Remote Monitoring and Artificial Intelligence: Outlook for 2050.

Remote monitoring and artificial intelligence will become common and intertwined in anesthesiology by 2050. In the intraoperative period, technology will lead to the development of integrated monitoring systems that will integrate multiple data streams and allow anesthesiologists to track patients more effectively. This will free up anesthesiologists to focus on more complex tasks, such as managing risk and making value-based decisions. This will also enable the continued integration of remote monitoring and control towers having profound effects on coverage and practice models. In the PACU and ICU, the technology will lead to the development of early warning systems that can identify patients who are at risk of complications, enabling early interventions and more proactive care. The integration of augmented reality will allow for better integration of diverse types of data and better decision-making. Postoperatively, the proliferation of wearable devices that can monitor patient vital signs and track their progress will allow patients to be discharged from the hospital sooner and receive care at home. This will require increased use of telemedicine, which will allow patients to consult with doctors remotely. All of these advances will require changes to legal and regulatory frameworks that will enable new workflows that are different from those familiar to today's providers.

Laboratory Result Reference Ranges Stratified for Patient-Reported Sex and Ethnicity Are More Closely Associated With Postoperative Outcomes Than Currently Used Reference Ranges: A Retrospective Data Analysis.

BACKGROUND: A single laboratory range for all individuals may fail to take into account underlying physiologic differences based on sex and genetic factors. We hypothesized that laboratory distributions differ based on self-reported sex and ethnicity and that ranges stratified by these factors better correlate with postoperative mortality and acute kidney injury (AKI). METHODS: Results from metabolic panels, complete blood counts, and coagulation panels for patients in outpatient encounters were identified from our electronic health record. Patients were grouped based on self-reported sex (2 groups) and ethnicity (6 groups). Stratified ranges were set to be the 2.5th/97.5th percentile for each sex/ethnic group. For patients undergoing procedures, each patient/laboratory result was classified as normal/abnormal using the stratified and nonstratified (traditional) ranges; overlap in the definitions was assessed between the 2 classifications by looking for the percentage of agreement in result classifications of normal/abnormal using the 2 methods. To assess which definitions of normal are most associated with adverse postoperative outcomes, the odds ratio (OR) for each outcome/laboratory result pair was assessed, and the frequency that the confidence intervals of ORs for the stratified versus nonstratified range did not overlap was examined. RESULTS: Among the 300 unique combinations (race × sex × laboratory type), median proportion overlap (meaning patient was either "normal" or "abnormal" for both methodologies) was 0.86 [q1, 0.80; q3, 0.89]. All laboratory results except 6 overlapped at least 80% of the time. The frequency of overlap did not differ among the racial/ethnic groups. In cases where the ORs were different, the stratified range was better associated with both AKI and mortality (P < .001). There was no trend of bias toward any specific sex/ethnic group. CONCLUSIONS: Baseline "normal" laboratory values differ across sex and ethnic groups, and ranges stratified by these groups are better associated with postoperative AKI and mortality as compared to the standard reference ranges.

Visual Analytics to Leverage Anesthesia Electronic Health Record.

BACKGROUND: Visual analytics is the science of analytical reasoning supported by interactive visual interfaces called dashboards. In this report, we describe our experience addressing the challenges in visual analytics of anesthesia electronic health record (EHR) data using a commercially available business intelligence (BI) platform. As a primary outcome, we discuss some performance metrics of the dashboards, and as a secondary outcome, we outline some operational enhancements and financial savings associated with deploying the dashboards. METHODS: Data were transferred from the EHR to our departmental servers using several parallel processes. A custom structured query language (SQL) query was written to extract the relevant data fields and to clean the data. Tableau was used to design multiple dashboards for clinical operation, performance improvement, and business management. RESULTS: Before deployment of the dashboards, detailed case counts and attributions were available for the operating rooms (ORs) from perioperative services; however, the same level of detail was not available for non-OR locations. Deployment of the yearly case count dashboards provided near-real-time case count information from both central and non-OR locations among multiple campuses, which was not previously available. The visual presentation of monthly data for each year allowed us to recognize seasonality in case volumes and adjust our supply chain to prevent shortages. The dashboards highlighted the systemwide volume of cases in our endoscopy suites, which allowed us to target these supplies for pricing negotiations, with an estimated annual cost savings of $250,000. Our central venous pressure (CVP) dashboard enabled us to provide individual practitioner feedback, thus increasing our monthly CVP checklist compliance from approximately 92% to 99%. CONCLUSIONS: The customization and visualization of EHR data are both possible and worthwhile for the leveraging of information into easily comprehensible and actionable data for the improvement of health care provision and practice management. Limitations inherent to EHR data presentation make this customization necessary, and continued open access to the underlying data set is essential.

