Reliability of Medical Information Provided by ChatGPT: Assessment Against Clinical Guidelines and Patient Information Quality Instrument.

BACKGROUND: ChatGPT-4 is the latest release of a novel artificial intelligence (AI) chatbot able to answer freely formulated and complex questions. In the near future, ChatGPT could become the new standard for health care professionals and patients to access medical information. However, little is known about the quality of medical information provided by the AI. OBJECTIVE: We aimed to assess the reliability of medical information provided by ChatGPT. METHODS: Medical information provided by ChatGPT-4 on the 5 hepato-pancreatico-biliary (HPB) conditions with the highest global disease burden was measured with the Ensuring Quality Information for Patients (EQIP) tool. The EQIP tool is used to measure the quality of internet-available information and consists of 36 items that are divided into 3 subsections. In addition, 5 guideline recommendations per analyzed condition were rephrased as questions and input to ChatGPT, and agreement between the guidelines and the AI answer was measured by 2 authors independently. All queries were repeated 3 times to measure the internal consistency of ChatGPT. RESULTS: Five conditions were identified (gallstone disease, pancreatitis, liver cirrhosis, pancreatic cancer, and hepatocellular carcinoma). The median EQIP score across all conditions was 16 (IQR 14.5-18) for the total of 36 items. Divided by subsection, median scores for content, identification, and structure data were 10 (IQR 9.5-12.5), 1 (IQR 1-1), and 4 (IQR 4-5), respectively. Agreement between guideline recommendations and answers provided by ChatGPT was 60% (15/25). Interrater agreement as measured by the Fleiss κ was 0.78 (P<.001), indicating substantial agreement. Internal consistency of the answers provided by ChatGPT was 100%. CONCLUSIONS: ChatGPT provides medical information of comparable quality to available static internet information. Although currently of limited quality, large language models could become the future standard for patients and health care professionals to gather medical information.

Prevention of conversion in posterior retroperitoneal adrenalectomy by measuring pre-operative anatomical conditions on cross-sectional imaging (computerised tomography or magnetic resonance imaging).

Background: In addition to the common laparoscopic lateral transperitoneal adrenalectomy (LTA), the posterior retroperitoneal adrenalectomy (PRA) is becoming increasingly important. Both techniques overlap in their indication, resulting in uncertainty about the preferred approach in some patients. We hypothesise that by determining anatomical characteristics on cross-sectional imaging computerised tomography or magnetic resonance imaging, we can show the limitations of the PRA and prevent patients from being converted to LTA. Methods: This retrospective study includes 14 patients who underwent PRA (n = 15) at a single institution between 2016 and 2018. Previously described parameters such as the retroperitoneal fat mass (RPF) were measured on pre-operative imaging. We compared data from one patient who had a conversion with those from 13 patients without conversion. Furthermore, we explored the influence of these parameters on the operative time. Results: Conversion to LTA was necessary during 1 PRA procedure. Fourteen PRAs in 13 patients were successfully completed. The mean body mass index was 30 kg/m2 and the mean operation time was 98 min. One patient who underwent a conversion had a substantially higher RPF (25 mm) compared to the patients with successfully completed PRA (median: 5.5 mm [P = 0.001]). Furthermore, the operation time strongly correlated with the RPF (P = 0.004, r = 0.713). Conclusions: Surgeons can use pre-operative imaging to assess the anatomical features to determine whether a PRA can be performed. Patients with an RPF under 14.3 mm can be safely treated with PRA. In contrast, LTA access should be considered for patients with a higher RPF (>25 mm).

Near-infrared autofluorescence of the parathyroid glands during thyroidectomy for the prevention of hypoparathyroidism: a prospective randomized clinical trial.

PURPOSE: Postoperative hypoparathyroidism remains the most often complication in thyroid surgery. Near-infrared autofluorescence (NIR-AF) is a modality to identify parathyroid glands (PG) in vivo with high accuracy, but its use in daily routine surgery is unclear so far. In this randomized controlled trial, we evaluate the ability of NIR-AF to prevent postoperative hypoparathyroidism following total thyroidectomy. METHODS: Patients undergoing total thyroidectomy were allocated in two groups with the use of NIR-AF in the intervention group or according to standard practice in the control group. The aim was to identify the PGs in an early most stage of the operation to prevent their devascularization or removal. Parathyroid hormone was measured pre- and postoperatively and on postoperative day (POD) 1. Serum calcium was measured on POD 1 and 2. Possible symptoms and calcium/calcitriol supplement were recorded. RESULTS: A total of 60 patients were randomized, of whom 30 underwent NIR-AF-based PG identification. Hypoparathyroidism at skin closure occurred in 7 out of 30 patients using NIR-AF, respectively, in 14 out of 30 patients in the control group (p=0.058). There was no significant difference in serum calcium and parathyroid hormone levels between both groups. Likewise, NIR-AF could not detect PGs at a higher rate. CONCLUSION: The use of NIR-AF may help surgeons identify and preserve PGs but did not significantly reduce the incidence of postoperative hypoparathyroidism in this trial. Larger case series have to clarify whether there is a benefit in routine thyroidectomy. TRIAL REGISTRATION NUMBER: DRKS00009242 (German Clinical Trial Register). Registration date: 03.09.2015.

