It has been long appreciated that pulmonary arterial hypertension (PAH) in children is a heterogeneous disease with considerable morbidity and high mortality. While the introduction of specific disease targeting therapies, including prostanoids, endothelin receptor antagonists and phosphodiesterase-5-inhibitors, has improved symptoms and prognosis of patients with PAH, this remains a progressive and often fatal disease.1 Given the substantial costs of oral PAH therapies and—more importantly—the invasiveness and high maintenance of intravenous or subcutaneous prostanoid therapy, careful selection of patients likely to benefit from treatment is highly desirable. Despite considerable research efforts, however, a universally accepted risk stratification algorithm for children with PAH is still unavailable.

Experienced clinicians have always appreciated that no risk marker should be seen in isolation and that the degree of pulmonary hypertension (ie, the absolute value of mean or systolic pulmonary arterial (PA) pressure) per se is insufficient to predict the outcome of patients. This is because even highly elevated PA pressures can be well tolerated as long as the RV is able to cope with increased ventricular afterload. The response of the RV to chronically increased afterload is highly individual and cannot currently be anticipated. Moreover, falling PA pressures on serial clinical assessment should be interpreted as signs of RV failure rather than pulmonary vascular disease regression (an unlikely scenario), unless clearly explained by newly established or escalated medical therapy. This empirical clinical observation was formally confirmed by the seminal work on primary PAH (nowadays referred to as idiopathic PAH) published in 1991. Combining data from multiple clinical centres in the USA participating in a registry for the characterisation of PAH in the 1980s, D'Alonzo et al derived a formula to estimate …