A Retrospective Analysis Demonstrates That a Failure to Document Key Comorbid Diseases in the Anesthesia Preoperative Evaluation Associates With Increased Length of Stay and Mortality.

BACKGROUND: The introduction of electronic health records (EHRs) has helped physicians access relevant medical information on their patients. However, the design of EHRs can make it hard for clinicians to easily find, review, and document all of the relevant data, leading to documentation that is not fully reflective of the complete history. We hypothesized that the incidence of undocumented key comorbid diseases (atrial fibrillation [afib], congestive heart failure [CHF], chronic obstructive pulmonary disease [COPD], diabetes, and chronic kidney disease [CKD]) in the anesthesia preoperative evaluation was associated with increased postoperative length of stay (LOS) and mortality. METHODS: Charts of patients >18 years who received anesthesia in an inpatient facility were reviewed in this retrospective study. For each disease, a precise algorithm was developed to look for key structured data (medications, lab results, structured medical history, etc) in the EHR. Additionally, the checkboxes from the anesthesia preoperative evaluation were queried to determine the presence or absence of the documentation of the disease. Differences in mortality were modeled with logistic regression, and LOS was analyzed using linear regression. RESULTS: A total of 91,011 cases met inclusion criteria (age 18-89 years; 52% women, 48% men; 70% admitted from home). Agreement between the algorithms and the preoperative note was >84% for all comorbidities other than chronic pain (63.5%). The algorithm-detected disease not documented by the anesthesia team in 34.5% of cases for chronic pain (vs 1.9% of cases where chronic pain was documented but not detected by the algorithm), 4.0% of cases for diabetes (vs 2.1%), 4.3% of cases for CHF (vs 0.7%), 4.3% of cases for COPD (vs 1.1%), 7.7% of cases for afib (vs 0.3%), and 10.8% of cases for CKD (vs 1.7%). To assess the association of missed documentation with outcomes, we compared patients where the disease was detected by the algorithm but not documented (A+/P-) with patients where the disease was documented (A+/P+). For all diseases except chronic pain, the missed documentation was associated with a longer LOS. For mortality, the discrepancy was associated with increased mortality for afib, while the differences were insignificant for the other diseases. For each missed disease, the odds of mortality increased 1.52 (95% confidence interval [CI], 1.42-1.63) and the LOS increased by approximately 11%, geometric mean ratio of 1.11 (95% CI, 1.10-1.12). CONCLUSIONS: Anesthesia preoperative evaluations not infrequently fail to document disease for which there is evidence of disease in the EHR data. This missed documentation is associated with an increased LOS and mortality in perioperative patients.

Successful Implementation of a Perioperative Data Warehouse Using Another Hospital's Published Specification From Epic's Electronic Health Record System.

