A systematic review of large language models and their implications in medical education.

INTRODUCTION: In the past year, the use of large language models (LLMs) has generated significant interest and excitement because of their potential to revolutionise various fields, including medical education for aspiring physicians. Although medical students undergo a demanding educational process to become competent health care professionals, the emergence of LLMs presents a promising solution to challenges like information overload, time constraints and pressure on clinical educators. However, integrating LLMs into medical education raises critical concerns and challenges for educators, professionals and students. This systematic review aims to explore LLM applications in medical education, specifically their impact on medical students' learning experiences. METHODS: A systematic search was performed in PubMed, Web of Science and Embase for articles discussing the applications of LLMs in medical education using selected keywords related to LLMs and medical education, from the time of ChatGPT's debut until February 2024. Only articles available in full text or English were reviewed. The credibility of each study was critically appraised by two independent reviewers. RESULTS: The systematic review identified 166 studies, of which 40 were found by review to be relevant to the study. Among the 40 relevant studies, key themes included LLM capabilities, benefits such as personalised learning and challenges regarding content accuracy. Importantly, 42.5% of these studies specifically evaluated LLMs in a novel way, including ChatGPT, in contexts such as medical exams and clinical/biomedical information, highlighting their potential in replicating human-level performance in medical knowledge. The remaining studies broadly discussed the prospective role of LLMs in medical education, reflecting a keen interest in their future potential despite current constraints. CONCLUSIONS: The responsible implementation of LLMs in medical education offers a promising opportunity to enhance learning experiences. However, ensuring information accuracy, emphasising skill-building and maintaining ethical safeguards are crucial. Continuous critical evaluation and interdisciplinary collaboration are essential for the appropriate integration of LLMs in medical education.

Automating the Classification of Complexity of Medical Decision-Making in Patient-Provider Messaging in a Patient Portal.

BACKGROUND: Patient portals are consumer health applications that allow patients to view their health information. Portals facilitate the interactions between patients and their caregivers by offering secure messaging. Patients communicate different needs through portal messages. Medical needs contain requests for delivery of care (e.g. reporting new symptoms). Automating the classification of medical decision complexity in portal messages has not been investigated. MATERIALS AND METHODS: We trained two multiclass classifiers, multinomial Naïve Bayes and random forest on 500 message threads, to quantify and label the complexity of decision-making into four classes: no decision, straightforward, low, and moderate. We compared the performance of the models to using only the number of medical terms without training a machine learning model. RESULTS: Our analysis demonstrated that machine learning models have better performance than the model that did not use machine learning. Moreover, machine learning models could quantify the complexity of decision-making that the messages contained with 0.59, 0.45, and 0.58 for macro, micro, and weighted precision and 0.63,0.41, and 0.63 for macro, micro, and weighted recall. CONCLUSIONS: This study is one of the first to attempt to classify patient portal messages by whether they involve medical decision-making and the complexity of that decision-making. Machine learning classifiers trained on message content resulted in better message thread classification than classifiers that employed medical terms in the messages alone.

Association of Genetic Risk of Obesity with Postoperative Complications Using Mendelian Randomization.

BACKGROUND: The extent to which obesity and genetics determine postoperative complications is incompletely understood. METHODS: We performed a retrospective study using two population cohorts with electronic health record (EHR) data. The first included 736,726 adults with body mass index (BMI) recorded between 1990 and 2017 at Vanderbilt University Medical Center. The second cohort consisted of 65,174 individuals from 12 institutions contributing EHR and genome-wide genotyping data to the Electronic Medical Records and Genomics (eMERGE) Network. Pairwise logistic regression analyses were used to measure the association of BMI categories with postoperative complications derived from International Classification of Disease-9 codes, including postoperative infection, incisional hernia, and intestinal obstruction. A genetic risk score was constructed from 97 obesity-risk single-nucleotide polymorphisms for a Mendelian randomization study to determine the association of genetic risk of obesity on postoperative complications. Logistic regression analyses were adjusted for sex, age, site, and race/principal components. RESULTS: Individuals with overweight or obese BMI (≥25 kg/m2) had increased risk of incisional hernia (odds ratio [OR] 1.7-5.5, p < 3.1 × 10-20), and people with obesity (BMI ≥ 30 kg/m2) had increased risk of postoperative infection (OR 1.2-2.3, p < 2.5 × 10-5). In the eMERGE cohort, genetically predicted BMI was associated with incisional hernia (OR 2.1 [95% CI 1.8-2.5], p = 1.4 × 10-6) and postoperative infection (OR 1.6 [95% CI 1.4-1.9], p = 3.1 × 10-6). Association findings were similar after limitation of the cohorts to those who underwent abdominal procedures. CONCLUSIONS: Clinical and Mendelian randomization studies suggest that obesity, as measured by BMI, is associated with the development of postoperative incisional hernia and infection.

