Case of positional dyspnoea and hypoxia secondary to intracardiac shunting

Abstract

Platypnea-orthodexia syndrome (POS) is a rare but well-characterised condition where hypoxaemia and breathlessness occur while upright but resolve once recumbent. Early recognition can result in excellent outcomes for patients and can prevent unnecessary investigations for patients, especially if they present repeatedly to hospital after missed diagnosis. We present a case of a 75-year-old woman with a chronic history of breathlessness who was picked up after observations at a routine outpatient clinic. Early recognition of the POS allowed for appropriate investigations to take place identifying a patent foramen ovale (PFO). She was referred to the tertiary centre for closure of her PFO with complete resolution of her symptoms.

Background

Breathlessness is a common cause of presentation to hospital with a wide differential diagnosis. It is easy to overlook rarer causes, and hence having a systematic approach can prove vital when the cause is not so apparent. This case will highlight the diagnostic challenges when faced with a patient who desaturated significantly, particularly while in the upright position.

Case presentation

Mrs A, a 75-year-old woman, was admitted to our respiratory ward from the outpatient stroke clinic. She was noted to be breathless with saturations of 68% measured on the pulse oximeter. She was immediately given oxygen via nasal canula and transferred to the emergency department.

She had a background of stroke 16 years ago, hypothyroidism, hypertension and hyperuricaemia. She was also noted to have mild dilatation of her descending aorta and enlarged right ventricle on an echo performed in 2019. She was otherwise fit and well and independent with all activities of daily living.

On admission her home medications included levothyroxine 50 mcg once a morning, bisoprolol 1.25 mg once a day, atorvastatin 40 mg once a day, clopidogrel 75 mg once a day, doxazosin 1 mg once a day, omeprazole 20 mg once a day and spironolactone 25 mg once a morning.

Her initial observations were saturation of 82% on 15 L oxygen via a non-rebreathe mask, respiratory rate 18 beats per minute, blood pressure 106/73 mm Hg, heart rate 83 beats per minute and temperature 36.1 ºC. On examination she had mild dyspnoea but was able to speak in full sentences, her lung fields were clear on auscultation and there were no appreciable heart murmurs or added sounds.

On further questioning she had been feeling breathless for the last 6 months which had been gradually getting worse. She denied having any cough, fevers, chest pain or leg oedema. She was otherwise well and independent with all activities of daily living.

Investigations and differential diagnosis

Her initial blood tests showed a mild neutrophilia with normal C reactive protein, normal liver function tests and a mild rise in creatinine and no dynamic change in repeat serial troponins.

Her arterial blood gas done in the emergency department performed on 4 L oxygen showed a pO₂ 5.74 kPa, saO₂ 79.4%, pCO₂ 3.50 kPa, pH 7.462 and HCO₃ 21.3 mmol/L, and Alveolar-arterial oxygen gradient was 24.1 kPa (age adjusted normal range 3.0 kPa).

Her chest radiograph showed no abnormality, and a CT pulmonary angiogram was negative for pulmonary embolism, showed normal lung parenchyma but noted an aneurysmal dilatation of the aortic room. This ruled out common causes of hypoxia such as pneumonia, pulmonary embolism and pulmonary oedema. Coronavirus (COVID-19) pneumonitis was ruled out by a negative PCR test.

She went on to have further specialised investigations; a contrast bubble transthoracic echocardiogram (TTE) demonstrated a large right to left interatrial shunt which was then further evaluated by a transoesophageal echocardiogram (TOE) which only showed a small patent foramen ovale (PFO) with small resting shunt at rest. Other findings included mild to moderate aortic regurgitation with a dilated aortic root and ascending aorta. There was no evidence of pulmonary hypertension with an estimated right ventricular systolic pressure of 24 mm Hg. During the TOE her saturations remained normal; therefore, the conditions for the shunt may not have been present during the procedure. The working diagnosis currently was the platypnea-orthodexia syndrome (POS). She was referred onto a tertiary cardiology unit for percutaneous closure of PFO. During the procedure a TOE was performed which demonstrated a PFO with an approximately 10 mm long tunnel with right to left shunting noted. When agitated saline was injected to the inferior vena cava, there was significant opacification of the left atrium and left ventricle. This was not seen when agitated saline with 0.5 mL of own blood was injected to the right ventricle, making a significant pulmonary arteriovenous malformation unlikely. Following this her PFO was successfully occluded.

Treatment

Initial management of the patient included correction of the hypoxaemia with supplemental oxygen, but due to the likely right to left shunt the benefit of this was limited. The patient was mainly nursed in the recumbent position while the initial investigations were carried out.

Once the working diagnosis of POS was considered, she was referred to a tertiary care centre for trial of PFO closure with excellent outcomes.

She was followed up 3 months post discharge, and her breathlessness and exercise tolerance had improved greatly. Her oxygen saturations on home monitoring remained between 96% and 99%.