BACKGROUND: Many hospitals have replaced their legacy anesthesia information management system with an enterprise-wide electronic health record system. Integrating the anesthesia data within the context of the global hospital information infrastructure has created substantive challenges for many organizations. A process to build a perioperative data warehouse from Epic was recently published from the University of California Los Angeles (UCLA), but the generalizability of that process is unknown. We describe the implementation of their process at the University of Miami (UM). METHODS: The UCLA process was tested at UM, and performance was evaluated following the configuration of a reporting server and transfer of the required Clarity tables to that server. Modifications required for the code to execute correctly in the UM environment were identified and implemented, including the addition of locally specified elements in the database. RESULTS: The UCLA code to build the base tables in the perioperative data warehouse executed correctly after minor modifications to match the local server and database architecture at UM. The 26 stored procedures in the UCLA process all ran correctly using the default settings provided and populated the base tables. After modification of the item lists to reflect the UM implementation of Epic (eg, medications, laboratory tests, physiologic monitors, and anesthesia machine parameters), the UCLA code ran correctly and populated the base tables. The data from those tables were used successfully to populate the existing perioperative data warehouse at UM, which housed data from the legacy anesthesia information management system of the institution. The time to pull data from Epic and populate the perioperative data warehouse was 197 ± 47 minutes (standard deviation [SD]) on weekdays and 260 ± 56 minutes (SD) on weekend days, measured over 100 consecutive days. The longer times on weekends reflect the simultaneous execution of database maintenance tasks on the reporting server. The UCLA extract process has been in production at UM for the past 18 months and has been invaluable for quality assurance, business process, and research activities. CONCLUSIONS: The data schema developed at UCLA proved to be a practical and scalable method to extract information from the Epic electronic health system database into the perioperative data warehouse in use at UM. Implementing the process developed at UCLA to build a comprehensive perioperative data warehouse from Epic is an extensible process that other hospitals seeking more efficient access to their electronic health record data should consider.

Automated Assessment of Existing Patient's Revised Cardiac Risk Index Using Algorithmic Software.

BACKGROUND: Previous work in the field of medical informatics has shown that rules-based algorithms can be created to identify patients with various medical conditions; however, these techniques have not been compared to actual clinician notes nor has the ability to predict complications been tested. We hypothesize that a rules-based algorithm can successfully identify patients with the diseases in the Revised Cardiac Risk Index (RCRI). METHODS: Patients undergoing surgery at the University of California, Los Angeles Health System between April 1, 2013 and July 1, 2016 and who had at least 2 previous office visits were included. For each disease in the RCRI except renal failure-congestive heart failure, ischemic heart disease, cerebrovascular disease, and diabetes mellitus-diagnosis algorithms were created based on diagnostic and standard clinical treatment criteria. For each disease state, the prevalence of the disease as determined by the algorithm, International Classification of Disease (ICD) code, and anesthesiologist's preoperative note were determined. Additionally, 400 American Society of Anesthesiologists classes III and IV cases were randomly chosen for manual review by an anesthesiologist. The sensitivity, specificity, accuracy, positive predictive value, negative predictive value, and area under the receiver operating characteristic curve were determined using the manual review as a gold standard. Last, the ability of the RCRI as calculated by each of the methods to predict in-hospital mortality was determined, and the time necessary to run the algorithms was calculated. RESULTS: A total of 64,151 patients met inclusion criteria for the study. In general, the incidence of definite or likely disease determined by the algorithms was higher than that detected by the anesthesiologist. Additionally, in all disease states, the prevalence of disease was always lowest for the ICD codes, followed by the preoperative note, followed by the algorithms. In the subset of patients for whom the records were manually reviewed, the algorithms were generally the most sensitive and the ICD codes the most specific. When computing the modified RCRI using each of the methods, the modified RCRI from the algorithms predicted in-hospital mortality with an area under the receiver operating characteristic curve of 0.70 (0.67-0.73), which compared to 0.70 (0.67-0.72) for ICD codes and 0.64 (0.61-0.67) for the preoperative note. On average, the algorithms took 12.64 ± 1.20 minutes to run on 1.4 million patients. CONCLUSIONS: Rules-based algorithms for disease in the RCRI can be created that perform with a similar discriminative ability as compared to physician notes and ICD codes but with significantly increased economies of scale.

Opening the Black Box: Understanding the Science Behind Big Data and Predictive Analytics.

Big data, smart data, predictive analytics, and other similar terms are ubiquitous in the lay and scientific literature. However, despite the frequency of usage, these terms are often poorly understood, and evidence of their disruption to clinical care is hard to find. This article aims to address these issues by first defining and elucidating the term big data, exploring the ways in which modern medical data, both inside and outside the electronic medical record, meet the established definitions of big data. We then define the term smart data and discuss the transformations necessary to make big data into smart data. Finally, we examine the ways in which this transition from big to smart data will affect what we do in research, retrospective work, and ultimately patient care.

