Hepatopulmonary Syndrome: Monitoring at Your Fingertip

Abstract

Hepatopulmonary syndrome (HPS) is the triad of arterial hypoxemia due to pulmonary vascular dilatation induced by liver disease [1]. Diagnosis of HPS relies on measurement of arterial blood gas (ABG) to determine the severity of hypoxemia, whereas vascular dilatation is demonstrated by positive contrast-enhanced transthoracic echocardiography (CE). An invasive test, ABG is usually obtained via a radial artery stick. Approximately 5 cc arterial blood is drawn into a sealed, heparinized syringe and inserted into a blood-gas analyzer within 30 min of collection. Oxygenation is quantified by two measures. PaO2 in mmHg (partial pressure of oxygen dissolved in the plasma; normal considered by most authorities[80 mmHg at sea level) or the calculated alveolar–arterial oxygen gradient (A–a O2 grad) in mmHg (difference between alveolar oxygen and PaO2; normal is age-dependent and\15–20 mmHg). A-a O2 grad is the more sensitive measure that takes into account the degree of ventilation within the lung [1], but it is not calculated in all pulmonary function laboratories. Importantly, oxygenation in patients with HPS can differ significantly (up to 20 mm Hg) as body position changes (orthodeoxia) with PaO2 supine [ sitting [ standing and is worse with exercise [1]. Fortunately, an easier measure of arterial oxygenation has evolved—pulse oximetry [2]. Simply, pulse oximetry allows noninvasive measurement of arterial hemoglobin saturation by oxygen, without the risks associated with arterial puncture [2]. Via a sensor placed on the finger tip (earlobe or forehead as needed), pulse oximetry determines the percentage of oxygen that saturates circulating hemoglobin (SaO2) [2]. This measurement is based upon light absorption within arterial blood at specific wavelengths; red (640 nm) and infrared (940 nm) [3]. Unfortunately, this method cannot distinguish between oxygen and carbon monoxide attached to hemoglobin [2]. The measurement can be one-time or continuous over time. The latter is particularly important during sleep in the setting of advanced liver disease, because SaO2 can frequently drop below 90% [4]. Is there a close relationship between PaO2 and SaO2? Indeed there is via the oxyhemoglobin dissociation curve [2]. Although beyond the scope of this editorial, recall this relationship is sigmoid, not linear. Therefore, under normal conditions, big change in PaO2 (100 ? 50 mmHg) can occur in association with a relatively small change in SaO2 (98 ? 89%). By any measure (PaO2 or A-a O2 grad) hypoxemia has common causes (40–50%) in advanced liver disease; COPD, hepatic hydrothorax, interstitial lung disorders, massive ascites, or obesity, to name a few. Hypoxemia due to HPS is of particular interest because oxygenation can worsen after the diagnosis is made, but it is unclear how rapidly and why change occurs without obvious deterioration in hepatic function. Such change is not trivial, because current organ procurement/liver allocation policy states that once PaO2 is less than 60 mmHg (sitting position, breathing room air) MELD (model for end-stage live disease) exception may be granted to enhance priority for pediatric and adult liver transplantation [5]. How can change in oxygenation be sequentially monitored in HPS patients? Kochar and colleagues in this issue [6] describe pulse oximetry to assess change in oxygenation over time in HPS. They reported periodic, one-time SaO2 measurements obtained in the sitting position, breathing room air. Retrospectively, worsening oxygenation was documented M. J. Krowka (&) Division of Pulmonary and Critical Care Medicine, Mayo Clinic, Rochester, MN 55905, USA e-mail: Krowka@mayo.edu