Care Delivered by Pediatric Surgical Specialties Through Patient Portal Messaging.

BACKGROUND: Patient portals are online applications that typically allow users to interact with providers using secure messaging. Portal messaging use and content have not been studied in pediatric surgical specialties. MATERIALS AND METHODS: We obtained all message threads initiated by pediatric patients/caregivers and sent to pediatric surgical providers through the Vanderbilt University Medical Center patient portal from June 1, 2014 to December 31, 2014. We collected patient demographics and providers' surgical specialties. We determined the number of message threads and individual messages sent by patients/caregivers and providers by specialty. Message content was analyzed by semantic types using a validated consumer health taxonomy. RESULTS: Most threads were about male (176, 60.3%), white (239, 81.8%), non-Hispanic (278, 95.2%) patients with a median age of 6 y (range: 0-21 y). A total of 292 message threads containing 1679 individual messages were sent with mean 5.8 (standard deviation [SD] 5.0) messages per thread. Messages were sent more frequently regarding younger patients (P = 0.001). Physicians directly contributed to 161 (55%) message threads. Otolaryngology received the most threads (123, 42.1%) and messages (790, 47.1%). Specialties exchanging the most messages per thread were cardiac surgery (mean 7.0, SD 11.7), and dermatology (7.0, SD 6.9). Most message threads (273, 93.5%) involved delivery of medical care with 123 (42.1%) involving appointments/scheduling; 99 (33.9%) medical problems; 81 (27.7%) treatments; 68 (23.3%) testing; and 29 (9.9%) referrals. CONCLUSIONS: Pediatric surgeons deliver substantial care within portal messages exchanged with pediatric patients and caregivers. Institutions adopting portals should consider effects on provider workload and potential disparities in access to care.

Enhancing Recovery After Kasai Portoenterostomy With Epidural Analgesia.

BACKGROUND: Biliary atresia (BA) is a rare obstructive cholangiopathy that presents in early infancy. The Kasai portoenterostomy (PE) improves survival with the native liver. Epidural analgesia is an appealing option to control pain in this fragile patient population, yet its safety, efficacy, and potential benefits remain unproven. METHODS: Patients undergoing PE for BA between 2001 and 2016 at a single institution were identified by ICD codes. Preoperative laboratories, procedure details, and recovery outcomes were reviewed retrospectively. Outcomes of interest were need for postoperative mechanical ventilation, pain scores, normalized opioid administration, return of bowel function, and length of hospital stay after PE. RESULTS: Of 47 infants undergoing PE for BA, 25 received epidural analgesia, and 22 did not. Infants with epidurals received less systemic opioids over the first 96 h postoperatively than those without (P < 0.001). Epidurals were associated with lower pain scores between 6 and 30 h postoperatively (P = 0.01 to 0.04), during which the highest median 6-h interval pain score was 0.2 (IQR 0-1.3) for patients with epidurals yet 2.1 (IQR 1.2-3.3) for patients without. Patients with epidurals (88%, n = 22) were more commonly extubated before leaving the operating room than those without (59%, n = 13; P = 0.02). No significant difference was observed in time to first bowel movement (P = 0.48) or first oral feed (P = 0.81). However, infants with epidurals had shorter hospital stays after PE than those without (6 d [IQR 5-7] versus 8 d [IQR 6.3-11], P = 0.01). No major complications were associated with epidural catheters. CONCLUSIONS: Epidural analgesia in patients undergoing PE for BA appears safe and effectively controls pain while minimizing the need for systemic opioids. Reduced need for mechanical ventilation postoperatively and shortened hospital stays serve as further evidence for using epidurals to enhance recovery after PE.

Measuring the Value of a Clinical Practice Guideline for Children With Perforated Appendicitis.