Perioperative Temperature Measurement Considerations Relevant to Reporting Requirements for National Quality Programs Using Data From Anesthesia Information Management Systems.

BACKGROUND: Perioperative hypothermia may increase the incidences of wound infection, blood loss, transfusion, and cardiac morbidity. US national quality programs for perioperative normothermia specify the presence of at least 1 "body temperature" ≥35.5°C during the interval from 30 minutes before to 15 minutes after the anesthesia end time. Using data from 4 academic hospitals, we evaluated timing and measurement considerations relevant to the current requirements to guide hospitals wishing to report perioperative temperature measures using electronic data sources. METHODS: Anesthesia information management system databases from 4 hospitals were queried to obtain intraoperative temperatures and intervals to the anesthesia end time from discontinuation of temperature monitoring, end of surgery, and extubation. Inclusion criteria included age >16 years, use of a tracheal tube or supraglottic airway, and case duration ≥60 minutes. The end-of-case temperature was determined as the maximum intraoperative temperature recorded within 30 minutes before the anesthesia end time (ie, the temperature that would be used for reporting purposes). The fractions of cases with intervals >30 minutes between the last intraoperative temperature and the anesthesia end time were determined. RESULTS: Among the hospitals, averages (binned by quarters) of 34.5% to 59.5% of cases had intraoperative temperature monitoring discontinued >30 minutes before the anesthesia end time. Even if temperature measurement had been continued until extubation, averages of 5.9% to 20.8% of cases would have exceeded the allowed 30-minute window. Averages of 8.9% to 21.3% of cases had end-of-case intraoperative temperatures <35.5°C (ie, a quality measure failure). CONCLUSIONS: Because of timing considerations, a substantial fraction of cases would have been ineligible to use the end-of-case intraoperative temperature for national quality program reporting. Thus, retrieval of postanesthesia care unit temperatures would have been necessary. A substantive percentage of cases had end-of-case intraoperative temperatures below the 35.5°C threshold, also requiring postoperative measurement to determine whether the quality measure was satisfied. Institutions considering reporting national quality measures for perioperative normothermia should consider the technical and logistical issues identified to achieve a high level of compliance based on the specified regulatory language.

Creation and Validation of an Automated Algorithm to Determine Postoperative Ventilator Requirements After Cardiac Surgery.

BACKGROUND: In medical practice today, clinical data registries have become a powerful tool for measuring and driving quality improvement, especially among multicenter projects. Registries face the known problem of trying to create dependable and clear metrics from electronic medical records data, which are typically scattered and often based on unreliable data sources. The Society for Thoracic Surgery (STS) is one such example, and it supports manually collected data by trained clinical staff in an effort to obtain the highest-fidelity data possible. As a possible alternative, our team designed an algorithm to test the feasibility of producing computer-derived data for the case of postoperative mechanical ventilation hours. In this article, we study and compare the accuracy of algorithm-derived mechanical ventilation data with manual data extraction. METHODS: We created a novel algorithm that is able to calculate mechanical ventilation duration for any postoperative patient using raw data from our EPIC electronic medical record. Utilizing nursing documentation of airway devices, documentation of lines, drains, and airways, and respiratory therapist ventilator settings, the algorithm produced results that were then validated against the STS registry. This enabled us to compare our algorithm results with data collected by human chart review. Any discrepancies were then resolved with manual calculation by a research team member. RESULTS: The STS registry contained a total of 439 University of California Los Angeles cardiac cases from April 1, 2013, to March 31, 2014. After excluding 201 patients for not remaining intubated, tracheostomy use, or for having 2 surgeries on the same day, 238 cases met inclusion criteria. Comparing the postoperative ventilation durations between the 2 data sources resulted in 158 (66%) ventilation durations agreeing within 1 hour, indicating a probable correct value for both sources. Among the discrepant cases, the algorithm yielded results that were exclusively correct in 75 (93.8%) cases, whereas the STS results were exclusively correct once (1.3%). The remaining 4 cases had inconclusive results after manual review because of a prolonged documentation gap between mechanical and spontaneous ventilation. In these cases, STS and algorithm results were different from one another but were both within the transition timespan. This yields an overall accuracy of 99.6% (95% confidence interval, 98.7%-100%) for the algorithm when compared with 68.5% (95% confidence interval, 62.6%-74.4%) for the STS data (P < .001). CONCLUSIONS: There is a significant appeal to having a computer algorithm capable of calculating metrics such as total ventilator times, especially because it is labor intensive and prone to human error. By incorporating 3 different sources into our algorithm and by using preprogrammed clinical judgment to overcome common errors with data entry, our results proved to be more comprehensive and more accurate, and they required a fraction of the computation time compared with manual review.

