High Rates of Fabricated and Inaccurate References in ChatGPT-Generated Medical Content.

Background The availability of large language models such as Chat Generative Pre-trained Transformer (ChatGPT, OpenAI) has enabled individuals from diverse backgrounds to access medical information. However, concerns exist about the accuracy of ChatGPT responses and the references used to generate medical content. Methods This observational study investigated the authenticity and accuracy of references in medical articles generated by ChatGPT. ChatGPT-3.5 generated 30 short medical papers, each with at least three references, based on standardized prompts encompassing various topics and therapeutic areas. Reference authenticity and accuracy were verified by searching Medline, Google Scholar, and the Directory of Open Access Journals. The authenticity and accuracy of individual ChatGPT-generated reference elements were also determined. Results Overall, 115 references were generated by ChatGPT, with a mean of 3.8±1.1 per paper. Among these references, 47% were fabricated, 46% were authentic but inaccurate, and only 7% were authentic and accurate. The likelihood of fabricated references significantly differed based on prompt variations; yet the frequency of authentic and accurate references remained low in all cases. Among the seven components evaluated for each reference, an incorrect PMID number was most common, listed in 93% of papers. Incorrect volume (64%), page numbers (64%), and year of publication (60%) were the next most frequent errors. The mean number of inaccurate components was 4.3±2.8 out of seven per reference. Conclusions The findings of this study emphasize the need for caution when seeking medical information on ChatGPT since most of the references provided were found to be fabricated or inaccurate. Individuals are advised to verify medical information from reliable sources and avoid relying solely on artificial intelligence-generated content.

Recent Trend in Artificial Intelligence-Assisted Biomedical Publishing: A Quantitative Bibliometric Analysis.

The rapid advancements in artificial intelligence (AI) technology in recent years have led to its integration into biomedical publishing. However, the extent to which AI has contributed to developing biomedical literature is unclear. This study aimed to identify trends in AI-generated content within peer-reviewed biomedical literature. We first tested the sensitivity and specificity of commercially available AI-detection software (Originality.AI, Collingwood, Ontario, Canada). Next, we conducted a MEDLINE (Medical Literature Analysis and Retrieval System Online) search to identify randomized controlled trials with available abstracts indexed between January 2020 and March 2023. We randomly selected 30 abstracts per quarter during this period and pasted the abstracts into the AI detection software to determine the probability of AI-generated content. The software yielded 100% sensitivity, 95% specificity, and excellent overall discriminatory ability with an area under the receiving operating curve of 97.6%. Among the 390 MEDLINE-indexed abstracts included in the analysis, the prevalence with a high probability (≥ 90%) of AI-generated text increased during the study period from 21.7% to 36.7% (p=0.01) based on a chi-square test for trend. The increasing prevalence of AI-generated text during the study period was also observed in various sensitivity analyses using AI probability thresholds ranging from 50% to 99% (all p≤0.01). The results of this study suggest that the prevalence of AI-assisted publishing in peer-reviewed journals has been increasing in recent years, even before the widespread adoption of ChatGPT (OpenAI, San Francisco, California, United States) and similar tools. The extent to which natural writing characteristics of the authors, utilization of common AI-powered applications, and introduction of AI elements during the post-acceptance publication phase influence AI detection scores warrants further study.

Antimicrobial and resource utilization with T2 magnetic resonance for rapid diagnosis of bloodstream infections: systematic review with meta-analysis of controlled studies.

Objectives: To compare antimicrobial and resource utilization with T2 Magnetic Resonance (T2MR) versus blood culture (BC) in patients with suspected bloodstream infection.Methods: We systematically searched MEDLINE, EMBASE, and CENTRAL for randomized trials or observational controlled studies of patients with suspected bloodstream infection receiving a diagnosis with T2MR or BC. Using an inverse variance meta-analysis model, we reported mortality using the risk ratio (RR) and the remaining outcomes as the mean difference (MD).Results: Fourteen studies were included in the meta-analysis. Time to detection (MD = -81 hours; p < 0.001) and time to species identification (MD = -77 hours; p < 0.001) were faster with T2MR. Patients testing positive on T2MR received targeted antimicrobial therapy faster (-42 hours; p < 0.001) and patients testing negative on T2MR were de-escalated from empirical therapy faster (-7 hours; p = 0.02) vs. BC. Length of intensive care unit stay (MD = -5.0 days; p = 0.03) and hospital stay (MD = -4.8 days; p = 0.03) were shorter with T2MR. Mortality rates were comparable between T2MR and BC (28.9% vs. 29.9%, RR = 1.02, p = 0.86).Conclusion: Utilization of T2MR for identification of bloodstream pathogens provides faster time to detection, faster transition to targeted microbial therapy, faster de-escalation of empirical therapy, shorter ICU and hospital stay, and with comparable mortality rate versus BC.

