Skip to main content
BMJ Case Reports logoLink to BMJ Case Reports
. 2019 Nov 26;12(11):e231499. doi: 10.1136/bcr-2019-231499

Platypnoea and orthodeoxia in the hepatopulmonary syndrome

Anna Surridge 1, Jane Atkinson 1, Avinash Aujayeb 1,
PMCID: PMC6887442  PMID: 31776152

Abstract

A 71-year-old female patient with alcohol-induced cirrhosis presented with symptoms of dyspnoea. Previous extensive investigations had detected no apparent cause. Platypnoea and orthodeoxia were observed. A bubble echocardiogram revealed significant intracardiac shunting and a diagnosis of hepatopulmonary syndrome was made. The patient was discharged on home oxygen and referred for liver transplantation.

Keywords: Respiratory medicine, Gastroenterology

Background

There are numerous pulmonary complications of chronic liver disease: hepatic hydrothorax, spontaneous bacterial pleuritis, pulmonary infections, acute respiratory distress syndrome, portopulmonary hypertension and hepatopulmonary syndrome (HPS). These are all associated with significant morbidity and mortality.

HPS is a rare disease and requires a high index of suspicion and specialist investigations. The only cure is through liver transplantation.1 2

Case presentation

A 71-year-old female patient presented acutely with a short history of increased dyspnoea and reduced exercise tolerance. There was no associated cough, chest pain or haemoptysis.

She had never smoked and had no occupational dust exposure. Her medical history included alcohol-induced liver cirrhosis. She had been abstinent for almost 2 years and was well-known to the gastroenterological team. Her liver function tests had slightly deteriorated over that period with a rise in her alanine transferase to 68 U/L (normal<40 U/L), albumin 33 g/L (normal 35–50 g/L) and bilirubin 46 μmol/L (normal<21 μmol/L).

Her medication list included lansoprazole 30 mg per day and thiamine 100 mg two times per day.

She was normotensive, had a heart rate of 67 beats per minute, a respiratory rate of 20 breaths per minute and oxygen saturations of 90% on room air.

On examination, she had a normal cardiovascular, respiratory and gastrointestinal examination. There was no lymphadenopathy, clubbing, jaundice, spider naevi, liver flap, clinically demonstrated ascites or ankle oedema.

Investigations

Her blood tests showed normal values for haemoglobin, white cell count, urea, creatinine, prothrombin time and glucose. Her liver function tests showed no deterioration from previous investigations. Her D-dimer level was within the normal age-adjusted range and her score on the Wells’ criteria was low. Both the ECG and chest radiograph were normal. Arterial blood gas analysis showed type 1 respiratory failure. There was a normal pH and carbon dioxide but a low arterial PaO2 at 7.1 kPa (10.7–13.3 kPa).

Previous investigations were reviewed. Within the last 2 years, she has had two similar admissions with dyspnoea. A CT of her chest and upper abdomen was normal. A transthoracic echocardiogram was also normal and showed no elevated right-sided heart pressures.

To assess her so far undiagnosed hypoxia, she was referred to a respiratory physician. A targeted history of her symptoms revealed platypnoea; worsening of dyspnoea when standing up. Oxygen saturations were 90% on air and dropped to 85% within a minute of standing. She had to sit down to relieve herself of her increased dyspnoea; thus, orthodeoxia was demonstrated.

A bubble echocardiogram was performed. The images (figures 1–4) show a clear and significant intracardiac shunt from the fifth phase onwards.

Figure 1.

Figure 1

Capture of echocardiogram at the first beat showing no bubbles in left side.

Figure 2.

Figure 2

Capture of echocardiogram at the third beat showing some bubbles in left side.

Figure 3.

Figure 3

Capture of echocardiogram at the fourth beat showing an increasing number of bubbles in left side.

Figure 4.

Figure 4

Capture of echocardiogram at the fifth beat showing a significant number of bubbles in left side.

Differential diagnosis

Unexplained hypoxia has a wide range of differential diagnoses. Physiologically, it is due to a mismatch between ventilation and perfusion. Normal haemoglobin levels ruled out anaemia. A normal chest radiograph ruled out primary pulmonary causes, such as hepatic hydrothorax, pneumothorax and pneumonia. A normal CT scan within the last year corroborated this. Normal D-dimer levels and a low Well’s score made a pulmonary embolism very unlikely. A normal echocardiogram and the absence of peripheral oedema ruled out pulmonary hypertension.

Platypnoea and orthodeoxia are very specific symptoms and can be signs of a rare condition called HPS.1–3 A bubble echocardiogram showing an intracardiac shunt clinches the diagnoses.

