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.

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.

A retrospective study of the impact of supraglottic airway devices on the appearance of neck masses in children undergoing serial magnetic resonance imaging.

BACKGROUND: General anesthesia and sedation are used routinely for magnetic resonance imaging (MRI) studies in children to optimize image quality. Airway devices such as supraglottic airways (SGAs) can alter the appearance of cervical soft tissue regions on an MRI and increase the risk of misdiagnosis. This phenomenon has not been well described in vivo. AIMS: We conducted this retrospective study to determine how often SGAs affected the appearance of neck masses in children who received multiple anesthetics for MRIs with and without an SGA. METHODS: We retrieved data on children 17 years old and younger who had multiple MRIs between January 2005 and January 2015. Inclusion criteria were patients with neck masses who had a SGA for at least one MRI and either a natural airway or endotracheal tube (ETT) for another MRI. We reviewed MRI images and imaging reports to determine if SGAs affected the appearance of neck masses. RESULTS: Twelve of the 921 patients who received anesthesia for neck MRIs during the study period met the inclusion criteria. SGAs affected the appearance of the neck mass in 11 of the 12 patients. CONCLUSIONS: Supraglottic airways can significantly alter the appearance of neck masses in children undergoing MRIs and affect radiologists' ability to assess those masses. Communication with the radiologist prior to the induction of anesthesia is crucial when using supraglottic devices in this patient population. It may be more prudent to use a different airway device and/or anesthetic technique when MRIs of these neck masses are undertaken.

Extubation in the operating room after pediatric liver transplant: A retrospective cohort study.

BACKGROUND: Early extubation immediately following liver transplantation is increasingly common in adult practice. Some pediatric institutions have begun to adopt this strategy. Careful patient selection is essential in minimizing risk. METHODS: This retrospective cohort study evaluated infants and children who underwent liver transplantation between July 2011 and December 2014. Our primary objective was to determine early extubation rate. Secondary objectives were to identify clinical factors associated with successful early extubation compared with delayed extubation and to examine significant postoperative complications, intensive care unit length of stay, and hospital length of stay. RESULTS: The early extubation rate was 57.8% (37/64, confidence interval [CI] 44.8%-70.1%) over this 3.5-year period, increasing from 42% in 2012 to 58% by the end of 2014. The patients in the early extubation group were more likely to be older than the delayed extubation group (mean [SD], 7 [5.3] years vs 3.5 [5.5] years, difference between the mean [95% CI], 3.5 [0.8, 6.2] years); were to have come from home on the day of surgery (78.4% vs 25.9%); and were less likely to be listed as United Network for Organ Sharing status 1A (2.7% vs 25.9%). The early extubation group received less packed red blood cell volume (mean [SD], 9 [13.2] mL/kg vs 40.6 [48.5] mL/kg, difference between the mean [95% CI], 31.6 [95% CI 14.9, 48.3] mL/kg) and fresh-frozen plasma (mean 2.7 [SD 9.5] vs 13.3 [SD15.1], difference between the mean [95% CI], 10.5 [4.4,16.7] mL/kg). None of the patients in the early extubation group required reintubation in the first 24 hours following transplant and none experienced hepatic artery thrombosis. The early extubation group had a shorter average postoperative PICU stay (mean 3.8 [SD 2.1] days vs 17.6 [SD 31.3] days, difference between the mean [95% CI], 9.5 [4.3, 14.7] days) and a shorter postoperative hospital stay overall (mean 10.7 [SD 4.3] days vs 29.7 [SD 43.1] days, difference between the mean [95% CI], 19.1 [8.6, 29.6] days). CONCLUSION: More than half of our pediatric liver transplant patients were successfully extubated in the operating room immediately following surgery. We believe early extubation to be safe when employed in selected subpopulations of pediatric patients undergoing liver transplantation.

Neural Network Classifier for Automatic Detection of Invasive Versus Noninvasive Airway Management Technique Based on Respiratory Monitoring Parameters in a Pediatric Anesthesia.