Assessing the severity of acute pancreatitis (ASAP) in Switzerland: A nationwide survey on severity assessment in daily clinical practice.

BACKGROUND/OBJECTIVES: We aim to assess which tools for severity stratification in acute pancreatitis are used in today's daily clinical practice and to what extent the new Atlanta classification is being implemented by the medical community in Switzerland. METHODS: The heads of surgical, medical and emergency departments of Swiss hospitals (n = 83) that directly treat patients with acute pancreatitis were given access to an online survey and asked to forward the questionnaire to their team. The questionnaire consisted of 16 items, including questions about the specialty background of the participants, the allocation of patients with AP, severity assessment, patient management, the role of imaging procedures, and future perspectives. RESULTS: A total of 233 participants from 63 hospitals responded (response rate, 74%). A vast majority of participants [198 (87%)] does assess severity. The most frequently used tools are the Ranson [108 (87%)] and APACHE II scores [28 (23%)]. A majority of the participants were not satisfied with the currently available tools to assess severity [130 (59%)]. A minority [15 (12%)] use the revised Atlanta classification to assess the degree of severity in AP. CONCLUSIONS: The Ranson score remains the dominant risk stratification tool in clinical practice in Switzerland, followed by the APACHE II score. Other modern instruments, such as the Atlanta 2012 classification, have not yet earned broad recognition and have not reached daily practice. Further efforts must be made to expand physicians' awareness of their existence and significance.

Laboratory markers predicting severity of acute pancreatitis.

Acute pancreatitis (AP) is an inflammatory disease of highly variable severity, ranging from mild cases with low mortality to severe cases with high mortality. Numerous biomarkers have been studied as potential early predictors of the severity of this disease so that treatment can be optimally tailored to prevent complications. We aim to present and discuss the most relevant biomarkers for early severity assessment in AP that have been studied to date. We review the current literature on biomarkers that have been used to predict the severity in AP. C-reactive protein (CRP) is still considered to be the gold standard, with a cut-off value of 150 mg/ml 48 h after disease onset. Other markers, including procalcitonin (PCT) and interleukin 6 (IL-6) have been implemented in some hospitals, but are not used on a routine basis. Most other markers, including acute phase proteins (LBP, SAA, PTX3), cytokines (Il-8, TNF-a, MIF), activation peptides of pancreatic proteases (TAP, CAPAP, PLAP), antiproteases (AAT, a2M), adhesion molecules (ICAM-1, selectins, E-cadherin) and leukocyte-derived enzymes (PA2, PMN-E) have shown some promising results but have not been routinely implemented. Furthermore, new and interesting biomarkers (Copeptin, TRX-1, Ang-2, E-2) have shown good results, but more research is needed to determine if they could play a role in the future. Various reasons why new markers for disease severity have not been adopted in daily routine include low accuracy, cumbersome laboratory techniques and high cost. Despite these difficulties, research is still very active in finding new markers to predict the severity of AP.

Preoperative identification of small parathyroid adenomas-better done by fluorocholine positron emission tomography/computed tomography.

Background: Preoperative localization imaging studies are crucial for safe and successful parathyroidectomy in patients with primary hyperparathyroidism (pHPT), especially in focused approaches. A common imaging sequence is ultrasound followed by scintigraphy. These techniques, but not 18F-fluorocholine positron emission tomography/computed tomography (PET/CT), show lower detection rates in multiglandular disease (MGD), which is associated with smaller adenomas. In this study, we evaluate the accuracy of these modalities in small parathyroid adenomas (PAs) and discuss the potential sequence of preoperative localization diagnostics. Methods: Patients undergoing parathyroidectomy for pHPT were retrospectively categorized into small adenoma (specimen diameter <10 mm) and large adenoma. The groups were compared for accuracy of preoperative imaging studies, short-term and long-term outcomes. Results: Among 147 patients retrospectively analyzed in this study, 38 small PAs were found. Preoperative correct quadrant prediction for small adenomas was significantly lower for ultrasound (P=0.03) and single-photon emission computed tomography/CT (SPECT/CT) (P<0.01) but not for choline PET/CT. While PET/CT was performed significantly more often in small PAs (P<0.01), it showed highly significant superiority over the other imaging modalities in accurate preoperative localization in both small (P<0.0001) and large PAs (P<0.01). There was no difference in calcium and parathyroid hormone (PTH) levels at latest follow-up with slightly more recurrences in small adenomas (P=0.08). Conclusions: Choline PET/CT showed a better diagnostic yield especially for small and multiple adenomas and was better in prediction of the correct localization. It could therefore serve as a second-line imaging modality.