Treatment

Home and ambulatory oxygen were provided to keep saturations between 92% and 94%. A referral to the gastroenterology service was made. An inpatient abdominal ultrasound ruled out hepatocellular carcinoma and an urgent referral for liver transplantation was sent.

Outcome and follow-up

Symptoms and exercise tolerance have improved greatly with oxygen therapy. The patient has been referred for consideration of a liver transplant and is awaiting preassessment at the time of writing.

Discussion

HPS is a condition that only affects patients with liver cirrhosis, irrespective of the degree of liver disease. It is a triad of liver disease and/or portal hypertension, PaO2<80 mm Hg while breathing ambient air and documented intrapulmonary vascular dilatation by contrast-enhanced echocardiography or lung perfusion scanning with radioactive albumin.1

The prevalence ranges from 4% to 32% depending on the population studied and diagnostic cut-offs of oxygen used.2

The pathogenesis is complicated.1–5 Liver fibrosis increases the production of proinflammatory cytokines, such as endothelin 1 and tumour necrosis factor-A. This leads to pulmonary vasodilatation mediated by excessive nitrous oxide production as well as accumulation of excessive macrophages and monocytes that promote angiogenesis and pulmonary capillary proliferation. Increased blood flow with preserved alveolar oxygenation causes the formation of intrapulmonary shunts between the arterial and venous circulation, which leads to ventilation-perfusion mismatch followed by arterial hypoxaemia. Another mechanism is diffusion-perfusion impairment (also known as alveolar-capillary oxygen disequilibrium) which is due to the vasodilation of alveolar capillaries. The PaO2 in the alveolus is insufficient to achieve equilibration with blood in the vessel. As such, with supplemental oxygen, the gradient for diffusion increases and oxygenation improves, although a caveat to this is explained below.2 3

The PaO2 suggests that the severity of HPS was moderate. PaO2 is normally ≥80 mm Hg (10.7 kPa) on air, moderate HPS has a PaO2≥60 and <80 mm Hg (≥8 and <10.7 kPa) on air, severe HPS has a PaO2≥50 and <60 mm Hg on air and very severe HPS has a PaO2<50 mm Hg (6.7 kPa) on air or <300 mm Hg (40 kPa) on 100% oxygen.3

Intrapulmonary vascular dilatations can occur in two distinct patterns. Type I lesions are more common and have diffuse pulmonary vascular dilatations at the precapillary level close to the normal gas exchange units of the lung. Type II lesions are more concrete, localised dilatations and have large arteriovenous communications distant from the gas exchange. Type I lesions are responsive to oxygen therapy, but type 2 lesions have a poor to non-existent response to oxygen therapy as they have true large shunts.1–3 Coil embolotherapy can significantly improve oxygenation in patients from the latter category.1

Our patient probably has a mixture of type 1 and 2 lesions. On formal oxygen testing, her resting oxygen saturations on air were 88%. With 2 L of oxygen via nasal cannula, saturations rose to 90% but increasing concentrations of oxygen did not increase her saturations. She felt symptomatically better at 90% on 2 L, and felt that she could now walk without dyspnoea and 2 L via nasal cannula were, hence, prescribed.

HPS presents insidiously with progressive dyspnoea and a high degree of suspicion is required for the diagnosis. Patients will normally have signs of chronic liver disease but the presence of spider naevi (also known as spider angiomas) particularly raise the suspicion for HPS as the causative mechanism for this phenomenon, through angiogenesis, is the same. A higher prevalence of HPS has been associated with patients who have spider naevi. Our patient had a significant number of spider naevi.1 2

Platypnoea is a classic symptom but is not pathognomonic. It is an increase in the feeling of breathlessness when standing up which improves when lying down. Orthodeoxia is the decrease in PaO2 (typically by more than 4 mm of Hg or 0.5 kPa) or a reduction in saturations by more than 5%. Our patient had a drop of 5.6%. Platypnoea-orthodeoxia syndrome is rare and present in HPS. The syndrome is also described in patients with right-to-left shunts due to a persistent foramen ovale or pulmonary arteriovenous malformations.6 Orthodeoxia affects up to 88% of patients with HPS compared with less than 5% of patients with cirrhosis alone. This is due to intracardiac shunting, intrapulmonary shunting, ventilation-perfusion mismatch or a combination of these.7

There are two investigations of choice: a transthoracic contrast echocardiogram (also known as bubble study) and radionuclide perfusion scanning. The former is less invasive and was locally available.