Children undergoing general anesthesia require airway monitoring by an anesthesia provider. The airway may be supported with noninvasive devices such as face mask or invasive devices such as a laryngeal mask airway or an endotracheal tube. The physiologic data stored provides an opportunity to apply machine learning algorithms distinguish between these modes based on pattern recognition. We retrieved three data sets from patients receiving general anesthesia in 2015 with either mask, laryngeal mask airway or endotracheal tube. Patients underwent myringotomy, tonsillectomy, adenoidectomy or inguinal hernia repair procedures. We retrieved measurements for end-tidal carbon dioxide, tidal volume, and peak inspiratory pressure and calculated statistical features for each data element per patient. We applied machine learning algorithms (decision tree, support vector machine, and neural network) to classify patients into noninvasive or invasive airway device support. We identified 300 patients per group (mask, laryngeal mask airway, and endotracheal tube) for a total of 900 patients. The neural network classifier performed better than the boosted trees and support vector machine classifiers based on the test data sets. The sensitivity, specificity, and accuracy for neural network classification are 97.5%, 96.3%, and 95.8%. In contrast, the sensitivity, specificity, and accuracy of support vector machine are 89.1%, 92.3%, and 88.3% and with the boosted tree classifier they are 93.8%, 92.1%, and 91.4%. We describe a method to automatically distinguish between noninvasive and invasive airway device support in a pediatric surgical setting based on respiratory monitoring parameters. The results show that the neural network classifier algorithm can accurately classify noninvasive and invasive airway device support.

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.

Increasing cumulative exposure to volatile anesthetic agents is associated with poorer neurodevelopmental outcomes in children with hypoplastic left heart syndrome.

OBJECTIVES: Despite improved survival in children with hypoplastic left heart syndrome (HLHS), significant concern persists regarding their neurodevelopmental (ND) outcomes. Previous studies have identified patient factors, such as prematurity and genetic syndromes, to be associated with worse ND outcomes. However, no consistent relationships have been identified among modifiable management factors, including cardiopulmonary bypass strategies, and ND outcomes after cardiac surgery in infancy. Studies in immature animals, including primates, have demonstrated neurodegeneration and apoptosis in the brain after certain levels and extended durations of anesthetic exposure. Retrospective human studies have also suggested relationships between adverse ND effects and anesthetic exposure. METHODS: Cumulative minimum alveolar concentration hours (MAC-hrs) of exposure to volatile anesthetic agents (VAA) (desflurane, halothane, isoflurane, and sevoflurane) were collected from an anesthetic database and medical record review for 96 patients with HLHS or variants. ND testing was performed between ages 4 and 5 years, including full-scale IQ, verbal IQ, performance IQ, and processing speed. Four generalized linear modes were hypothesized a priori and tested using a Gaussian (normal) distribution with an identity link. RESULTS: Cumulative VAA exposure ranged from 0 to 35.3 MAC-hrs (median 7.5 hours). Using specified covariates identified previously as significant predictors of ND outcomes, statistically significant relationships were identified between total MAC-hrs exposure and worse full-scale IQ and verbal IQ scores (P's < .05) alone and after adjusting for relevant covariates. CONCLUSIONS: Increased cumulative MAC-hrs exposure to VAA is associated with worse ND outcomes in certain domains in children with HLHS and variants.

Anesthetic management of the first pediatric bilateral hand transplant.

PURPOSE: The purpose of this case report is to describe the anesthetic and case management of the first vascularized composite allograft pediatric bilateral hand transplant. CLINICAL DETAILS: Our patient was an eight-year-old male with a medical history of Staphylococcus aureus sepsis at one year of age that resulted in end-stage renal disease as well as bilateral upper and lower extremity amputations. After referral for bilateral hand transplantation, the transplantation team, with expertise in all aspects of perioperative care (surgery, anesthesiology, nephrology, renal transplantation, pediatric intensive care, and therapeutic pharmacy), was consulted to help develop anesthetic and other perioperative protocols for surgery. Prior to activation of the transplantation team, the lead surgeon evaluated potential donors by comparing a three-dimensional printed model of the recipient's forearm with the donor's upper extremities to ensure an adequate match. The anesthesia team inserted bilateral ultrasound-guided infraclavicular catheters to provide a sympathetic block to facilitate blood flow to the upper extremities and to provide both intraoperative and postoperative pain control. The patient remained in the operating room for 13 hr 37 min for a surgical time of ten hours 39 min. He remained in the hospital for 34 days after the procedure and was then transferred to an inpatient rehabilitation facility for a further 15 days. The patient is currently doing well in a postoperative rehabilitation program. He has demonstrated motor power to the hands using the forearm muscles but is not expected to reach his maximum sensory function for at least one to two years. CONCLUSION: This report describes the anesthetic management of the first pediatric bilateral hand transplant. This procedure required considerable preoperative planning and communication between various teams to ensure all resources needed to deliver the care for this complex and novel transplant surgery were readily available.

A Technical Evaluation of Wireless Connectivity from Patient Monitors to an Anesthesia Information Management System During Intensive Care Unit Surgery.

