Cardiac intensivists frequently assess patient readiness to wean off mechanical ventilation with an extubation readiness trial despite it being no more effective than clinician judgement alone. We evaluated the utility of high-frequency physiologic data and machine learning for improving the prediction of extubation failure in children with cardiovascular disease.

This was a retrospective analysis of clinical registry data and streamed physiologic extubation readiness trial data from one paediatric cardiac ICU (12/2016-3/2018). We analysed patients’ final extubation readiness trial. Machine learning methods (classification and regression tree, Boosting, Random Forest) were performed using clinical/demographic data, physiologic data, and both datasets. Extubation failure was defined as reintubation within 48 hrs. Classifier performance was assessed on prediction accuracy and area under the receiver operating characteristic curve.

Of 178 episodes, 11.2% (N = 20) failed extubation. Using clinical/demographic data, our machine learning methods identified variables such as age, weight, height, and ventilation duration as being important in predicting extubation failure. Best classifier performance with this data was Boosting (prediction accuracy: 0.88; area under the receiver operating characteristic curve: 0.74). Using physiologic data, our machine learning methods found oxygen saturation extremes and descriptors of dynamic compliance, central venous pressure, and heart/respiratory rate to be of importance. The best classifier in this setting was Random Forest (prediction accuracy: 0.89; area under the receiver operating characteristic curve: 0.75). Combining both datasets produced classifiers highlighting the importance of physiologic variables in determining extubation failure, though predictive performance was not improved.

Physiologic variables not routinely scrutinised during extubation readiness trials were identified as potential extubation failure predictors. Larger analyses are necessary to investigate whether these markers can improve clinical decision-making.

Understanding the impact of the COVID-19 pandemic on paediatric non-COVID-19-related care, as well as patient and caregiver concerns and stressors, is critical for informing healthcare delivery. It was hypothesised that high care disruptions and psychological stress would be observed among paediatric and adult CHD patients in the early phase of the pandemic.

A cross-sectional, international, electronic survey study was completed. Eligible participants included parents of children with acquired or CHD, adults with CHD, or caregivers of adults with CHD.

A total of 1220 participants from 25 countries completed the survey from 16 April to 4 May, 2020. Cardiac care disruption was significant with 38% reporting delays in pre-pandemic scheduled cardiac surgeries and 46% experiencing postponed cardiac clinic visits. The majority of respondents (75%) endorsed moderate to high concern about the patient with heart disease becoming ill from COVID-19. Worry about returning for in-person care was significantly greater than worry of harm to patient due to postponed care. Clinically significant psychological stress was high across the sample including children (50%), adults with CHD (42%), and caregivers (42%).

The early phase of the COVID-19 pandemic contributed to considerable disruptions in cardiac care for patients with paediatric and adult CHD. COVID-19-related fears are notable with potential to impact willingness to return to in-person care. Psychological stress is also very high necessitating intervention. Further study of the impact of delays in care on clinical outcomes is warranted.

Registry-based trials have emerged as a potentially cost-saving study methodology. Early estimates of cost savings, however, conflated the benefits associated with registry utilisation and those associated with other aspects of pragmatic trial designs, which might not all be as broadly applicable. In this study, we sought to build a practical tool that investigators could use across disciplines to estimate the ranges of potential cost differences associated with implementing registry-based trials versus standard clinical trials.

We built simulation Markov models to compare unique costs associated with data acquisition, cleaning, and linkage under a registry-based trial design versus a standard clinical trial. We conducted one-way, two-way, and probabilistic sensitivity analyses, varying study characteristics over broad ranges, to determine thresholds at which investigators might optimally select each trial design.

Registry-based trials were more cost effective than standard clinical trials 98.6% of the time. Data-related cost savings ranged from $4300 to $600,000 with variation in study characteristics. Cost differences were most reactive to the number of patients in a study, the number of data elements per patient available in a registry, and the speed with which research coordinators could manually abstract data. Registry incorporation resulted in cost savings when as few as 3768 independent data elements were available and when manual data abstraction took as little as 3.4 seconds per data field.

Registries offer important resources for investigators. When available, their broad incorporation may help the scientific community reduce the costs of clinical investigation. We offer here a practical tool for investigators to assess potential costs savings.

To evaluate the association between novel pre- and post-operative biomarker levels and 30-day unplanned readmission or mortality after paediatric congenital heart surgery.

Children aged 18 years or younger undergoing congenital heart surgery (n = 162) at Johns Hopkins Hospital from 2010 to 2014 were enrolled in the prospective cohort. Collected novel pre- and post-operative biomarkers include soluble suppression of tumorgenicity 2, galectin-3, N-terminal prohormone of brain natriuretic peptide, and glial fibrillary acidic protein. A model based on clinical variables from the Society of Thoracic Surgery database was developed and evaluated against two augmented models.