The additional intrapulmonary capillaries act as a physiologic shunt to allow a higher volume of oxygenated blood to bypass the alveoli. Thus, shunt assessments are elevated in HPS.1–4 Agitated saline was injected intravenously. Normally, the bubbles would only be seen on the right side of the heart because of filtration by the capillaries. However, the bubbles will be seen on the left if a right-to-left intracardiac or intrapulmonary shunt is present. With the former, the bubbles appear within one cardiac cycle, but in the latter, the bubbles appear after three to eight beats. In our patient, the appearance was noted after the fifth beat. Technetium-99m-labelled macroaggregated albumin scanning is the other diagnostic test that can be performed. The macroaggregates should be trapped in the capillary bed, but in the presence of HPS, there is a significant uptake in the kidneys or the brain. However, such a scan cannot differentiate between an intracardiac and intrapulmonary shunt.1 4

Long-term supplemental oxygen is used as supportive therapy and has greatly benefited our patient, restoring her quality of life. However, the only definitive treatment is liver transplantation. Observational studies have shown complete or near-complete resolution of HPS with improved oxygenation and shunt in the majority of patients within the initial 6–12 months with no increased mortality. The discussion of the evidence base behind liver transplantation for patients with HPS and their associated Model for End-stage Liver Disease score is beyond the scope of this case report.1–6

Patient’s perspective.

I have been troubled by breathlessness for over a year now and have not had a diagnosis until now. I used to drink too much alcohol but have, now, not had a drop of drink for many years and I didn’t think my liver would be affecting me like that. My gastroenterologist had said that my condition was stable. I am glad that I now have a diagnosis, although I am troubled by the fact that I might need a new liver and scared by what this means for me in the long-term. The oxygen that has been supplied to me helps but I find it quite restrictive. I have asked for a more portable device so that I can get around more. I am more than happy to contribute to this publication as I think more and more doctors need to made aware of this very rare condition that I have as I have probably had it for a long time and it has been undiagnosed for that long.

Learning points.

  • Hepatopulmonary syndrome presents insidiously and requires a high degree of clinical suspicion.

  • Platypnoea and orthodeoxia are classically associated with the syndrome but not pathogonomic.

  • The investigation of choice is a transthoracic contrast echocardiogram (also known as bubble study).

  • Long-term oxygen therapy is beneficial as supportive therapy.

  • The only cure is through liver transplantation.

Acknowledgments

Thank you to Dr Praveen Rajashekar who referred the patient to the liver transplant team.

Footnotes

Contributors: AA, JA and AS all contributed equal parts to the manuscript. AA edited the final version.

Funding: The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.

Competing interests: None declared.

Patient consent for publication: Obtained.

Provenance and peer review: Not commissioned; externally peer reviewed.

References

  • 1. Ho V. Current concepts in the management of hepatopulmonary syndrome. Vasc Health Risk Manag 2008;4:1035–41. 10.2147/VHRM.S3608 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 2. Rodríguez-Roisin R, Krowka MJ. Hepatopulmonary syndrome--a liver-induced lung vascular disorder. N Engl J Med 2008;358:2378–87. 10.1056/NEJMra0707185 [DOI] [PubMed] [Google Scholar]
  • 3. Gómez FP, Martínez-Pallí G, Barberà JA, et al. Gas exchange mechanism of orthodeoxia in hepatopulmonary syndrome. Hepatology 2004;40:660–6. 10.1002/hep.20358 [DOI] [PubMed] [Google Scholar]
  • 4. Lang PA, Stoller JK. Hepatopulmonary syndrome in adults: prevalence, causes, clinical manifestations, and diagnosis. Available: https://www.uptodate.com/contents/hepatopulmonary-syndrome-in-adults-prevalence-causes-clinical-manifestations-and-diagnosis [Accessed 18 Oct 2019].
  • 5. Younis I, Sarwar S, Butt Z, et al. Clinical characteristics, predictors, and survival among patients with hepatopulmonary syndrome. Ann Hepatol 2015;14:354–60. 10.1016/S1665-2681(19)31275-X [DOI] [PubMed] [Google Scholar]
  • 6. Rojas E, Aktas A, Parikh H, et al. Platypnea-orthodeoxia syndrome in a patient with cryptogenic liver cirrhosis: an elusive cause of hypoxemia. Cureus 2019;11:e3846 10.7759/cureus.3846 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7. Agrawal A, Palkar A, Talwar A. The multiple dimensions of Platypnea-Orthodeoxia syndrome: a review. Respir Med 2017;129:31–8. 10.1016/j.rmed.2017.05.016 [DOI] [PubMed] [Google Scholar]

Articles from BMJ Case Reports are provided here courtesy of BMJ Publishing Group

RESOURCES