Safety of parathyroidectomy in older vs. younger patients with primary hyperparathyroidism.

BACKGROUND: Primary hyperparathyroidism is a prevalent endocrinopathy for which surgery is the only curative option. Parathyroidectomy is primarily recommended in younger and symptomatic patients, while there are still concerns regarding surgical complications in older patients. We therefore assessed the association of age with surgical outcomes in patients undergoing parathyroidectomy in a large population in Switzerland. METHODS: Population-based cohort study of adult patients with primary hyperparathyroidism undergoing parathyroidectomy in Switzerland between 2012 and 2018. The cohort was divided into four age groups (<50 years, 50-64 years, 65-74 years, ≥75 years). The primary outcome was a composite of in-hospital postoperative complications. Secondary outcomes were intensive care unit (ICU) admission, unplanned 30-day-readmission, and prolonged length of hospital stay. RESULTS: We studied 2642 patients with a median (IQR) age of 62 (53-71) years. Overall, 111 patients had complications including surgical re-intervention, hypocalcemia, and vocal cord paresis. As compared to <50 year-old patients, older patients had no increased risk for in-hospital complications after surgery (50-64 years: odds ratio (OR): 0.51 (95% CI, 0.28 to 0.92); 65-74 years: OR: 0.72 (95% CI, 0.39 to 1.33); ≥75 years: OR: 1.03 (95% CI, 0.54 to 1.95), respectively. There was also no association of age and rates of ICU-admission and unplanned 30-day-readmission, but oldest patients had longer hospital stays (OR: 2.38 (95% CI, 1.57 to 3.60)). CONCLUSION: ≥50 year-old patients undergoing parathyroidectomy had comparable risk of in-hospital complications as compared with younger ones. These data support parathyroidectomy in even older patients with primary hyperparathyroidism as performed in clinical routine.

Cancer-associated retinopathy as the leading symptom in colon cancer.

Cancer-associated retinopathy (CAR) is a rare paraneoplastic visual syndrome. Its early detection may lead to the diagnosis of the causative malignancy. As many different types of malignancies are known to be associated with CAR, it is important that clinicians are aware of the phenomenon of CAR.

Residents' performance in open versus laparoscopic bench-model cholecystectomy in a hands-on surgical course.

INTRODUCTION: Laparoscopy has become the gold standard for many abdominal procedures. Among young surgeons, experience in laparoscopic surgery increasingly outweighs experience in open surgery. This study was conducted to compare residents' performance in laparoscopic versus open bench-model task. METHODS: In an international surgical skills course, we compared trainees' performance in open versus laparoscopic cholecystectomy in a cadaveric animal bench-model. Both exercises were evaluated by board-certified surgeons using an 8-item checklist and by the trainees themselves. RESULTS: 238 trainees with a median surgical experience of 24 months (interquartile range 14-48) took part. Twenty-two percent of the trainees had no previous laparoscopic and 62% no previous open cholecystectomy experience. Significant differences were found in the overall score (median difference of 1 (95% CI: 1, 1), p < 0.001), gallbladder perforation rate (73% vs. 29%, p < 0.001), safe dissection of the Calot's triangle (98% vs. 90%, p = 0.001) and duration of surgery (42 (13) minutes vs. 26 (10) minutes (mean differences 17.22 (95% CI: 15.37, 19.07), p < 0.001)), all favouring open surgery. The perforation rate in open and laparoscopic cholecystectomies was not consistently decreasing with increasing years of experience or number of previously performed procedures. Self-assessment was lower than the assessment by board-certified surgeons. CONCLUSION: Despite lower experience in open compared to laparoscopic cholecystectomy, better performance was observed in open task. It may be explained by a wider access with easier preparation. Open cholecystectomy is the rescue manoeuvre and therefore, it is important to provide also enough training opportunities in open surgery.

Incarcerated obturator hernia-laparoscopic repair with intraoperative view of the corona mortis.

Obturator hernias are a rare form of abdominal wall hernias. We present a case of a patient with an obturator hernia diagnosed by the classical signs of lower abdominal pain, a positive Howship-Romberg sign (painful internal rotation of the hip) and a computed tomography scan showing a herniated loop of small bowel. During the emergency laparoscopic hernia repair (transabdominal preperitoneal approach) a variant vessel, the corona mortis, was detected.