Related Experiment Videos
Measured versus predicted oxygen consumption in children with congenital heart disease
1Department of Anaesthesiology, Hospital for Children and Adolescents, University of Helsinki, Finland.
Insights
Predicting oxygen consumption (VO2) in children with congenital heart disease using regression or AI methods is inaccurate. Direct measurement of VO2 is essential for preoperative evaluation in these patients.
Area of Science:
- Pediatric Cardiology
- Cardiopulmonary Physiology
- Biomedical Engineering
Background:
- Accurate assessment of oxygen consumption (VO2) is crucial for managing children with congenital heart disease (CHD).
- Preoperative VO2 estimation aids in surgical planning and risk stratification.
- Current prediction methods may not adequately account for the complexities of CHD.
Purpose of the Study:
- To compare measured VO2 with values predicted by established regression equations and an artificial neural network (ANN) in children with CHD.
- To evaluate the accuracy (bias and precision) of different VO2 prediction techniques.
Main Methods:
- Retrospective study of 125 children undergoing preoperative cardiac catheterization.
- VO2 measured via indirect calorimetry.
- Predicted VO2 calculated using regression equations (Lindahl, Wessel, Lundell) and an ANN.
- Analysis of bias and precision of predicted versus measured VO2.
Main Results:
- Regression-based predictions showed varying degrees of bias and precision, with Lindahl's equation yielding the highest precision (+/- 42%) but significant bias.
- The ANN achieved better precision (+/- 29%) but still had a notable bias (6%).
- None of the prediction methods accurately estimated preoperative VO2 in this cohort.
Conclusions:
- Both regression and AI-based methods are unreliable for predicting preoperative VO2 in children with CHD.
- Direct measurement of VO2 is indispensable for accurate preoperative assessment in this patient population.
Objective:
To compare measured and predicted oxygen consumption (VO2) in children with congenital heart disease.
Design:
Retrospective study.
Setting:
The cardiac catheterisation laboratory in a university hospital.
Patients:
125 children undergoing preoperative cardiac catheterisation.
Interventions:
VO2 was measured using indirect calorimetry; the predicted values were calculated from regression equations published by Lindahl, Wessel et al, and Lundell et al. Stepwise linear regression and analysis of variance were used to evaluate the influence of age, sex, weight, height, cardiac malformation, and heart failure on the bias and precision of predicted VO2. An artificial neural network was trained and used to produce an estimate of VO2 employing the same variables. The various estimates for VO2 were evaluated by calculating their bias and precision values.
Results:
Lindahl's equation produced the highest precision (+/- 42%) of the regression based estimates. The corresponding average bias of the predicted VO2 was 3% (range -66% to 43%). When VO2 was predicted according to regression equations by Wessel and Lundell, the bias and precision were 0% and +/- 44%, and -16% and +/- 51%, respectively. The neural network predicted VO2 from variables included in the regression equations with a bias of 6% and precision +/- 29%; addition of further variables failed to improve this estimate.
Conclusions:
Both regression based and artificial intelligence based techniques were inaccurate for predicting preoperative VO2 in patients with congenital heart disease. Measurement of VO2 is necessary in the preoperative evaluation of these patients.