Nimma, Induja R.

Patients with ineffective esophageal motility continue to experience significant symptoms years after diagnosis. In this survey of 47 patients diagnosed ≥5 years earlier, 45% reported GERD symptoms and 75% reported dysphagia. Greater numbers of failed swallows correlated with worse GERD and dysphagia scores, highlighting the long-term clinical impact of IEM and its association with esophageal symptom burden.

Abstract Objectives To comprehensively evaluate the validity of International Classification of Diseases, Tenth Revision, Clinical Modification (ICD-10-CM) codes for both prevalent diagnoses and less common diseases, and to assess the performance of a large language model (LLM)-based system in validating these codes. Materials and Methods This retrospective study analyzed hospital admissions from the Medical Information Mart for Intensive Care IV (MIMIC-IV) database. We developed a validated LLM-based system using GPT-4o, refined through iterative prompt engineering, to assess ICD-10-CM code validity. We measured the positive predictive value (PPV) of ICD-10-CM codes, PPV of principal and secondary diagnoses, and the performance of an LLM-based system in code validation. Results Among 865 079 assigned codes, the PPV was 84.6% (95% CI, 84.5%-84.6%). Principal diagnoses had a PPV of 93.9% (95% CI, 93.7%-94.1%), while secondary diagnoses had a PPV of 83.8% (95% CI, 83.7%-83.9%). The LLM system demonstrated high performance in validating ICD codes, achieving 93.6% accuracy, 95.4% sensitivity, and 85.2% specificity. Among correctly assigned secondary diagnoses, the majority (67.9%) represented historical or baseline conditions, while 32.1% reflected active conditions that deviated from baseline status; 22.3% of these emerged after hospital admission. PPV decreases with later diagnosis positions, with the largest decline occurring between principal and secondary diagnoses. Discussion and Conclusion In this large-scale evaluation, ICD-10-CM codes exhibited generally high accuracy, though variability existed by position and condition type. A validated LLM system performed comparably to physician review and offers a scalable means to improve coding accuracy. These findings support the potential for integrating LLM-based auditing into routine workflows to strengthen the quality of administrative and research data.

ABSTRACT Objective To comprehensively evaluate the validity of ICD-10-CM codes for both prevalent diagnoses and less common diseases, and to assess the performance of a large language model (LLM)-based system in validating these codes. Materials and Methods This retrospective study analyzed hospital admissions from the Medical Information Mart for Intensive Care (MIMIC-IV) database. We developed a validated LLM-based system using GPT-4o, refined through iterative prompt engineering, to assess ICD-10-CM code validity. We measured the PPV of ICD-10-CM codes, PPV of principal and secondary diagnoses, and the performance of an LLM-based system in code validation. Results Among 865,079 assigned codes, the PPV was 84.6% (95% CI, 84.5%-84.6%). Principal diagnoses had a PPV of 93.9% (95% CI, 93.7%-94.1%), while secondary diagnoses had a PPV of 83.8% (95% CI, 83.7%-83.9%). The LLM system demonstrated high performance in validating ICD codes, achieving 93.6% accuracy, 95.4% sensitivity and 85.2% specificity. Among correctly assigned secondary diagnoses, the majority (67.9%) represented historical or baseline conditions, while 32.1% reflected active conditions that deviated from baseline status; 22.3% of these emerged after hospital admission. PPV decreases with later diagnosis positions, with the largest decline occurring between principal and secondary diagnoses. Discussion and Conclusion In this large-scale evaluation, ICD-10-CM codes exhibited generally high accuracy, though variability existed by position and condition type. A validated LLM system performed comparably to physician review and offers a scalable means to improve coding accuracy. These findings support the potential for integrating LLM-based auditing into routine workflows to strengthen the quality of administrative and research data.