A Systematic Approach to Creation of a Perioperative Data Warehouse.

Extraction of data from the electronic medical record is becoming increasingly important for quality improvement initiatives such as the American Society of Anesthesiologists Perioperative Surgical Home. To meet this need, the authors have built a robust and scalable data mart based on their implementation of EPIC containing data from across the perioperative period. The data mart is structured in such a way so as to first simplify the overall EPIC reporting structure into a series of Base Tables and then create several Reporting Schemas each around a specific concept (operating room cases, obstetrics, hospital admission, etc.), which contain all of the data required for reporting on various metrics. This structure allows centralized definitions with simplified reporting by a large number of individuals who access only the Reporting Schemas. In creating the database, the authors were able to significantly reduce the number of required table identifiers from >10 to 3, as well as to correct errors in linkages affecting up to 18.4% of cases. In addition, the data mart greatly simplified the code required to extract data, making the data accessible to individuals who lacked a strong coding background. Overall, this infrastructure represents a scalable way to successfully report on perioperative EPIC data while standardizing the definitions and improving access for end users.

Post-operative urinary retention is impacted by neuromuscular block reversal agent choice: A retrospective cohort study in US hospital setting.

STUDY OBJECTIVE: Perioperative neuromuscular blocking agents are pharmacologically reversed to minimize complications associated with residual neuromuscular block. Neuromuscular block reversal with anticholinesterases (e.g., neostigmine) require coadministration of an anticholinergic agent (e.g., glycopyrrolate) to mitigate muscarinic activity; however, sugammadex, devoid of cholinergic activity, does not require anticholinergic coadministration. Single-institution studies have found decreased incidence of post-operative urinary retention associated with sugammadex reversal. This study used a multicenter database to better understand the association between neuromuscular block reversal technique and post-operative urinary retention. DESIGN: Retrospective cohort study utilizing large healthcare database. SETTING: Non-profit, non-governmental and community and teaching hospitals and health systems from rural and urban areas. PATIENTS: 61,898 matched adult inpatients and 95,500 matched adult outpatients. INTERVENTIONS: Neuromuscular block reversal with sugammadex or neostigmine plus glycopyrrolate. MEASUREMENTS: Incidence of post-operative urinary retention by neuromuscular block reversal agent and the independent association of neuromuscular block reversal technique and risk of post-operative urinary retention. MAIN RESULTS: The incidence of post-operative urinary retention was 2-fold greater among neostigmine with glycopyrrolate compared to sugammadex patients (5.0% vs 2.4% inpatients; 0.9% vs 0.4% outpatients; both p < 0.0001). Multivariable logistic regression identified reversal with neostigmine to be independently associated with greater risk of post-operative urinary retention (inpatients: odds ratio, 2.20; 95% confidence interval, 2.00 to 2.41; p < 0.001; outpatients: odds ratio, 2.57; 95% confidence interval, 2.13 to 3.10; p < 0.001). Post-operative urinary retention-related visits within 2 days following discharge were five-fold higher among those reversed with neostigmine than sugammadex among inpatients (0.05% vs. 0.01%, respectively; p = 0.018) and outpatients (0.5% vs. 0.1%; p < 0.0001). CONCLUSION: Though this study suggests that neuromuscular block reversal with neostigmine can increase post-operative urinary retention risk, additional studies are needed to fully understand the association.