Discussion

POS is a rare but well characterised syndrome, defined as hypoxia and breathlessness while in the upright position which improves once recumbent. It can occur with either cardiac or extra-cardiac abnormalities.

Approximately 25% of the population have a PFO¹ with no symptoms at all. This case highlights how a seemingly innocuous PFO which can cause occasional hypoxaemia develops into a significant clinical problem with difficulty in diagnosis. The presence of this anatomical variant with a functional change such as a dilated aortic root, pericardial effusion or pericarditis which causes deviation of the atrial septum, allows the POS to arise in these patients. It is important to note that right to left shunts can occur in this condition without the presence of pulmonary hypertension; the combination of the anatomical and functional variants allow the streaming of deoxygenated blood from the inferior vena cava into the PFO while upright.² Having a high degree of clinical suspicion is important in being able to recognise this syndrome allowing appropriate investigations to take place.

Given the significant desaturation while upright a differential diagnosis of POS was considered. With the hypoxaemia being positional the ventilation-perfusion (V/Q) ratio was likely being reduced in the upright position and corrected once recumbent. V/Q mismatch is one of the causes of hypoxaemia. Ventilation is the product of the tidal volume and respiratory rate whereas perfusion is the volume of blood delivered to the alveoli per unit time. In normal physiology ventilation is greater at the apices of the lung compared with perfusion while at the lung bases the reverse is true. In normal physiology the overall V/Q ratio is around 0.8. Abnormalities can occur in both ventilation and perfusion causing hypoxaemia. Figure 1 summarises the different causes of the POS.

Figure 1.

Differential diagnoses which can cause the platypnea-orthodexia syndrome and how to differentiate between them.

A low V/Q ratio is associated with reduced ventilation relative to perfusion. This can be positional in certain pulmonary pathology such as basal interstitial fibrosis. In the upright position, most of the perfusion to the lungs is to the basal areas where the ventilation is suboptimal due to the fibrosis but corrected once recumbent as the well-ventilated apices of the lung would then be well perfused. A normal CT chest was useful in ruling this out as a cause for the patient. Other causes of reduced ventilation that could be positional include obesity hypoventilation syndrome; however, this is often worse in the supine position which was not the case here.

An extremely low V/Q ratio implies a right to left shunt where deoxygenated blood bypasses the alveoli enabling it to mix with oxygenated blood in the left side of the circulation. This can occur in several conditions, and each was explored as a possible cause for our patient’s presentation.

Pulmonary arteriovenous shunts when located in the bases of the lung can cause POS. While upright the blood flow to these shunts is increased reducing the V/Q ratio causing hypoxaemia. This was ruled out in our patient by a normal CT pulmonary angiogram. This phenomenon can also occur in hepatopulmonary syndrome.³ Liver disease progression leads to an increase in nitric oxide which causes dilatation of the pulmonary vasculature, and on standing there is increased perfusion of the lung bases relative to ventilation causing a V/Q mismatch. This was ruled out in our patient by normal liver function tests.

Intracardiac shunts are often a result of congenital heart defects. There can be atrio-ventricular septal defects (AVSDs), atrial septal defects (ASDs) or ventricular septal defects (VSDs). Normally the greater pressures present in the left side of the circulation prevent this unwanted mixing from occurring, but this is not the case in certain conditions.

Eisenmenger’s syndrome occurs due to chronic left to right shunting of blood, leading to increasing pulmonary pressures overcoming pressures in the left atrium allowing for a reversal of the shunt causing mixing of deoxygenated blood in left-sided circulation causing hypoxia. As this syndrome was not dependent on positional status and the absence of pulmonary hypertension on her echocardiogram made this an unlikely cause of her symptoms.

More fitting with our case is a transient right to left shunt which occurs when in the upright position but not when recumbent. The higher left atrial pressures are normally sufficient to prevent the right to left shunt, except when there is distortion of atrial anatomy which leads to preferential direction of deoxygenated venous blood usually from the inferior vena cava through a PFO or an ASD. The effect may be greater in the upright than the supine position, and this situation is one of the causes of the POS.⁴ It has been described also in patients with distorted anatomy due to pneumonectomy or dilatation of the aortic root.⁵ In our patient this was confirmed by the finding of the contrast TTE showing a large right to left interatrial shunt likely due to the PFO along with the TOE findings which showed the moderate aortic regurgitation, dilated aortic root arch along with the previously identified PFO. With the diagnosis of POS likely, the referral for trial of PFO closure proved successful with excellent outcomes.

Learning points.

• For the platypnea-orthodexia syndrome (POS) to exist there must be a functional and anatomical component— contrast echo is a useful starting point to demonstrate these.

• Beware that pulmonary pressures may be normal in the POS.

• This syndrome is often picked up later in life.

• The syndrome is potentially treatable with PFO closure, and early diagnosis can prevent unnecessary investigations.

Ethics statements

Patient consent for publication

Consent obtained directly from patient(s).