Hepatopulmonary syndrome (HPS) is defined as the triad of CLD, arterial hypoxaemia and abnormal intrapulmonary shunting [1, 3]. Hypoxaemia is a significant consequence of HPS and is typically due to intrapulmonary shunting of blood from intrapulmonary vascular dilatations [4]. HPS can occur as a result of CLD and is independent of aetiology [5]. HPS can also occur in the context of PHT, but patients will often have stable and compensated liver disease [5]. While the incidence of HPS is not widely reported, retrospective studies have suggested a prevalence of 9–20% and 0.5% in children with biliary atresia and extra-hepatic portal vein obstruction, respectively [6, 7]. HPS can be challenging to recognise, especially in the early stages when the patient can be asymptomatic [8]. However, as HPS progresses patients may experience dyspnoea, hypoxaemia and cyanosis [8].

Platypnoea, defined as shortness of breath relieved by lying down, is considered an important indicator to suspect and screen patients for HPS [1]. Furthermore, orthodeoxia is deemed to be a specific sign of HPS although importantly it does not occur in all patients [1]. Instead, patients may report symptoms such as breathlessness during activity and during mobilisation [3].

When there is clinical suspicion of HPS, it is important to carry out documentation of arterial hypoxaemia [1, 9]. Although it allows more reliable measurement of hypoxaemia, gaining an arterial oxygen value in paediatrics can be challenging due to the invasive nature of arterial blood gases [3]. However, pulse oximetry can be used as a screening tool to detect hypoxaemia and has been shown to have high sensitivity and specificity [1]. Pulse oximetry should not be used in isolation since it is insufficient in diagnosing HPS alone; instead it should be used in conjunction with other measures to raise the clinical suspicion of HPS [10]. It is important to mention that children with subclinical HPS may have an oxygen saturation measured by pulse oximetry of >98% and therefore HPS may be harder to recognise [3]. Serial haemoglobin levels have been suggested as a tool to raise an index of suspicion in these patients [3]. Therefore, clinicians should remain cautious about the potential of HPS in children with liver disease.

Technetium-99m-labelled macroaggregated albumin (99mTc-MAA) lung perfusion scans are used in the diagnosis of HPS, allowing for the detection of intrapulmonary shunting and intrapulmonary vascular dilatation (IPVD) [3, 10]. However, its low sensitivity, especially in patients with mild-to-moderate HPS, and invasive nature open up the opportunity for other measures to be used [10, 11]. One method described is the use of an oxyhaemoglobin dissociation curve as a noninvasive measure of oxygen abnormalities [11]. This method can be used to establish ventilation/perfusion mismatch as well as a right-to-left shunt in children with CLD, acting as a tool for screening and monitoring HPS in these patients [11]. Such methods may have an important role in clinical practice due to their noninvasive nature compared with 99mTc-MAA lung perfusion scans, as well as their potential to allow for early diagnosis of HPS in children with CLD [11]. Contrast enhanced echocardiography, using agitated saline with microbubbles, can also be used to confirm the presence of IPVD and differentiate intracardiac shunting from intrapulmonary shunting [1]. The presence of agitated saline bubbles in the left atrium on echocardiography, typically after 3–5 cardiac cycles is suggestive of intrapulmonary shunting and HPS [1].

While there is a greater understanding of outcomes of HPS in adults, previous studies have highlighted higher mortality and poorer quality of life in patients with CLD and HPS compared to those without HPS [3]. Outcomes post-LT are of particular importance since LT is the only treatment for HPS [1]. Furthermore, failure to recognise HPS in a patient may delay appropriate treatment, which has implications such as progressive polycythaemia and home oxygen requirements [3]. A small study highlighted that patients with pre-transplant polycythaemia required longer stays in intensive care and had a higher requirement for oxygen therapy [3]. These adverse outcomes from delayed treatment highlight the requirement for timely diagnosis of HPS, which may be challenging in asymptomatic children. Screening for HPS could be enforced to increase the chances of timely diagnosis and treatment of HPS to improve outcomes for these children.

Another consequence of CLD is portopulmonary hypertension (PoPH). The hallmarks of PoPH include raised pulmonary arterial pressure, increased pulmonary vascular resistance with pulmonary arterial occlusion, or a left-ventricular end-diastolic pressure of <15 mmHg [9]. The mechanisms via which PoPH occurs as a consequence of PHT is not known, although several hypotheses have been proposed. Almost all patients who have PoPH have a hyperdynamic circulation and high cardiac output resulting in increased stress of the pulmonary circulation [9]. As a consequence of the increased resistance, remodelling of the pulmonary vasculature is undertaken, which can result in pulmonary hypertension. Common presentations include dyspnoea on exertion, fatigue and syncope, with the latter being less common [12].

PoPH is associated with poor survival, especially when patients do not receive treatment or undergo a LT [13]. However, transplantation does not ensure regression of symptoms, with many patients still requiring treatment after LT [13].