Precision Anesthesia in 2050.

Over the last few decades, the field of anesthesia has advanced far beyond its humble beginnings. Today's anesthetics are better and safer than ever, thanks to innovations in drugs, monitors, equipment, and patient safety.1-4 At the same time, we remain limited by our herd approach to medicine. Each of our patients is unique, but health care today is based on a one-size-fits-all approach, while our patients grow older and more medically complex every year. By 2050, we believe that precision medicine will play a central role across all medical specialties, including anesthesia. In addition, we expect that health care and consumer technology will continually evolve to improve and simplify the interactions between patients, providers, and the health care system. As demonstrated by 2 hypothetical patient experiences, these advancements will enable more efficient and safe care, earlier and more accurate diagnoses, and truly personalized treatment plans.

Comparing ChatGPT and a Single Anesthesiologist's Responses to Common Patient Questions: An Exploratory Cross-Sectional Survey of a Panel of Anesthesiologists.

Increased patient access to electronic medical records and resources has resulted in higher volumes of health-related questions posed to clinical staff, while physicians' rising clinical workloads have resulted in less time for comprehensive, thoughtful responses to patient questions. Artificial intelligence chatbots powered by large language models (LLMs) such as ChatGPT could help anesthesiologists efficiently respond to electronic patient inquiries, but their ability to do so is unclear. A cross-sectional exploratory survey-based study comprised of 100 anesthesia-related patient question/response sets based on two fictitious simple clinical scenarios was performed. Each question was answered by an independent board-certified anesthesiologist and ChatGPT (GPT-3.5 model, August 3, 2023 version). The responses were randomized and evaluated via survey by three blinded board-certified anesthesiologists for various quality and empathy measures. On a 5-point Likert scale, ChatGPT received similar overall quality ratings (4.2 vs. 4.1, p = .81) and significantly higher overall empathy ratings (3.7 vs. 3.4, p < .01) compared to the anesthesiologist. ChatGPT underperformed the anesthesiologist regarding rate of responses in agreement with scientific consensus (96.6% vs. 99.3%, p = .02) and possibility of harm (4.7% vs. 1.7%, p = .04), but performed similarly in other measures (percentage of responses with inappropriate/incorrect information (5.7% vs. 2.7%, p = .07) and missing information (10.0% vs. 7.0%, p = .19)). In conclusion, LLMs show great potential in healthcare, but additional improvement is needed to decrease the risk of patient harm and reduce the need for close physician oversight. Further research with more complex clinical scenarios, clinicians, and live patients is necessary to validate their role in healthcare.

Machine Vision and Image Analysis in Anesthesia: Narrative Review and Future Prospects.

Machine vision describes the use of artificial intelligence to interpret, analyze, and derive predictions from image or video data. Machine vision-based techniques are already in clinical use in radiology, ophthalmology, and dermatology, where some applications currently equal or exceed the performance of specialty physicians in areas of image interpretation. While machine vision in anesthesia has many potential applications, its development remains in its infancy in our specialty. Early research for machine vision in anesthesia has focused on automated recognition of anatomical structures during ultrasound-guided regional anesthesia or line insertion; recognition of the glottic opening and vocal cords during video laryngoscopy; prediction of the difficult airway using facial images; and clinical alerts for endobronchial intubation detected on chest radiograph. Current machine vision applications measuring the distance between endotracheal tube tip and carina have demonstrated noninferior performance compared to board-certified physicians. The performance and potential uses of machine vision for anesthesia will only grow with the advancement of underlying machine vision algorithm technical performance developed outside of medicine, such as convolutional neural networks and transfer learning. This article summarizes recently published works of interest, provides a brief overview of techniques used to create machine vision applications, explains frequently used terms, and discusses challenges the specialty will encounter as we embrace the advantages that this technology may bring to future clinical practice and patient care. As machine vision emerges onto the clinical stage, it is critically important that anesthesiologists are prepared to confidently assess which of these devices are safe, appropriate, and bring added value to patient care.

