses (45%). Additionally, ChatGPT-3.5 provided supplementary information in 24 responses (83%), and ChatGPT-4 provided supplemental information in 27 responses (93%). In terms of incompleteness, ChatGPT-3.5 displayed this in 11 responses (38%), while ChatGPT-4 showed incompleteness in 8 responses (23%). Conclusion ChatGPT shows promise for clinical decision-making, but both patients and healthcare providers should exercise caution to ensure safety and quality of care. While these results are encouraging, further research is necessary to validate the use of large language models in clinical settings.

Prior research suggests metformin has anti-cancer effects, yet data are limited. We examined the association between diabetes treatment (metformin versus sulfonylurea) and risk of incident diabetes-related and non- diabetes-related cancers in US veterans.This retrospective cohort study included US veterans, without cancer, aged ≥ 55 years, who were new users of metformin or sulfo-nylureas for diabetes between 2001 to 2012. Cox proportional hazards models, with propensity score-matched inverse probability of treatment weighting (IPTW) were constructed. A total of 88,713 veterans (mean age 68.6 ± 7.8 years; 97.7% male; 84.1% White, 12.6% Black, 3.3% other race) were followed for 4.2 ± 3.0 years. Among metformin users (n = 60,476), there were 858 incident diabetes-related cancers (crude incidence rate [IR; per 1,000 person-years] = 3.4) and 3,533 non-diabetes-related cancers (IR = 14.1). Among sulfonylurea users (n = 28,237), there were 675 incident diabetes-related cancers (IR = 5.5) and 2,316 non-diabetes-related cancers (IR = 18.9). After IPTW adjustment, metformin use was associated with a lower risk of incident diabetes-related cancer (hazard ratio [HR] = 0.66, 95% CI 0.58-0.75) compared to sulfonylurea use. There was no association between treatment group (metformin versus sulfonylurea) and non-diabetes-related cancer (HR = 0.96, 95% CI 0.89-1.02). Of diabetes-related cancers, metformin users had lower incidence of liver (HR = 0.39, 95% CI 0.28-0.53), colorectal (HR = 0.75, 95% CI 0.62-0.92), and esophageal cancers (HR = 0.54, 95% CI 0.36-0.81). Among US veterans, metformin users had lower incidence of diabetes-related cancer, particularly liver, colorectal, and esophageal cancers, as compared to sulfonylurea users. Use of metformin was not associated with non-diabetes-related cancer. Further studies are needed to understand how metformin use impacts cancer incidence in different patient populations.

Objective: Large language models like chat generative pre-trained transformer (ChatGPT) have found success in various sectors, but their application in the medical field remains limited. This study aimed to assess the feasibility of using ChatGPT to provide accurate medical information to patients, specifically evaluating how well ChatGPT versions 3.5 and 4 aligned with the 2012 North American Spine Society (NASS) guidelines for lumbar disk herniation with radiculopathy. Methods: ChatGPT's responses to questions based on the NASS guidelines were analyzed for accuracy. Three new categories—overconclusiveness, supplementary information, and incompleteness—were introduced to deepen the analysis. Overconclusiveness referred to recommendations not mentioned in the NASS guidelines, supplementary information denoted additional relevant details, and incompleteness indicated omitted crucial information from the NASS guidelines. Results: Out of 29 clinical guidelines evaluated, ChatGPT-3.5 demonstrated accuracy in 15 responses (52%), while ChatGPT-4 achieved accuracy in 17 responses (59%). ChatGPT-3.5 was overconclusive in 14 responses (48%), while ChatGPT-4 exhibited overconclusiveness in 13 responses (45%). Additionally, ChatGPT-3.5 provided supplementary information in 24 responses (83%), and ChatGPT-4 provided supplemental information in 27 responses (93%). In terms of incompleteness, ChatGPT-3.5 displayed this in 11 responses (38%), while ChatGPT-4 showed incompleteness in 8 responses (23%). Conclusion ChatGPT shows promise for clinical decision-making, but both patients and healthcare providers should exercise caution to ensure safety and quality of care. While these results are encouraging, further research is necessary to validate the use of large language models in clinical settings.

Purpose: Varicocele is a common problem among infertile men. Varicocele repair (VR) is frequently performed to improve semen parameters and the chances of pregnancy. However, there is a lack of consensus about the diagnosis, indications for VR and its outcomes. The aim of this study was to explore global practice patterns on the management of varicocele in the context of male infertility. Materials and Methods: Sixty practicing urologists/andrologists from 23 countries contributed 382 multiple-choice-questions pertaining to varicocele management. These were condensed into an online questionnaire that was forwarded to clinicians involved in male infertility management through direct invitation. The results were analyzed for disagreement and agreement in practice patterns and, compared with the latest guidelines of international professional societies (American Urological Association [AUA], American Society for Reproductive Medicine [ASRM], and European Association of Urology [EAU]), and with evidence emerging from recent systematic reviews and meta-analyses. Additionally, an expert opinion on each topic was provided based on the consensus of 16 experts in the field. Results: The questionnaire was answered by 574 clinicians from 59 countries. The majority of respondents were urologists/ uro-andrologists. A wide diversity of opinion was seen in every aspect of varicocele diagnosis, indications for repair, choice of technique, management of sub-clinical varicocele and the role of VR in azoospermia. A significant proportion of the responses were at odds with the recommendations of AUA, ASRM, and EAU. A large number of clinical situations were identified where no guidelines are available. Conclusions This study is the largest global survey performed to date on the clinical management of varicocele for male infertility. It demonstrates: 1) a wide disagreement in the approach to varicocele management, 2) large gaps in the clinical practice guidelines from professional societies, and 3) the need for further studies on several aspects of varicocele management in infertile men.