OBJECTIVE: To determine the incremental cost-effectiveness of a clinical practice guideline (CPG) compared with "usual care" for treatment of perforated appendicitis in children. Secondary objective was to compare cost analyses using hospital accounting system data versus data in the Pediatric Health Information System (PHIS). BACKGROUND: Value-based surgical care (outcomes relative to costs) is frequently touted, but outcomes and costs are rarely measured together. METHODS: During an 18-month period, 122 children with perforated appendicitis at a tertiary referral children's hospital were treated using an evidence-based CPG. Clinical outcomes and costs for the CPG cohort were compared with patients in the 30-month period before CPG implementation (n = 191 children). RESULTS: With CPG-directed care, intra-abdominal abscess rate decreased from 0.24 to 0.10 (adjusted risk ratio 0.44, 95% confidence interval [CI] 0.26-0.75). The rate of any adverse event decreased from 0.30 to 0.23 (adjusted risk ratio 0.82, 95% CI 0.58-1.17). Mean total hospital costs per patient (hospital accounting system) decreased from $16,466 to $10,528 (adjusted absolute difference-$5451, 95% CI -$7755 to -$3147), leading to estimated adjusted total savings of $665,022 during the study period. Costs obtained from the PHIS database also showed reduction with CPG-directed care (-$6669, 95% CI -$8949 to -$4389 per patient). In Bayesian cost-effectiveness analyses, likelihood that CPG was the dominant strategy was 91%. CONCLUSIONS: An evidence-based CPG increased the value of surgical care for children with perforated appendicitis by improving outcomes and lowering costs. Hospital cost accounting data and pre-existing cost data within the PHIS database provided similar results.

Complexity of medical decision-making in care provided by surgeons through patient portals.

BACKGROUND: Patient portals are online applications that allow patients to interact with healthcare organizations and view information. Portal messages exchanged between patients and providers contain diverse types of communications, including delivery of medical care. The types of communications and complexity of medical decision-making in portal messages sent to surgeons have not been studied. MATERIALS AND METHODS: We obtained all message threads initiated by patients and exchanged with surgical providers through the Vanderbilt University Medical Center patient portal from June 1 to December 31, 2014. Five hundred randomly selected messages were manually analyzed by two research team members to determine the types of communication (i.e., informational, medical, logistical, or social), whether medical care was delivered, and complexity of medical decision-making as defined for outpatient billing in each message thread. RESULTS: A total of 9408 message threads were sent to 401 surgical providers during the study period. In the 500 threads selected for detailed analysis, 1293 distinct issues were communicated, with an average of 2.6 issues per thread. Medical needs were communicated in 453 message threads (90.6%). Further, 339 message threads (67.8%) contained medical decision-making. Overall complexity of medical decision-making was straightforward in 210 messages (62%), low in 102 messages (30%), and moderate in 27 messages (8%). No highly complex decisions were made over portal messaging. CONCLUSIONS: Through patient portal messages, surgeons deliver substantial medical care with varied levels of medical complexity. Models for compensation of online care must be developed as consumer and surgeon adoption of these technologies increases.

Review of information technology for surgical patient care.

BACKGROUND: Electronic health records (EHRs), computerized provider order entry (CPOE), and patient portals have experienced increased adoption by health care systems. The objective of this study was to review evidence regarding the impact of such health information technologies (HIT) on surgical practice. MATERIALS AND METHODS: A search of Medline, EMBASE, CINAHL, and the Cochrane Library was performed to identify data-driven, nonsurvey studies about the effects of HIT on surgical care. Domain experts were queried for relevant articles. Two authors independently reviewed abstracts for inclusion criteria and analyzed full text of eligible articles. RESULTS: A total of 2890 citations were identified. Of them, 32 observational studies and two randomized controlled trials met eligibility criteria. EHR or CPOE improved appropriate antibiotic administration for surgical procedures in 13 comparative observational studies. Five comparative observational studies indicated that electronically generated operative notes had increased accuracy, completeness, and availability in the medical record. The Internet as an information resource about surgical procedures was generally inadequate. Surgical patients and providers demonstrated rapid adoption of patient portals, with increasing proportions of online versus inperson outpatient surgical encounters. CONCLUSIONS: The overall quality of evidence about the effects of HIT in surgical practice was low. Current data suggest an improvement in appropriate perioperative antibiotic administration and accuracy of operative reports from CPOE and EHR applications. Online consumer health educational resources and patient portals are popular among patients and families, but their impact has not been studied well in surgical populations. With increasing adoption of HIT, further research is needed to optimize the efficacy of such tools in surgical care.