A machine-learning approach for decision support and risk stratification of pediatric perioperative patients based on the APRICOT dataset.

BACKGROUND: Pediatric anesthesia has evolved to a high level of patient safety, yet a small chance remains for serious perioperative complications, even in those traditionally considered at low risk. In practice, prediction of at-risk patients currently relies on the American Society of Anesthesiologists Physical Status (ASA-PS) score, despite reported inconsistencies with this method. AIMS: The goal of this study was to develop predictive models that can classify children as low risk for anesthesia at the time of surgical booking and after anesthetic assessment on the procedure day. METHODS: Our dataset was derived from APRICOT, a prospective observational cohort study conducted by 261 European institutions in 2014 and 2015. We included only the first procedure, ASA-PS classification I to III, and perioperative adverse events not classified as drug errors, reducing the total number of records to 30 325 with an adverse event rate of 4.43%. From this dataset, a stratified train:test split of 70:30 was used to develop predictive machine learning algorithms that could identify children in ASA-PS class I to III at low risk for severe perioperative critical events that included respiratory, cardiac, allergic, and neurological complications. RESULTS: Our selected models achieved accuracies of >0.9, areas under the receiver operating curve of 0.6-0.7, and negative predictive values >95%. Gradient boosting models were the best performing for both the booking phase and the day-of-surgery phase. CONCLUSIONS: This work demonstrates that prediction of patients at low risk of critical PAEs can be made on an individual, rather than population-based, level by using machine learning. Our approach yielded two models that accommodate wide clinical variability and, with further development, are potentially generalizable to many surgical centers.

STBUR: Sleep trouble breathing and unrefreshed questionnaire: Evaluation of screening tool for postanesthesia care and disposition.

BACKGROUND: The Snoring, Trouble Breathing, and Un-Refreshed (STBUR) questionnaire is a five-question screening tool for pediatric sleep-disordered breathing and risk for perioperative respiratory adverse events in children. The utility of this questionnaire as a preoperative risk-stratification tool has not been investigated. In view of limited availability of screening tools for preoperative pediatric sleep-disordered breathing, we evaluated the questionnaire's performance for postanesthesia adverse events that can impact postanesthesia care and disposition. METHODS: The retrospective study protocol was approved by the institutional research board. The data were analyzed using two different definitions for a positive screening based on a five-point scale: low threshold (scores 1 to 5) and high threshold (score of 5). The primary outcome was based on the following criteria: (a) supplemental oxygen therapy following postanesthesia care unit (PACU) stay until hospital discharge, (b) greater than two hours during phase 1 recovery, (c) anesthesia emergency activation in the PACU, and (d) unplanned hospital admission. RESULTS: About 6025 patients completed the questionnaire during the preoperative evaluation. And 1522 patients had a low threshold score and 270 had a high-threshold score. We found statistically significant associations in three outcomes based on the low threshold score: supplemental oxygen therapy (negative-predictive value [NPV] 0.97, 95% CI 0.97-98), PACU recovery time (NPV 0.99, 95% CI 0.99-0.99) and escalation of care (NPV 0.98, 95% CI 0.97-0.98). Positive-predictive values were statistically significant for all outcomes except anesthesia emergency in the PACU. CONCLUSION: The Snoring, Trouble Breathing, and Un-Refreshed questionnaire identified patients at higher risk for prolonged phase 1 recovery, oxygen therapy requirement, and escalation of care. The questionnaire's high-negative predictive value and specificity may make it useful as a screening tool to identify patients at low risk for prolonged stay in PACU.

Early identification of impending cardiac arrest in neonates and infants in the cardiovascular ICU: a statistical modelling approach using physiologic monitoring data.