Surgical procedures performed at the bedside in the neonatal intensive care unit (NICU) at The Children's Hospital of Philadelphia were documented using paper anesthesia records in contrast to the operating rooms, where an anesthesia information management system (AIMS) was used for all cases. This was largely because of logistical problems related to connecting cables between the bedside monitors and our portable AIMS workstations. We implemented an AIMS for documentation in the NICU using wireless adapters to transmit data from bedside monitoring equipment to a portable AIMS workstation. Testing of the wireless AIMS during simulation in the presence of an electrosurgical generator showed no evidence of interference with data transmission. Thirty NICU surgical procedures were documented via the wireless AIMS. Two wireless cases exhibited brief periods of data loss; one case had an extended data gap because of adapter power failure. In comparison, in a control group of 30 surgical cases in which wired connections were used, there were no data gaps. The wireless AIMS provided a simple, unobtrusive, portable alternative to paper records for documenting anesthesia records during NICU bedside procedures.

Assessment and Planning for a Pediatric Bilateral Hand Transplant Using 3-Dimensional Modeling: Case Report.

Children are not typically considered for hand transplantation for various reasons, including the difficulty of finding an appropriate donor. Matching donor-recipient hands and forearms based on size is critically important. If the donor's hands are too large, the recipient may not be able to move the fingers effectively. Conversely, if the donor's hands are too small, the appearance may not be appropriate. We present an 8-year-old child evaluated for a bilateral hand transplant following bilateral amputation. The recipient forearms and model hands were modeled from computed tomography imaging studies and replicated as anatomic models with a 3-dimensional printer. We modified the scale of the printed hand to produce 3 proportions, 80%, 100% and 120%. The transplant team used the anatomical models during evaluation of a donor for appropriate match based on size. The donor's hand size matched the 100%-scale anatomical model hand and the transplant team was activated. In addition to assisting in appropriate donor selection by the transplant team, the 100%-scale anatomical model hand was used to create molds for prosthetic hands for the donor.

Airway management complications in children with difficult tracheal intubation from the Pediatric Difficult Intubation (PeDI) registry: a prospective cohort analysis.

BACKGROUND: Despite the established vulnerability of children during airway management, remarkably little is known about complications in children with difficult tracheal intubation. To address this concern, we developed a multicentre registry (Pediatric Difficult Intubation [PeDI]) to characterise risk factors for difficult tracheal intubation, establish the success rates of various tracheal intubation techniques, catalogue the complications of children with difficult tracheal intubation, and establish the effect of more than two tracheal intubation attempts on complications. METHODS: The PeDI registry consists of prospectively collected tracheal intubation data from 13 children's hospitals in the USA. We established standard data collection methods before implementing the secure web-based registry. After establishing standard definitions, we collected and analysed patient, clinician, and practice data and tracheal intubation outcomes. We categorised complications as severe or non-severe. FINDINGS: Between August, 2012, and January, 2015, 1018 difficult paediatric tracheal intubation encounters were done. The most frequently attempted first tracheal intubation techniques were direct laryngoscopy (n=461, 46%), fibre-optic bronchoscopy (n=284 [28%]), and indirect video laryngoscopy (n=183 [18%]) with first attempt success rates of 16 (3%) of 461 with direct laryngoscopy, 153 (54%) of 284 with fibre-optic bronchoscopy, and 101 (55%) of 183 with indirect video laryngoscopy. Tracheal intubation failed in 19 (2%) of cases. 204 (20%) children had at least one complication; 30 (3%) of these were severe and 192 (19%) were non-severe. The most common severe complication was cardiac arrest, which occurred in 15 (2%) patients. The occurrence of complications was associated with more than two tracheal intubation attempts, a weight of less than 10 kg, short thyromental distance, and three direct laryngoscopy attempts before an indirect technique. Temporary hypoxaemia was the most frequent non-severe complication. INTERPRETATION: More than two direct laryngoscopy attempts in children with difficult tracheal intubation are associated with a high failure rate and an increased incidence of severe complications. These results suggest that limiting the number of direct laryngoscopy attempts and quickly transitioning to an indirect technique when direct laryngoscopy fails would enhance patient safety. FUNDING: None.

Transient and reproducible loss of motor-evoked potential signals after intravenous levetiracetam in a child undergoing craniotomy for resection of astrocytoma.

Transcranial electrical motor-evoked potential (tceMEP) monitoring is used in complex intracranial and spinal surgeries to detect and prevent neurological injury. We present a case of transient, reproducible loss of tceMEPs after an infusion of levetiracetam during craniotomy and tumor resection in a child. Cessation of the infusion resulted in restoration of baseline tceMEPs. When the infusion was resumed at the end of the procedure, a similar decrease in tceMEPs was seen as before, after the infusion was stopped. The surgery and postoperative course proceeded without incident, and the patient experienced a full recovery.

TAOK3, a novel genome-wide association study locus associated with morphine requirement and postoperative pain in a retrospective pediatric day surgery population.