Peri-operative risk factors for delayed gastric emptying after a pancreaticoduodenectomy.

BACKGROUND: Delayed gastric emptying (DGE) is a frequent cause of morbidity, prolonged hospital stay and readmission after a pancreaticoduodenectomy (PD). We sought to evaluate predictive peri-operative factors for DGE after a PD. METHODS: Four hundred and sixteen consecutive patients who underwent a PD at our tertiary referral centre were identified. Univariate and multivariate (MV) logistic regression models were used to assess peri-operative factors associated with the development of clinically significant DGE and a post-operative pancreatic fistula (POPF). RESULTS: DGE occurred in 24% of patients (n = 98) with Grades B and C occurring at 13.5% (n = 55) and 10.5% (n = 43), respectively. Using MV regression, a body mass index (BMI) ≥35 [odds ratio (OR) = 3.19], operating room (OR) length >5.5 h (OR = 2.72) and prophylactic octreotide use (OR = 2.04) were independently associated with an increased risk of DGE. DGE patients had a significantly longer median hospital stay (12 versus 7 days), higher 90-day readmission rates (32% versus 18%) and an increased incidence of a pancreatic fistula (59% versus 27%). When controlling for POPF, only OR length >5.5 h (OR 2.73) remained significantly associated with DGE. CONCLUSIONS: DGE remains a significant cause of morbidity, increased hospital stay and readmission after PD. Our findings suggest patients with a BMI ≥35 or longer OR times have a higher risk of DGE either independently or through the development of POPF. These patients should be considered for possible enteral feeding tube placement along with limited octreotide use to decrease the potential risk and consequences of DGE.

Driving Precision of Pediatric VTE Risk-stratification through Genetics.

This study addresses rising incidence of pediatric venous thromboembolism by validating a VTE phenotype and developing a polygenic risk score (PRS) using UK Biobank data. Our findings demonstrate predictive value of the PRS, enhancing VTE risk assessment in clinical settings. Future steps involve integrating the PRS into risk stratification models.

Developing and evaluating pediatric phecodes (Peds-Phecodes) for high-throughput phenotyping using electronic health records.

OBJECTIVE: Pediatric patients have different diseases and outcomes than adults; however, existing phecodes do not capture the distinctive pediatric spectrum of disease. We aim to develop specialized pediatric phecodes (Peds-Phecodes) to enable efficient, large-scale phenotypic analyses of pediatric patients. MATERIALS AND METHODS: We adopted a hybrid data- and knowledge-driven approach leveraging electronic health records (EHRs) and genetic data from Vanderbilt University Medical Center to modify the most recent version of phecodes to better capture pediatric phenotypes. First, we compared the prevalence of patient diagnoses in pediatric and adult populations to identify disease phenotypes differentially affecting children and adults. We then used clinical domain knowledge to remove phecodes representing phenotypes unlikely to affect pediatric patients and create new phecodes for phenotypes relevant to the pediatric population. We further compared phenome-wide association study (PheWAS) outcomes replicating known pediatric genotype-phenotype associations between Peds-Phecodes and phecodes. RESULTS: The Peds-Phecodes aggregate 15 533 ICD-9-CM codes and 82 949 ICD-10-CM codes into 2051 distinct phecodes. Peds-Phecodes replicated more known pediatric genotype-phenotype associations than phecodes (248 vs 192 out of 687 SNPs, P < .001). DISCUSSION: We introduce Peds-Phecodes, a high-throughput EHR phenotyping tool tailored for use in pediatric populations. We successfully validated the Peds-Phecodes using genetic replication studies. Our findings also reveal the potential use of Peds-Phecodes in detecting novel genotype-phenotype associations for pediatric conditions. We expect that Peds-Phecodes will facilitate large-scale phenomic and genomic analyses in pediatric populations. CONCLUSION: Peds-Phecodes capture higher-quality pediatric phenotypes and deliver superior PheWAS outcomes compared to phecodes.

Revolutionizing Healthcare with ChatGPT: An Early Exploration of an AI Language Model's Impact on Medicine at Large and its Role in Pediatric Surgery.