OBJECTIVE: To develop a physiological data-driven model for early identification of impending cardiac arrest in neonates and infants with cardiac disease hospitalised in the cardiovascular ICU. METHODS: We performed a single-institution retrospective cohort study (11 January 2013-16 September 2015) of patients ≤1 year old with cardiac disease who were hospitalised in the cardiovascular ICU at a tertiary care children's hospital. Demographics and diagnostic codes of cardiac arrest were obtained via the electronic health record. Diagnosis of cardiac arrest was validated by expert clinician review. Minute-to-minute physiological monitoring data were recorded via bedside monitors. A generalized linear model was used to compute a minute by minute risk score. Training and test data sets both included data from patients who did and did not develop cardiac arrest. An optimal risk-score threshold was derived based on the model's discriminatory capacity for impending arrest versus non-arrest. Model performance measures included sensitivity, specificity, accuracy, likelihood ratios, and post-test probability of arrest. RESULTS: The final model consisting of multiple clinical parameters was able to identify impending cardiac arrest at least 2 hours prior to the event with an overall accuracy of 75% (sensitivity = 61%, specificity = 80%) and observed an increase in probability of detection of cardiac arrest from a pre-test probability of 9.6% to a post-test probability of 21.2%. CONCLUSIONS: Our findings demonstrate that a predictive model using physiologic monitoring data in neonates and infants with cardiac disease hospitalised in the paediatric cardiovascular ICU can identify impending cardiac arrest on average 17 hours prior to arrest.

Interactive pediatric emergency checklists to the palm of your hand - How the Pedi Crisis App traveled around the world.

BACKGROUND: Cognitive aids help clinicians manage critical events and have been shown to improve outcomes by providing critical information at the point of care. Critical event guidelines, such as the Society of Pediatric Anesthesia's Critical Events Checklists described in this article, can be distributed globally via interactive smartphone apps. From October 1, 2013 to January 1, 2014, we performed an observational study to determine the global distribution and utilization patterns of the Pedi Crisis cognitive aid app that the Society for Pediatric Anesthesia developed. We analyzed distribution and utilization metrics of individuals using Pedi Crisis on iOS (Apple Inc., Cupertino, CA) devices worldwide. We used Google Analytics software (Google Inc., Mountain View, CA) to monitor users' app activity (eg, screen views, user sessions). METHODS: The primary outcome measurement was the number of user-sessions and geographic locations of Pedi Crisis user sessions. Each user was defined by the use of a unique Apple ID on an iOS device. RESULTS: Google Analytics correlates session activity with geographic location based on local Internet service provider logs. Pedi Crisis had 1 252 active users (both new and returning) and 4 140 sessions across 108 countries during the 3-month study period. Returning users used the app longer and viewed significantly more screens that new users (mean screen views: new users 1.3 [standard deviation +/-1.09, 95% confidence interval 1.22-1.55]; returning users 7.6 [standard deviation +/-4.19, 95% confidence interval 6.73-8.39]P<.01) CONCLUSIONS: Pedi Crisis was used worldwide within days of its release and sustained utilization beyond initial publication. The proliferation of handheld electronic devices provides a unique opportunity for professional societies to improve the worldwide dissemination of guidelines and evidence-based cognitive aids.

The timing and prevalence of intraoperative hypotension in infants undergoing laparoscopic pyloromyotomy at a tertiary pediatric hospital.

BACKGROUND: Intraoperative hypotension may be associated with adverse outcomes in children undergoing surgery. Infants and neonates under 6 months of age have less autoregulatory cerebral reserve than older infants, yet little information exists regarding when and how often intraoperative hypotension occurs in infants. AIMS: To better understand the epidemiology of intraoperative hypotension in infants, we aimed to determine the prevalence of intraoperative hypotension in a generally uniform population of infants undergoing laparoscopic pyloromyotomy. METHODS: Vital sign data from electronic records of infants who underwent laparoscopic pyloromyotomy with general anesthesia at a children's hospital between January 1, 1998 and October 4, 2013 were analyzed. Baseline blood pressure (BP) values and intraoperative BPs were identified during eight perioperative stages based on anesthesia event timestamps. We determined the occurrence of relative (systolic BP <20% below baseline) and absolute (mean arterial BP <35 mmHg) intraoperative hypotension within each stage. RESULTS: A total of 735 full-term infants and 82 preterm infants met the study criteria. Relative intraoperative hypotension occurred in 77%, 72%, and 58% of infants in the 1-30, 31-60, and 61-90 days age groups, respectively. Absolute intraoperative hypotension was seen in 21%, 12%, and 4% of infants in the 1-30, 31-60, and 61-90 days age groups, respectively. Intraoperative hypotension occurred primarily during surgical prep and throughout the surgical procedure. Preterm infants had higher rates of absolute intraoperative hypotension than full-term infants. CONCLUSIONS: Relative intraoperative hypotension was routine and absolute intraoperative hypotension was common in neonates and infants under 91 days of age. Preterm infants and infants under 61 days of age experienced the highest rates of absolute and relative intraoperative hypotension, particularly during surgical prep and throughout surgery.

