In paediatric cardiovascular disease, predicting how patients will respond to treatments, which treatments to use, and when to treat can be difficult to define due to small patient numbers and limited outcome data. MD-Paedigree will re-use the models developed in Health-e-Child and Sim-e-Child and extend them to cardiomyopathies.
In paediatric cardiovascular disease, predicting how patients will respond to treatments (operations, catheter interventions, pharmacology), which treatments to use, and when to treat can be difficult to define due to small patient numbers and limited outcome data. When children present with new onset heart failure, there are five possible outcomes: full recovery, dilated cardiomyopathy (DCM) requiring drug therapy, DCM requiring transplantation or mechanical support, another diagnosis (other forms of cardiomyopathy, metabolic disease) or death. At presentation, however, it is very difficult to predict which group any patient will end up in. Data suggests that good systolic function and younger age are good prognostic indicators for survival, but better prognosticators are necessary.
Over the last decade, there has been a huge investment into information technology and computer modelling to build models of the heart that are able to gather any kind of clinical information and produce realistic representations of the cardiovascular system. Modelling of patient bioinformatic data may provide better insight into prognosis of cardiomyopathies, which would help in patient management and in telling families how their child will progress. Would he/she recover completely or would he/she require heart transplant? These models have now reached high levels of reproducibility, opening new avenues for more efficient, safer, and cost‐effective patient management. However, their comprehensive validation is still limited.
MD-Paedigree will re-use the models developed in Health-e-Child and Sim-e-Child and extend them to cardiomyopathies. The objective is to capture the main features of the cardiovascular system, including the heart, arteries and peripheral circulation, to predict cardiomyopathy progression and plan therapies like heart transplant and ventricular assist devices. Investigative data provided by imaging, pressure monitoring, clinical observations and exercise will be used to build these models and to validate them, by comparing model prediction with actual outcome. By merging all scattered information obtained from different diagnostic tools in clinical practice, and obtaining a generative model of heart function in children, our model will provide cardiologists the tools to deliver patients the best possible medical care.