Introduction/Background: Atrial fibrillation (AF), a common postoperative arrhythmia, is associated with increased morbidity and mortality across various surgical populations. In patients undergoing orthotopic liver transplantation (LT), the presence of AF has been linked to poorer clinical outcomes. However, data on the prognostic significance of new-onset AF during early post-transplant readmissions in orthotopic LT recipients remain limited. Hypothesis: We postulated that incident AF occurring during 90-day post-LT readmissions is independently associated with increased in-hospital mortality, prolonged length of stay, and higher healthcare costs. Methods: We conducted a retrospective cohort study using data from the National Readmission Database (2016–2019). Adult LT recipients (aged >18 years) readmitted within 90 days were identified via ICD-10 coding. Patients who developed new-onset atrial fibrillation (AF) during readmission were compared to those without AF. Primary outcomes included in-hospital mortality, length of stay (LOS), and total hospitalization costs. Multivariable logistic and linear regression models, adjusted for demographic and clinical confounders, were used. Results: Of 31,557 adult LT recipients (mean age 55.5 years, 35% female), 8,449 (27.6%) (mean age 55.3 years, 36.7% women) were readmitted within 90 days. Among those readmitted, 222 patients (2.6%) developed new-onset atrial fibrillation (AF), and 84 (1.0%) died during the readmission period. New-onset AF was associated with a 7.6-fold increased risk of in-hospital mortality (95% CI: 3.0–19.0; P < 0.05). Additionally, AF onset was linked to a mean increase in LOS of 6.2 days (95% CI: 6.2–7.1; P = 0.001) and an average increase in hospitalization costs of $75,396 (95% CI: $11,932–$138,860; P = 0.02). All associations achieved statistical significance. Conclusion(s): New-onset AF during early post-transplant readmissions is associated with significantly worse clinical outcomes and increased healthcare utilization. These findings underscore the need for proactive rhythm monitoring and risk-stratified arrhythmia management during the liver transplant recovery period.

ABSTRACT Severe acute pancreatitis can cause systemic inflammation and multiorgan failure. We present the case of a 60-year-old woman who presented with necrotizing pancreatitis and subsequently developed a sigmoid colon perforation. The perforation presumably occurred because of extravasation of pancreatic enzymes into the abdominal cavity, resulting in colonic wall injury. Our case highlights the rare colonic complications of severe acute pancreatitis.

AbstractBackground and AimEffective clinical event classification is essential for clinical research and quality improvement. The validation of artificial intelligence (AI) models like Generative Pre‐trained Transformer 4 (GPT‐4) for this task and comparison with conventional methods remains unexplored.MethodsWe evaluated the performance of the GPT‐4 model for classifying gastrointestinal (GI) bleeding episodes from 200 medical discharge summaries and compared the results with human review and an International Classification of Diseases (ICD) code‐based system. The analysis included accuracy, sensitivity, and specificity evaluation, using ground truth determined by physician reviewers.ResultsGPT‐4 exhibited an accuracy of 94.4% in identifying GI bleeding occurrences, outperforming ICD codes (accuracy 63.5%, P < 0.001). GPT‐4's accuracy was either slightly lower or statistically similar to individual human reviewers (Reviewer 1: 98.5%, P < 0.001; Reviewer 2: 90.8%, P = 0.170). For location classification, GPT‐4 achieved accuracies of 81.7% and 83.5% for confirmed and probable GI bleeding locations, respectively, with figures that were either slightly lower or comparable with those of human reviewers. GPT‐4 was highly efficient, analyzing the dataset in 12.7 min at a cost of 21.2 USD, whereas human reviewers required 8–9 h each.ConclusionOur study indicates GPT‐4 offers a reliable, cost‐efficient, and faster alternative to current clinical event classification methods, outperforming the conventional ICD coding system and performing comparably to individual expert human reviewers. Its implementation could facilitate more accurate and granular clinical research and quality audits. Future research should explore scalability, prompt and model tuning, and ethical implications of high‐performance AI models in clinical data processing.

A woman in her 40s presented to the emergency department (ED) with a 3-week history of nausea, vomiting and diarrhoea. Blood cultures were positive for Serratia and Pantoea agglomerans spp. One month before her ED visit, she underwent targeted uterine artery embolisation with particles by an interventional radiologist. Uterine artery embolisation is considered a safe alternative to surgical removal of fibroids or hysterectomy. The patient was initially treated with targeted antibiotics for a large infected uterine fibroid but ultimately required a hysterectomy for source control. To our knowledge, this is the first documented case of P. agglomerans infecting a uterine fibroid.