Perioperative Smartphone Apps and Devices for Patient-Centered Care.

Smartphones have grown in ubiquity and computing power, and they play an ever-increasing role in patient-centered health care. The "medicalized smartphone" not only enables web-based access to patient health resources, but also can run patient-oriented software applications and be connected to health-related peripheral devices. A variety of patient-oriented smartphone apps and devices are available for use to facilitate patient-centered care throughout the continuum of perioperative care. Ongoing advances in smartphone technology and health care apps and devices should expand their utility for enhancing patient-centered care in the future.

A review of analytics and clinical informatics in health care.

Federal investment in health information technology has incentivized the adoption of electronic health record systems by physicians and health care organizations; the result has been a massive rise in the collection of patient data in electronic form (i.e. "Big Data"). Health care systems have leveraged Big Data for quality and performance improvements using analytics-the systematic use of data combined with quantitative as well as qualitative analysis to make decisions. Analytics have been utilized in various aspects of health care including predictive risk assessment, clinical decision support, home health monitoring, finance, and resource allocation. Visual analytics is one example of an analytics technique with an array of health care and research applications that are well described in the literature. The proliferation of Big Data and analytics in health care has spawned a growing demand for clinical informatics professionals who can bridge the gap between the medical and information sciences.

Using machine learning to predict five-year transplant-free survival among infants with hypoplastic left heart syndrome.

Hypoplastic left heart syndrome (HLHS) is a congenital malformation commonly treated with palliative surgery and is associated with significant morbidity and mortality. Risk stratification models have often relied upon traditional survival analyses or outcomes data failing to extend beyond infancy. Individualized prediction of transplant-free survival (TFS) employing machine learning (ML) based analyses of outcomes beyond infancy may provide further valuable insight for families and healthcare providers along the course of a staged palliation. Data from both the Pediatric Heart Network (PHN) Single Ventricle Reconstruction (SVR) trial and Extension study (SVR II), which extended cohort follow up for five years was used to develop ML-driven models predicting TFS. Models incrementally incorporated features corresponding to successive phases of care, from pre-Stage 1 palliation (S1P) through the stage 2 palliation (S2P) hospitalization. Models trained with features from Pre-S1P, S1P operation, and S1P hospitalization all demonstrated time-dependent area under the curves (td-AUC) beyond 0.70 through 5 years following S1P, with a model incorporating features through S1P hospitalization demonstrating particularly robust performance (td-AUC 0.838 (95% CI 0.836-0.840)). Machine learning may offer a clinically useful alternative means of providing individualized survival probability predictions, years following the staged surgical palliation of hypoplastic left heart syndrome.

Longitudinal Outcomes of Cumulative Impact Exposure on Oculomotor Functioning in Professional Motorsport Drivers.