Corrigendum to "Validation of a convolutional neural network that reliably identifies electromyographic compound motor action potentials following train-of-four stimulation: an algorithm development experimental study" [BJA Open 8 (2023) 100236].

[This corrects the article DOI: 10.1016/j.bjao.2023.100236.].

Postoperative Acute Kidney Injury is Associated with Persistent Renal Dysfunction: A Multicenter Propensity Matched Cohort Study.

Background: The risk of developing a persistent reduction in renal function after postoperative acute kidney injury (pAKI) is not well-established. Objective: Perform a multi-center retrospective propensity matched study evaluating whether patients that develop pAKI have a greater decline in long-term renal function than patients that did not develop postoperative AKI. Design: Multi-center retrospective propensity matched study. Setting: Anesthesia data warehouses at three tertiary care hospitals were queried. Patients: Adult patients undergoing surgery with available preoperative and postoperative creatinine results and without baseline hemodialysis requirements. Measurements: The primary outcome was a decline in follow-up glomerular filtration rate (GFR) of 40% relative to baseline, based on follow-up outpatient visits from 0-36 months after hospital discharge. A propensity score matched sample was used in Kaplan-Meier analysis and in a piecewise Cox model to compare time to first 40% decline in GFR for patients with and without pAKI. Results: A total of 95,208 patients were included. The rate of pAKI ranged from 9.9% to 13.7%. In the piecewise Cox model, pAKI significantly increased the hazard of a 40% decline in GFR. The common effect hazard ratio was 13.35 (95% CI: 10.79 to 16.51, p<0.001) for 0-6 months, 7.07 (5.52 to 9.05, p<0.001) for 6-12 months, 6.02 (4.69 to 7.74, p<0.001) for 12-24 months, and 4.32 (2.65 to 7.05, p<0.001) for 24-36 months. Limitations: Retrospective; Patients undergoing ambulatory surgery without postoperative lab tests drawn before discharge were not captured; certain variables like postoperative urine output were not reliably available. Conclusion: Postoperative AKI significantly increases the risk of a 40% decline in GFR up to 36 months after the index surgery across three institutions.

Assessing calibration and bias of a deployed machine learning malnutrition prediction model within a large healthcare system.

Malnutrition is a frequently underdiagnosed condition leading to increased morbidity, mortality, and healthcare costs. The Mount Sinai Health System (MSHS) deployed a machine learning model (MUST-Plus) to detect malnutrition upon hospital admission. However, in diverse patient groups, a poorly calibrated model may lead to misdiagnosis, exacerbating health care disparities. We explored the model's calibration across different variables and methods to improve calibration. Data from adult patients admitted to five MSHS hospitals from January 1, 2021 - December 31, 2022, were analyzed. We compared MUST-Plus prediction to the registered dietitian's formal assessment. Hierarchical calibration was assessed and compared between the recalibration sample (N = 49,562) of patients admitted between January 1, 2021 - December 31, 2022, and the hold-out sample (N = 17,278) of patients admitted between January 1, 2023 - September 30, 2023. Statistical differences in calibration metrics were tested using bootstrapping with replacement. Before recalibration, the overall model calibration intercept was -1.17 (95% CI: -1.20, -1.14), slope was 1.37 (95% CI: 1.34, 1.40), and Brier score was 0.26 (95% CI: 0.25, 0.26). Both weak and moderate measures of calibration were significantly different between White and Black patients and between male and female patients. Logistic recalibration significantly improved calibration of the model across race and gender in the hold-out sample. The original MUST-Plus model showed significant differences in calibration between White vs. Black patients. It also overestimated malnutrition in females compared to males. Logistic recalibration effectively reduced miscalibration across all patient subgroups. Continual monitoring and timely recalibration can improve model accuracy.

External validation of a predictive model for reintubation after cardiac surgery: A retrospective, observational study.