Clinical Utility of Presacral Neurectomy as an Adjunct to Conservative Endometriosis Surgery: Systematic Review and Meta-Analysis of Controlled Studies.

The objective of this review was to compare the efficacy and safety of conservative surgery with or without adjunctive presacral neurectomy (PN) for chronic endometriosis-related pelvic pain. In a systematic review with meta-analysis, randomized or nonrandomized controlled studies of conservative endometriosis surgery with or without adjunctive PN were included. Main outcomes were treatment failure (the proportion of women in which surgery failed to adequately resolve midline pain) and the frequency of operative and postoperative complications. A total of 7 studies with 8 group comparisons (3 randomized) representing 503 women (250 PN; 253 Control) were included. Over 34 months median follow-up, crude rates of treatment failure were 15.0% with PN and 40.9% with Controls (risk ratio = 0.43, 95% CI = 0.30 to 0.60, p < 0.001). The risk of postoperative constipation was higher with PN vs. Controls (12.5% vs. 0%, p = 0.024). No treatment group differences were observed for the risk of operative complications (0.6% vs. 0%, p = 0.498), reoperation (4.1% vs. 3.0%, p = 0.758) or urinary incontinence (5.0% vs. 0%, p = 0.195). Overall, in well-selected patients, conservative surgery with adjunctive PN may provide greater relief from midline pain and a similarly low rate of operative complications relative to conservative surgery alone but may increase the risk of constipation postoperatively. However, results were derived from mainly older and lower quality studies. Since then, surgical techniques to treat endometriosis have been improved and the effect of PN observed in prior studies should be confirmed in future studies in women in whom radical excision of deep infiltrating lesions is obtained.

Asymptomatic severe aortic stenosis with left ventricular dysfunction: watchful waiting or valve replacement?

Aortic stenosis (AS) is the most common valvular heart disorder in older adults. Patients with severe AS are generally treated nonsurgically if asymptomatic and referred to aortic valve replacement when symptoms develop. However, patients with severe asymptomatic AS with left ventricular dysfunction may benefit from early aortic valve replacement. Although operative mortality in patients with severe AS and left ventricular dysfunction is greater than in patients with preserved left ventricular function, the overall mortality risk is substantially lower than that of watchful waiting. Operative risk in patients with severe AS and left ventricular dysfunction is often overestimated and, consequently, most are not referred to surgery despite clinical data in support of early aortic valve replacement. Asymptomatic patients with echocardiographic confirmation of severe AS and left ventricular dysfunction should be referred for aortic valve replacement.

Safety and effectiveness of microvascular decompression for treatment of hemifacial spasm: a systematic review.

INTRODUCTION: Hemifacial spasm (HFS) is caused by vascular compression of the VII cranial nerve at its root exit zone from the brainstem. Microvascular decompression (MVD) is the only treatment option that offers the prospect of a definitive cure for HFS. We conducted the first systematic review on the safety and effectiveness of MVD for the treatment of HFS. METHODS: English-language studies on MVD for HFS published from 2000 to present were retrieved and summarized. The primary outcomes of interest were treatment success, recurrence rate and adverse events including mortality, stroke, cerebrospinal fluid leak, facial palsy (permanent and transient) and hearing deficit (permanent and transient). RESULTS: Twenty-two papers representing 5685 patients treated with MVD for HFS were included in this review. Complete resolution of symptoms following MVD was reported in 91.1% of patients over a median 2.9-year follow-up period. HFS symptoms recurred in 2.4% of patients and 1.2% underwent repeat MVD during the follow-up period. Transient complications included facial palsy (9.5%), hearing deficit (3.2%) and cerebrospinal fluid leak (1.4%). Permanent complications included hearing deficit (2.3%), facial palsy (0.9%), stroke (< 0.1%) and death (< 0.1%). CONCLUSIONS: MVD successfully relieves HFS in approximately 9 of 10 patients with low rates of symptom recurrence. Complications of this surgery are uncommon and generally transient.