Importance: Professional motorsport drivers are regularly exposed to biomechanical forces comparable with those experienced by contact and collision sport athletes, and little is known about the potential short-term and long-term neurologic sequelae. Objective: To determine whether cumulative impact exposure is associated with oculomotor functioning in motorsport drivers from the INDYCAR professional open-wheel automobile racing series. Design, Setting, and Participants: This is a longitudinal retrospective cohort study conducted across 3 racing seasons (2017-2019). Statistical analyses were conducted in November 2021. Data were retrieved from a secondary care setting associated with the INDYCAR series. INDYCAR series drivers who participated in 3 professional level racing seasons and were involved in at least 1 contact incident (ie, crash) in 2 of the 3 seasons were included in the study. Exposure: Cumulative acceleration and deceleration forces and total contact incidents (ie, crashes) measured via accident data recorder third generation chassis and ear accelerometers. Main Outcomes and Measures: Postseries oculomotor performance, including predictive saccades, vergence smooth pursuit, and optokinetic nystagmus, was measured annually with a head-mounted, clinical eye tracking system (Neurolign Dx 100). Results: Thirteen drivers (mean [SD] age, 29.36 [7.82] years; all men) sustained median resultant acceleration forces of 38.15 g (observed range, 12.01-93.05 g; 95% CI, 30.62-65.81 g) across 81 crashes. A 2-way multivariate analysis of variance did not reveal a statistically significant association between ear and chassis average resultant g forces, total number of contact incidents, and racing season assessed (F9,12 = 0.955; P = .54; Wilks Λ = 0.44). Conclusions and Relevance: In this cohort study of professional drivers from the INDYCAR series, there were no statistically significant associations among cumulative impact exposure, racing season assessed, and oculomotor performance. Longitudinal studies across racing seasons using multidimensional examination modalities (eg, neurocognitive testing, advanced imaging, biomarkers, and physical examination) are critical to understand potential neurological and neurobehavioral sequelae and long-term consequences of cumulative impact exposure.

Application of natural language processing to identify social needs from patient medical notes: development and assessment of a scalable, performant, and rule-based model in an integrated healthcare delivery system.

Objectives: To develop and test a scalable, performant, and rule-based model for identifying 3 major domains of social needs (residential instability, food insecurity, and transportation issues) from the unstructured data in electronic health records (EHRs). Materials and Methods: We included patients aged 18 years or older who received care at the Johns Hopkins Health System (JHHS) between July 2016 and June 2021 and had at least 1 unstructured (free-text) note in their EHR during the study period. We used a combination of manual lexicon curation and semiautomated lexicon creation for feature development. We developed an initial rules-based pipeline (Match Pipeline) using 2 keyword sets for each social needs domain. We performed rule-based keyword matching for distinct lexicons and tested the algorithm using an annotated dataset comprising 192 patients. Starting with a set of expert-identified keywords, we tested the adjustments by evaluating false positives and negatives identified in the labeled dataset. We assessed the performance of the algorithm using measures of precision, recall, and F1 score. Results: The algorithm for identifying residential instability had the best overall performance, with a weighted average for precision, recall, and F1 score of 0.92, 0.84, and 0.92 for identifying patients with homelessness and 0.84, 0.82, and 0.79 for identifying patients with housing insecurity. Metrics for the food insecurity algorithm were high but the transportation issues algorithm was the lowest overall performing metric. Discussion: The NLP algorithm in identifying social needs at JHHS performed relatively well and would provide the opportunity for implementation in a healthcare system. Conclusion: The NLP approach developed in this project could be adapted and potentially operationalized in the routine data processes of a healthcare system.

Survival analysis for pediatric heart transplant patients using a novel machine learning algorithm: A UNOS analysis.