Unplanned readmission or mortality within 30 days of cardiac surgery occurred among 21 (13%) children. The clinical model augmented with pre-operative biomarkers demonstrated a statistically significant improvement over the clinical model alone with a receiver-operating characteristics curve of 0.754 (95% confidence interval: 0.65–0.86) compared to 0.617 (95% confidence interval: 0.47–0.76; p-value: 0.012). The clinical model augmented with pre- and post-operative biomarkers demonstrated a significant improvement over the clinical model alone, with a receiver-operating characteristics curve of 0.802 (95% confidence interval: 0.72–0.89; p-value: 0.003).

Novel biomarkers add significant predictive value when assessing the likelihood of unplanned readmission or mortality after paediatric congenital heart surgery. Further exploration of the utility of these novel biomarkers during the pre- or post-operative period to identify early risk of mortality or readmission will aid in determining the clinical utility and application of these biomarkers into routine risk assessment.

Using existing data from clinical registries to support clinical trials and other prospective studies has the potential to improve research efficiency. However, little has been reported about staff experiences and lessons learned from implementation of this method in pediatric cardiology.

We describe the process of using existing registry data in the Pediatric Heart Network Residual Lesion Score Study, report stakeholders’ perspectives, and provide recommendations to guide future studies using this methodology.

The Residual Lesion Score Study, a 17-site prospective, observational study, piloted the use of existing local surgical registry data (collected for submission to the Society of Thoracic Surgeons-Congenital Heart Surgery Database) to supplement manual data collection. A survey regarding processes and perceptions was administered to study site and data coordinating center staff.

Survey response rate was 98% (54/55). Overall, 57% perceived that using registry data saved research staff time in the current study, and 74% perceived that it would save time in future studies; 55% noted significant upfront time in developing a methodology for extracting registry data. Survey recommendations included simplifying data extraction processes and tailoring to the needs of the study, understanding registry characteristics to maximise data quality and security, and involving all stakeholders in design and implementation processes.

Use of existing registry data was perceived to save time and promote efficiency. Consideration must be given to the upfront investment of time and resources needed. Ongoing efforts focussed on automating and centralising data management may aid in further optimising this methodology for future studies.

The Single Ventricle Reconstruction trial randomised neonates with hypoplastic left heart syndrome to a systemic-to-pulmonary-artery shunt strategy. Patients received care according to usual institutional practice. We analysed practice variation at the Stage II surgery to attempt to identify areas for decreased variation and process control improvement.

Prospectively collected data were available in the Single Ventricle Reconstruction public-use database. Practice variation across 14 centres was described for 397 patients who underwent Stage II surgery. Data are centre-level specific and reported as interquartile ranges across all centres, unless otherwise specified.

Preoperative Stage II median age and weight across centres were 5.4 months (interquartile range 4.9–5.7) and 5.7 kg (5.5–6.1), with 70% performed electively. Most patients had pre-Stage-II cardiac catheterisation (98.5–100%). Digoxin was used by 11/14 centres in 25% of patients (23–31%), and 81% had some oral feeds (68–84%). The majority of the centres (86%) performed a bidirectional Glenn versus hemi-Fontan. Median cardiopulmonary bypass time was 96 minutes (75–113). In aggregate, 26% of patients had deep hypothermic circulatory arrest >10 minutes. In 13/14 centres using deep hypothermic circulatory arrest, 12.5% of patients exceeded 10 minutes (8–32%). Seven centres extubated 5% of patients (2–40) in the operating room. Postoperatively, ICU length of stay was 4.8 days (4.0–5.3) and total length of stay was 7.5 days (6–10).

In the Single Ventricle Reconstruction Trial, practice varied widely among centres for nearly all perioperative factors surrounding Stage II. Further analysis may facilitate establishing best practices by identifying the impact of practice variation.

Although interstage mortality for infants with hypoplastic left heart syndrome has declined within the National Pediatric Cardiology Quality Improvement Collaborative, variation across centres persists. It remains unclear whether centres with lower interstage mortality have lower-risk patients or whether differences in care may explain this variation. We examined previously established risk factors across National Pediatric Cardiology Quality Improvement Collaborative centres with lower and higher interstage mortality rates.

Lower-mortality centres were defined as those with >25 consecutive interstage survivors. Higher-mortality centres were defined as those with cumulative interstage mortality rates >10%, which is a collaborative historic baseline rate. Baseline risk factors and perioperative characteristics were compared.

Seven lower-mortality centres were identified (n=331 patients) and had an interstage mortality rate of 2.7%, as compared with 13.3% in the four higher-mortality centres (n=173 patients, p<0.0001). Of all baseline risk factors examined, the only factor that differed between the lower- and higher-mortality centres was postnatal diagnosis (18.4 versus 31.8%, p=0.001). In multivariable analysis, there remained a significant mortality difference between the two groups of centres after adjusting for this variable: adjusted mortality rate was 2.8% in lower-mortality centres compared with 12.6% in higher-mortality centres, p=0.003. Secondary analyses identified multiple differences between groups in perioperative practices and other variables.