BACKGROUND: ChatGPT, a natural language processing model, has shown great promise in revolutionizing the field of medicine. This paper presents a comprehensive evaluation of the transformative potential of OpenAI's ChatGPT on healthcare and scientific research, with an exploration on its prospective capacity to impact the field of pediatric surgery. METHODS: Through an extensive review of the literature, we illuminate ChatGPT's applications in clinical healthcare and medical research while presenting the ethical considerations surrounding its use. RESULTS: Our review reveals the exciting work done so far evaluating the numerous potential uses of ChatGPT in clinical medicine and medical research, but it also shows that significant research and advancements in natural language processing models are still needed. CONCLUSION: ChatGPT has immense promise in transforming how we provide healthcare and how we conduct research. Currently, further robust research on the safety, effectiveness, and ethical considerations of ChatGPT is greatly needed. LEVEL OF STUDY: V.

Infectious Complications and Antibiotic Use in Gastroschisis.

Background: Gastroschisis is a challenging neonatal condition often with prolonged hospitalizations, need for parenteral nutrition, infectious complications, and can even result in death. Infection is reported to occur in up to two-thirds of patients with gastroschisis and is a strong risk factor for increased morbidity and mortality. Increased days with a central venous catheter, complex gastroschisis, and delayed abdominal wall closure have been consistently found to be associated with increased risk of infection, whereas sutureless gastroschisis closure has been associated with fewer infections. Although one of the most common complications of gastroschisis is infection, the use of antibiotic agents varies widely with variability in the literature to guide management. Antibiotic usage should be selective and short-term, especially in neonates with simple gastroschisis regardless of method for abdominal wall closure. Conclusions: Future initiatives should focus on development of evidence-based guidelines on the care of these patients with the goal of reducing variability and improve outcomes within and across institutions.

Developing and Evaluating Pediatric Phecodes (Peds-Phecodes) for High-Throughput Phenotyping Using Electronic Health Records.

Objective: Pediatric patients have different diseases and outcomes than adults; however, existing phecodes do not capture the distinctive pediatric spectrum of disease. We aim to develop specialized pediatric phecodes (Peds-Phecodes) to enable efficient, large-scale phenotypic analyses of pediatric patients. Materials and Methods: We adopted a hybrid data- and knowledge-driven approach leveraging electronic health records (EHRs) and genetic data from Vanderbilt University Medical Center to modify the most recent version of phecodes to better capture pediatric phenotypes. First, we compared the prevalence of patient diagnoses in pediatric and adult populations to identify disease phenotypes differentially affecting children and adults. We then used clinical domain knowledge to remove phecodes representing phenotypes unlikely to affect pediatric patients and create new phecodes for phenotypes relevant to the pediatric population. We further compared phenome-wide association study (PheWAS) outcomes replicating known pediatric genotype-phenotype associations between Peds-Phecodes and phecodes. Results: The Peds-Phecodes aggregate 15,533 ICD-9-CM codes and 82,949 ICD-10-CM codes into 2,051 distinct phecodes. Peds-Phecodes replicated more known pediatric genotype-phenotype associations than phecodes (248 versus 192 out of 687 SNPs, p<0.001). Discussion: We introduce Peds-Phecodes, a high-throughput EHR phenotyping tool tailored for use in pediatric populations. We successfully validated the Peds-Phecodes using genetic replication studies. Our findings also reveal the potential use of Peds-Phecodes in detecting novel genotype-phenotype associations for pediatric conditions. We expect that Peds-Phecodes will facilitate large-scale phenomic and genomic analyses in pediatric populations. Conclusion: Peds-Phecodes capture higher-quality pediatric phenotypes and deliver superior PheWAS outcomes compared to phecodes.

Quantifying the phenome-wide disease burden of obesity using electronic health records and genomics.

OBJECTIVE: High BMI is associated with many comorbidities and mortality. This study aimed to elucidate the overall clinical risk of obesity using a genome- and phenome-wide approach. METHODS: This study performed a phenome-wide association study of BMI using a clinical cohort of 736,726 adults. This was followed by genetic association studies using two separate cohorts: one consisting of 65,174 adults in the Electronic Medical Records and Genomics (eMERGE) Network and another with 405,432 participants in the UK Biobank. RESULTS: Class 3 obesity was associated with 433 phenotypes, representing 59.3% of all billing codes in individuals with severe obesity. A genome-wide polygenic risk score for BMI, accounting for 7.5% of variance in BMI, was associated with 296 clinical diseases, including strong associations with type 2 diabetes, sleep apnea, hypertension, and chronic liver disease. In all three cohorts, 199 phenotypes were associated with class 3 obesity and polygenic risk for obesity, including novel associations such as increased risk of renal failure, venous insufficiency, and gastroesophageal reflux. CONCLUSIONS: This combined genomic and phenomic systematic approach demonstrated that obesity has a strong genetic predisposition and is associated with a considerable burden of disease across all disease classes.