BACKGROUND: Impact of pretransplantation risk factors on mortality in the first year after heart transplantation remains largely unknown. Using machine learning algorithms, we selected clinically relevant identifiers that could predict 1-year mortality after pediatric heart transplantation. METHODS: Data were obtained from the United Network for Organ Sharing Database for years 2010-2020 for patients 0-17 years receiving their first heart transplant (N = 4150). Features were selected using subject experts and literature review. Scikit-Learn, Scikit-Survival, and Tensorflow were used. A train:test split of 70:30 was used. N-repeated k-fold validation was performed (N = 5, k = 5). Seven models were tested, Hyperparameter tuning performed using Bayesian optimization and the concordance index (C-index) was used for model assessment. RESULTS: A C-index above 0.6 for test data was considered acceptable for survival analysis models. C-indices obtained were 0.60 (Cox proportional hazards), 0.61 (Cox with elastic net), 0.64 (gradient boosting), 0.64 (support vector machine), 0.68 (random forest), 0.66 (component gradient boosting), and 0.54 (survival trees). Machine learning models show an improvement over the traditional Cox proportional hazards model, with random forest performing the best on the test set. Analysis of the feature importance for the gradient boosted model found that the top 5 features were the most recent serum total bilirubin, the travel distance from the transplant center, the patient body mass index, the deceased donor terminal Serum glutamic pyruvic transaminase/Alanine transaminase (SGPT/ALT), and the donor PCO2. CONCLUSIONS: Combination of machine learning and expert-based methodology of selecting predictors of survival for pediatric heart transplantation provides a reasonable prediction of 1- and 3-year survival outcomes. SHapley Additive exPlanations can be an effective tool for modeling and visualizing nonlinear interactions.

Examining Implicit Bias Differences in Pediatric Surgical Fellowship Letters of Recommendation Using Natural Language Processing.

OBJECTIVE: We analyzed the prevalence and type of bias in letters of recommendation (LOR) for pediatric surgical fellowship applications from 2016-2021 using natural language processing (NLP) at a quaternary care academic hospital. DESIGN: Demographics were extracted from submitted applications. The Valence Aware Dictionary for sEntiment Reasoning (VADER) model was used to calculate polarity scores. The National Research Council dataset was used for emotion and intensity analysis.  The Kruskal-Wallis H-test was used to determine statistical significance.  SETTING: This study took place at a single, academic, free standing quaternary care children's hospital with an ACGME accredited pediatric surgery fellowship. PARTICIPANTS: Applicants to a single pediatric surgery fellowship were selected for this study from 2016 to 2021. A total of 182 individual applicants were included and 701 letters of recommendation were analyzed. RESULTS: Black applicants had the highest mean polarity (most positive), while Hispanic applicants had the lowest.  Overall differences between polarity distributions were not statistically significant.   The intensity of emotions showed that differences in "anger" were statistically significant (p=0.03).  Mean polarity was higher for applicants that successfully matched in pediatric surgery. DISCUSSION: This study identified differences in LORs based on racial and gender demographics submitted as part of pediatric surgical fellowship applications to a single training program. The presence of bias in letters of recommendation can lead to inequities in demographics to a given program. While difficult to detect for humans, natural language processing is able to detect bias as well as differences in polarity and emotional intensity. While the types of emotions identified in this study are highly similar among race and gender groups, the intensity of these emotions revealed differences, with "anger" being most significant. CONCLUSION: From this work, it can be concluded that bias in LORs, as reflected as differences in polarity, which is likely a result of the intensity of the emotions being used and not the types of emotions being expressed.   Natural language processing shows promise in identification of subtle areas of bias that may influence an individual's likelihood of successful matching.

"P3": an adaptive modeling tool for post-COVID-19 restart of surgical services.

OBJECTIVE: To develop a predictive analytics tool that would help evaluate different scenarios and multiple variables for clearance of surgical patient backlog during the COVID-19 pandemic. MATERIALS AND METHODS: Using data from 27 866 cases (May 1 2018-May 1 2020) stored in the Johns Hopkins All Children's data warehouse and inputs from 30 operations-based variables, we built mathematical models for (1) time to clear the case backlog (2), utilization of personal protective equipment (PPE), and (3) assessment of overtime needs. RESULTS: The tool enabled us to predict desired variables, including number of days to clear the patient backlog, PPE needed, staff/overtime needed, and cost for different backlog reduction scenarios. CONCLUSIONS: Predictive analytics, machine learning, and multiple variable inputs coupled with nimble scenario-creation and a user-friendly visualization helped us to determine the most effective deployment of operating room personnel. Operating rooms worldwide can use this tool to overcome patient backlog safely.