Variation in interstage mortality rates between these two groups of centres does not appear to be explained by differences in baseline risk factors. Further study is necessary to evaluate variation in care practices to identify targets for improvement efforts.

The Pediatric Heart Network designed a career development award to train the next generation of clinician scientists in paediatric-cardiology-related research, a historically underfunded area. We sought to identify the strengths/weaknesses of the programme and describe the scholars’ academic achievements and the network’s return on investment.

Survey questions designed to evaluate the programme were sent to applicants – 13 funded and 19 unfunded applicants – and 20 mentors and/or principal investigators. Response distributions were calculated. χ2 tests of association assessed differences in ratings of the application/selection processes among funded scholars, unfunded applicants, and mentors/principal investigators. Scholars reported post-funding academic achievements.

Survey response rates were 88% for applicants and 100% for mentor/principal investigators. Clarity and fairness of the review were rated as “clear/fair” or “very clear/very fair” by 98% of respondents, but the responses varied among funded scholars, unfunded applicants, and mentors/principal investigators (clarity χ2=10.85, p=0.03; fairness χ2=16.97, p=0.002). Nearly half of the unfunded applicants rated feedback as “not useful” (47%). “Expanding their collaborative network” and “increasing publication potential” were the highest-rated benefits for scholars. Mentors/principal investigators found the programme “very” valuable for the scholars (100%) and the network (75%). The 13 scholars were first/senior authors for 97 abstracts and 109 manuscripts, served on 22 Pediatric Heart Network committees, and were awarded $9,673,660 in subsequent extramural funding for a return of ~$10 for every scholar dollar spent.

Overall, patient satisfaction with the Scholar Award was high and scholars met many academic markers of success. Despite this, programme challenges were identified and improvement strategies were developed.

Neurodevelopmental impairment is increasingly recognised as a potentially disabling outcome of CHD and formal evaluation is recommended for high-risk patients. However, data are lacking regarding the proportion of eligible children who actually receive neurodevelopmental evaluation, and barriers to follow-up are unclear. We examined the prevalence and risk factors associated with failure to attend neurodevelopmental follow-up clinic after infant cardiac surgery.

Survivors of infant (<1 year) cardiac surgery at our institution (4/2011-3/2014) were included. Socio-demographic and clinical characteristics were evaluated in neurodevelopmental clinic attendees and non-attendees in univariate and multivariable analyses.

A total of 552 patients were included; median age at surgery was 2.4 months, 15% were premature, and 80% had moderate–severe CHD. Only 17% returned for neurodevelopmental evaluation, with a median age of 12.4 months. In univariate analysis, non-attendees were older at surgery, had lower surgical complexity, fewer non-cardiac anomalies, shorter hospital stay, and lived farther from the surgical center. Non-attendee families had lower income, and fewer were college graduates or had private insurance. In multivariable analysis, lack of private insurance remained independently associated with non-attendance (adjusted odds ratio 1.85, p=0.01), with a trend towards significance for distance from surgical center (adjusted odds ratio 2.86, p=0.054 for ⩾200 miles).

The majority of infants with CHD at high risk for neurodevelopmental dysfunction evaluated in this study are not receiving important neurodevelopmental evaluation. Efforts to remove financial/insurance barriers, increase access to neurodevelopmental clinics, and better delineate other barriers to receipt of neurodevelopmental evaluation are needed.

Chylothorax after paediatric cardiac surgery incurs significant morbidity; however, a detailed understanding that does not rely on single-centre or administrative data is lacking. We described the present clinical epidemiology of postoperative chylothorax and evaluated variation in rates among centres with a multicentre cohort of patients treated in cardiac ICU.

This was a retrospective cohort study using prospectively collected clinical data from the Pediatric Cardiac Critical Care Consortium registry. All postoperative paediatric cardiac surgical patients admitted from October, 2013 to September, 2015 were included. Risk factors for chylothorax and association with outcomes were evaluated using multivariable logistic or linear regression models, as appropriate, accounting for within-centre clustering using generalised estimating equations.

A total of 4864 surgical hospitalisations from 15 centres were included. Chylothorax occurred in 3.8% (n=185) of hospitalisations. Case-mix-adjusted chylothorax rates varied from 1.5 to 7.6% and were not associated with centre volume. Independent risk factors for chylothorax included age <1 year, non-Caucasian race, single-ventricle physiology, extracardiac anomalies, longer cardiopulmonary bypass time, and thrombosis associated with an upper-extremity central venous line (all p<0.05). Chylothorax was associated with significantly longer duration of postoperative mechanical ventilation, cardiac ICU and hospital length of stay, and higher in-hospital mortality (all p<0.001).