Prenatal diagnosis of rapidly enlarging choledochal cyst with gastric outlet obstruction.

Choledochal cysts are congenital malformations of the biliary tract that involve aberrant configurations of the pancreaticobiliary ductal system. The pathology exists on a spectrum from fusiform dilation of the common bile duct to multiple dilations involving the intra- and extrahepatic bile ducts with potential risks of malignant transformation and hepatic fibrosis. Advancements in ultrasound technology have increased the incidence of prenatal diagnosis of choledochal cysts. Here, we present the case of a prenatally diagnosed initially asymptomatic Type I choledochal cyst with rapid progression in the neonatal period to a complete gastric outlet obstruction within the first month of life. We demonstrate the feasibility of cyst resection and reconstruction with Roux-en-Y hepaticojejunostomy in the neonatal age group. Finally, we discuss management of the case based on evolving imaging findings and laboratory evidence of impending liver dysfunction.

What are the risks and the benefits of current and emerging weight-loss medications?

Obesity is epidemic; new medications and therapeutic options are urgently needed to reduce the associated health care burden. The initial clinical strategy for weight loss is lifestyle modification involving a combination of diet, exercise, and behavior change. However, it is difficult for many to achieve and maintain weight loss solely through this approach. Only two drugs, orlistat and sibutramine, have been approved by the US Food and Drug Administration (FDA) to treat obesity long term, and both medications have undesirable side effects, leaving an enormous unmet need for efficacious and safe therapy for obesity. Other medications with weight-loss effects have been approved by the FDA for short-term treatment of obesity or for disorders other than obesity, but these also have potential adverse effects. This article discusses the perceived benefits and risks of these approved medications along with emerging drugs that have shown weight-loss effects.

Defining Phenotypes from Clinical Data to Drive Genomic Research.

The rise in available longitudinal patient information in electronic health records (EHRs) and their coupling to DNA biobanks has resulted in a dramatic increase in genomic research using EHR data for phenotypic information. EHRs have the benefit of providing a deep and broad data source of health-related phenotypes, including drug response traits, expanding the phenome available to researchers for discovery. The earliest efforts at repurposing EHR data for research involved manual chart review of limited numbers of patients but now typically involve applications of rule-based and machine learning algorithms operating on sometimes huge corpora for both genome-wide and phenome-wide approaches. We highlight here the current methods, impact, challenges, and opportunities for repurposing clinical data to define patient phenotypes for genomics discovery. Use of EHR data has proven a powerful method for elucidation of genomic influences on diseases, traits, and drug-response phenotypes and will continue to have increasing applications in large cohort studies.

Rapid resolution of hyperammonemia in neonates using extracorporeal membrane oxygenation as a platform to drive hemodialysis.

OBJECTIVE: We aimed to clarify the impact of extracorporeal membrane oxygenation (ECMO) as a platform to drive hemodialysis (HD) for ammonia clearance on outcomes of neonates with severe hyperammonemia. STUDY DESIGN: All neonates treated for hyperammonemia at a single children's hospital between 1992 and 2016 were identified. Patient characteristics and outcomes were compared between those receiving medical management or ECMO/HD. RESULT: Twenty-five neonates were treated for hyperammonemia, of which 13 (52%) received ECMO/HD. Peak ammonia levels among neonates treated with ECMO/HD were significantly higher than those medically managed (1041 [IQR 902-1581] µmol/L versus 212 [IQR 110-410] µmol/L; p = 0.009). Serum ammonia levels in the ECMO/HD cohort declined to the median of medically managed within 4.5 (IQR 2.9-7.0) hours and normalized within 7.3 (IQR 3.6-13.5) hours. All neonates survived ECMO/HD, and nine (69.2%) survived to discharge. CONCLUSION: ECMO/HD is an effective adjunct to rapidly clear severe hyperammonemia in newborns, reducing potential neurodevelopmental morbidity.