STUDY OBJECTIVE: Explore validation of a model to predict patients' risk of failing extubation, to help providers make informed, data-driven decisions regarding the optimal timing of extubation. DESIGN: We performed temporal, geographic, and domain validations of a model for the risk of reintubation after cardiac surgery by assessing its performance on data sets from three academic medical centers, with temporal validation using data from the institution where the model was developed. SETTING: Three academic medical centers in the United States. PATIENTS: Adult patients arriving in the cardiac intensive care unit with an endotracheal tube in place after cardiac surgery. INTERVENTIONS: Receiver operating characteristic (ROC) curves and concordance statistics were used as measures of discriminative ability, and calibration curves and Brier scores were used to assess the model's predictive ability. MEASUREMENTS: Temporal validation was performed in 1642 patients with a reintubation rate of 4.8%, with the model demonstrating strong discrimination (optimism-corrected c-statistic 0.77) and low predictive error (Brier score 0.044) but poor model precision and recall (Optimal F1 score 0.29). Combined domain and geographic validation were performed in 2041 patients with a reintubation rate of 1.5%. The model displayed solid discriminative ability (optimism-corrected c-statistic = 0.73) and low predictive error (Brier score = 0.0149) but low precision and recall (Optimal F1 score = 0.13). Geographic validation was performed in 2489 patients with a reintubation rate of 1.6%, with the model displaying good discrimination (optimism-corrected c-statistic = 0.71) and predictive error (Brier score = 0.0152) but poor precision and recall (Optimal F1 score = 0.13). MAIN RESULTS: The reintubation model displayed strong discriminative ability and low predictive error within each validation cohort. CONCLUSIONS: Future work is needed to explore how to optimize models before local implementation.

Derivation, External Validation and Clinical Implications of a deep learning approach for intracranial pressure estimation using non-cranial waveform measurements.

Importance: Increased intracranial pressure (ICP) is associated with adverse neurological outcomes, but needs invasive monitoring. Objective: Development and validation of an AI approach for detecting increased ICP (aICP) using only non-invasive extracranial physiological waveform data. Design: Retrospective diagnostic study of AI-assisted detection of increased ICP. We developed an AI model using exclusively extracranial waveforms, externally validated it and assessed associations with clinical outcomes. Setting: MIMIC-III Waveform Database (2000-2013), a database derived from patients admitted to an ICU in an academic Boston hospital, was used for development of the aICP model, and to report association with neurologic outcomes. Data from Mount Sinai Hospital (2020-2022) in New York City was used for external validation. Participants: Patients were included if they were older than 18 years, and were monitored with electrocardiograms, arterial blood pressure, respiratory impedance plethysmography and pulse oximetry. Patients who additionally had intracranial pressure monitoring were used for development (N=157) and external validation (N=56). Patients without intracranial monitors were used for association with outcomes (N=1694). Exposures: Extracranial waveforms including electrocardiogram, arterial blood pressure, plethysmography and SpO2. Main Outcomes and Measures: Intracranial pressure > 15 mmHg. Measures were Area under receiver operating characteristic curves (AUROCs), sensitivity, specificity, and accuracy at threshold of 0.5. We calculated odds ratios and p-values for phenotype association. Results: The AUROC was 0.91 (95% CI, 0.90-0.91) on testing and 0.80 (95% CI, 0.80-0.80) on external validation. aICP had accuracy, sensitivity, and specificity of 73.8% (95% CI, 72.0%-75.6%), 99.5% (95% CI 99.3%-99.6%), and 76.9% (95% CI, 74.0-79.8%) on external validation. A ten-percentile increment was associated with stroke (OR=2.12; 95% CI, 1.27-3.13), brain malignancy (OR=1.68; 95% CI, 1.09-2.60), subdural hemorrhage (OR=1.66; 95% CI, 1.07-2.57), intracerebral hemorrhage (OR=1.18; 95% CI, 1.07-1.32), and procedures like percutaneous brain biopsy (OR=1.58; 95% CI, 1.15-2.18) and craniotomy (OR = 1.43; 95% CI, 1.12-1.84; P < 0.05 for all). Conclusions and Relevance: aICP provides accurate, non-invasive estimation of increased ICP, and is associated with neurological outcomes and neurosurgical procedures in patients without intracranial monitoring.