Chylothorax after cardiac surgery in children is associated with significant morbidity and mortality. A five-fold variation in chylothorax rates was observed across centres. Future investigations should identify centres most adept at preventing and managing chylothorax and disseminate best practices.

With improvements in early survival following congenital heart surgery, it has become increasingly important to understand longer-term outcomes; however, routine collection of these data is challenging and remains very limited. We describe the development and initial results of a collaborative programme incorporating standardised longitudinal follow-up into usual care at the Children’s Hospital of Philadelphia (CHOP) and University of Michigan (UM).

We included children undergoing benchmark operations of the Society of Thoracic Surgeons. Considerations regarding personnel, patient/parent engagement, funding, regulatory issues, and annual data collection are described, and initial follow-up rates are reported.

The present analysis included 1737 eligible patients undergoing surgery at CHOP from January 2007 to December 2014 and 887 UM patients from January 2010 to December 2014. Overall, follow-up data, of any type, were obtained from 90.8% of patients at CHOP (median follow-up 4.3 years, 92.2% survival) and 98.3% at UM (median follow-up 2.8 years, 92.7% survival), with similar rates across operations and institutions. Most patients lost to follow-up at CHOP had undergone surgery before 2010. Standardised questionnaires assessing burden of disease/quality of life were completed by 80.2% (CHOP) and 78.4% (UM) via phone follow-up. In subsequent pilot testing of an automated e-mail system, 53.4% of eligible patients completed the follow-up questionnaire through this system.

Standardised follow-up data can be obtained on the majority of children undergoing benchmark operations. Ongoing efforts to support automated electronic systems and integration with registry data may reduce resource needs, facilitate expansion across centres, and support multi-centre efforts to understand and improve long-term outcomes in this population.

National organisations in several countries have recently released more restrictive guidelines for infective endocarditis prophylaxis, including the American Heart Association 2007 guidelines. Initial studies demonstrated no change in infective endocarditis rates over time; however, a recent United Kingdom study suggested an increase; current paediatric trends are unknown.

Children (<18 years) hospitalised with infective endocarditis at 29 centres participating in the Pediatric Health Information Systems Database from 2003 to 2014 were eligible for inclusion. Our primary analysis focussed on infective endocarditis most directly related to the change in guidelines and included community-acquired cases in those >5 years of age. Interrupted time series analysis was used to evaluate rates over time indexed to total hospitalisations.

A total of 841 cases were identified. The median age was 13 years (interquartile range 9–15 years). In the pre-guideline period, there was a slight increase in the rate of infective endocarditis by 0.13 cases/10,000 hospitalisations per semi-annual period. In the post-guideline period, the rate of infective endocarditis increased by 0.12 cases/10,000 hospitalisations per semi-annual period. There was no significant difference in the rate of change in the pre- versus post-guidelines period (p=0.895). Secondary analyses in children >5 years of age with CHD and in children hospitalised with any type of infective endocarditis at any age revealed similar results.

We found no significant change in infective endocarditis hospitalisation rates associated with revised prophylaxis guidelines over 11 years across 29 United States children’s hospitals.

Patients undergoing the Norwood operation consume considerable healthcare resources; however, detailed information regarding factors impacting hospitalisation costs is lacking. We evaluated the association of postoperative complications with hospital costs.

In the present study, we utilised a unique data set consisting of prospectively collected clinical data from the Pediatric Heart Network Single Ventricle Reconstruction trial linked at the patient level with cost data for 10 hospitals participating in the Children’s Hospital Association Case Mix database during the trial period. The relationship between complications and cost was modelled using linear regression, accounting for the skewed distribution of cost, adjusting for within-centre clustering and baseline patient characteristics.

A total of 334 eligible Norwood records (97.5%) were matched between data sets. Overall, 82% suffered from at least one complication (median 2; with a range from 0 to 33). Those with complications had longer postoperative length of stay (25 versus 12 days, p<0.001), more total ventilator days (7 versus 5 days, p<0.001), and higher in-hospital mortality (17.6 versus 3.4%, p<0.006). Mean adjusted hospital cost in those with a complication was $190,689 (95% CI $111,344–$326,577) versus $120,584 (95% CI $69,246–$209,983) in those without complications (p=0.002). Costs increased with the number of complications (1–2 complications=$132,800 versus 3–4 complications=$182,353 versus ⩾5 complications=$309,372 [p<0.001]).

This merged data set of clinical trial and cost data demonstrated that postoperative complications are common following the Norwood operation and are associated with worse clinical outcomes and higher costs. Efforts to reduce complications in this population may lead to improved outcomes and cost savings.