The outstanding diagnosis.

Dear Editor,

Altman and Robin first described platypnea-orthodeoxia syndrome (POS) in 1969,[1] although this physiological phenomenon had been originally described in 1956.[2] POS results from intra-cardiac, intra-pulmonary or intrahepatic arteriovenous shunts and ventilation-perfusion mismatch. We report this syndrome in a 66-year-old Caucasian female with an 80 pack-year smoking history 7 months after right pneumonectomy for hilar squamous cell carcinoma. One-month post-pneumonectomy, she noticed gradual onset dyspnea, more pronounced on standing upright and with activity, mandating her to use 3 L of oxygen which she did not use before surgery. Preoperative pulmonary function tests showed forced expiratory volume in 1 s (FEV1) of 1.69 L (79% predicted) and diffusing capacity for carbon monoxide (DLCO) of 66% predicted.

On admission, she was hemodynamically stable, but her oxygen saturations were significantly different in supine (98%) versus sitting up (85%) position on 80% high flow oxygen. Computerized tomography of chest with contrast ruled out pulmonary embolism or an infiltrating process. Given her postural desaturation, a likelihood of either cardiac or pulmonary shunt was entertained. A transthoracic echocardiogram (TTE) with agitated saline contrast was non-diagnostic due to motion artifact and a transesophageal echocardiogram (TEE) was pursued in supine position. A team of advanced airway specialists was at bedside during the TEE which demonstrated a fenestrated patent foramen ovale (PFO) with agitated saline [Figure 1 and Video 1] and bidirectional shunt flow [Figure 2 and Video 2]. Right and left heart catheterization done prior to PFO closure revealed normal coronaries, normal chamber pressures (right atrium = 10 mm Hg, pulmonary artery occlusion pressure 8 mm Hg) and an intracardiac echo probe showed positive agitated saline contrast study across inter-atrial septum. The PFO was closed with a 25 mm amplatzer cribriform device. Post procedure, the patient's oxygen saturation dramatically improved to 96% on room air. Follow-up TTE showed no residual shunting. She was discharged home on clopidogrel, aspiring and 2 L oxygen with exertion, which was weaned off in 2 weeks.

Figure 1

Transesophageal echocardiogram demonstrating patent foramen ovale after agitated saline infusion

Figure 2

Transesophageal echocardiogram showing bidirectional flow with color Doppler

Isolated reports of similar complications post-pneumonectomy have been reported after the original description of “postural cyanosis.”[2] All cases occurred after right pneumonectomy or lobectomy; usually around 8 weeks following surgery. This time lag may be from time taken for shifting and reorientation mediastinal structures and hyperinflation of opposite lung. The pathophysiologic mechanism of POS post-pneumonectomy is related to an anatomic component of atrial septal defect (ASD)/PFO and a functional component from re-direction of shunt flow in an upright posture.[3] Post-pneumonectomy, mediastinal shift causes the inter-atrial septum to stretch and opening up of a previously silent ASD/PFO. In upright position, the weight of the heart pulls the inter-atrial septum causing the PFO/ASD to widen further. The functional component may occur due to detrimental effect of the upright position on cardiac output and generation of pressure gradient across ASD/PFO caused by limited availability of vascular bed and increased pressure due to fluid filled hemithorax. This complication was corrected in most patients with ASD/PFO closure.[3]

Prompt recognition of this postural hypoxemia is important while assessing patients with dyspnea few weeks of right lung resection because closure of the interatrial communication frequently leads to cure as described in our patient.