Candidate gene studies have revealed limited genetic bases for opioid analgesic response variability. Genome-wide association studies facilitate impartial queries of common genetic variants, allowing identification of novel genetic contributions to drug effect. Illumina (Illumina Inc, San Diego, CA, USA) single nucleotide polymorphism (SNP) arrays were used to investigate SNP associations with total morphine requirement as a quantitative trait locus and with postoperative pain in a retrospective population of opioid-naïve children ages 4-18years who had undergone day surgery tonsillectomy and adenoidectomy. In an independent replication cohort, significant genome-wide association studies-identified SNPs were assayed using TaqMan probes. Among 617 comprehensively phenotyped children, the 277 subjects of European Caucasian (EC) ancestry demonstrated nominal association between morphine dose and a series of novel SNPs (top rs795484, P=1.01 × 10(-6) and rs1277441, P=2.77 × 10(-6)) at the TAOK3 locus. Age, body mass index, and physical status were included covariates. Morphine requirement averaged 132.4 µg/kg (SD 40.9). Each minor allele at rs795484 (guanine [G]>adenine [A]) contributed +17.6 µg/kg (95% confidence interval [CI] 10.7-24.4) to dose. Effect direction and magnitude were replicated in an independent cohort of 75 EC children (P<0.05). No association with morphine dose was detected in African Americans (AA) (n=241). Postoperative pain scores ≥ 7/10 were associated with rs795484 (G>A) in the EC cohort (odds ratio 2.35, 95% CI 1.56-3.52, P<0.00005) and this association replicated in AA children (odds ratio 1.76, 95% CI 1.14-2.71, P<0.01). Variants in TAOK3 encoding the serine/threonine-protein kinase, TAO3, are associated with increased morphine requirement in children of EC ancestry and with increased acute postoperative pain in both EC and AA subjects.

Visual analytical tool for evaluation of 10-year perioperative transfusion practice at a children's hospital.

Children are a vulnerable population in the operating room, and are particularly at risk of complications from unanticipated hemorrhage. The decision to prepare blood products prior to surgery varies depending on the personal experience of the clinician caring for the patient. We present the first application of a data visualization technique to study large datasets in the context of blood product transfusions at a tertiary pediatric hospital. The visual analytical interface allows real-time interaction with datasets from 230 000 procedure records. Clinicians can use the visual analytical interface to analyze blood product usage based on procedure- and patient-specific factors, and then use that information to guide policies for ordering blood products.

The efficacy of the Storz Miller 1 video laryngoscope in a simulated infant difficult intubation.

BACKGROUND: Several studies have shown video laryngoscopy to be a useful technique in the management of patients in whom glottic exposure by direct laryngoscopy is difficult. We conducted this study as a preliminary investigation comparing the Storz DCI Miller 1 video laryngoscope (VL, Karl Storz GmbH, Tuttlingen, Germany) and direct laryngoscopy with a Miller 1 laryngoscope (DL) in an infant manikin model simulating difficult direct laryngoscopy. We hypothesized that compared with DL, VL would provide a better glottic view but would be associated with a longer time to intubation because of the different skill set required when using video intubation. METHODS: A Laerdal infant airway management training manikin (Laerdal Medical, Wappingers Falls, NY) was adapted using cloth tape to limit cervical spine mobility. Thirty-two attending pediatric anesthesiologists attempted tracheal intubation of the infant manikin using VL and DL in randomized order. The best laryngeal view with each laryngoscope and time to intubation were documented. RESULTS: There was a significant difference in the distributions of laryngoscopy grades between VL and DL (P < 0.001), with the VL giving a better laryngeal view. Forty percent of anesthesiologists reported a Grade 3 or 4 view with DL; all of which were converted to Grades 1 and 2 with VL. The median grade with interquartile range was two (2-3) for DL and one (1-2) for VL (P < 0.001). Seventy-eight percent of participants reported an improvement of at least one grade in laryngeal view with VL compared with DL. There were two failed intubations using DL and none using VL. Time to intubation was similar between the two techniques. CONCLUSIONS: The Storz Miller 1 VL blade improved glottic exposure in a simulated difficult laryngoscopy compared with direct laryngoscopy with a standard Miller 1 blade without increasing the time to intubation.

Emergent excision of a prenatally diagnosed sacrococcygeal teratoma.

Prenatally diagnosed sacrococcygeal teratomas (SCT) have higher mortality rates than those diagnosed in the neonatal period. Natural history of SCT varies, and management depends on pathophysiology. Treatment may be minimally invasive or require open surgery. Intervention may take place in the prenatal period, or it may occur within minutes to days after birth. Optimal care requires close follow up and communication between members of a multidisciplinary team. We present a case of prenatally diagnosed SCT and address the evaluation, anesthetic considerations, and mechanisms needed to care for this high risk population.