Pulmonary artery thrombectomy – a life‐saving treatment in a patient with presumed COVID‐19 complicated by a massive pulmonary embolus

As new evidence emerges regarding the abnormal coagulation status in patients with novel coronavirus pneumonia (NCP), we are seeing more patients with venous thromboembolic (VTE) disease. ¹ , ² Hypercoagulability is not solely responsible for this evolving phenomenon; the other two variables in Virchow’s triad, namely venous stasis and epithelial injury are also fulfilled as patients lie immobile for prolonged periods on the intensive care unit (ICU). ³

New guidelines regarding thrombosis prevention and coagulopathy treatment have been published by Thrombosis UK to keep up with the ever changing management of patients with this unpredictable disease. ⁴

Case report

We discuss a patient with massive pulmonary embolism (PE) and suspected coronavirus disease 2019 (COVID‐19) with classic chest X‐ray (CXR), computed tomography (CT) ⁵ and clinical findings. A 58‐year‐old male presented to the emergency department by ambulance with a 9 day history of shortness of breath, non‐productive cough and feeling generally unwell. He had been self‐isolating for the last 10 days, denied foreign travel and had no known COVID‐19 contacts. He lived with his wife who also reported less severe symptoms of cough and shortness of breath. He was previously fit and well and on no regular medication.

On examination he had increased work of breathing and was unable to speak in full sentences. His heart sounds were normal and his abdomen was soft. His calves were soft and non‐oedematous. His Glascow Coma Score was 15/15. His blood results were as follows: international normalised ratio (INR) 1·1, activated partial thromboplastin time (aPTT) 22·7 s and fibrinogen was raised at >6 g/l. His white cell count was raised at 21·3 × 10⁹/l with a lymphopaenia of 0·98 × 10⁹/l and neutrophil count of 18·5 × 10⁹/l. Haemoglobin, hepatic function and renal function was normal. A nasopharyngeal swab to isolate severe acute respiratory syndrome coronavirus 2 (SARS‐CoV‐2) was negative. The patient had four subsequent swabs during his admission, all of which failed to detect the virus. No other respiratory viruses were isolated.

His CXR on admission (Fig 1) showed bilateral peripheral airspace opacifications, which are deemed classical findings according to British Society of Thoracic Imaging (BSTI) guidelines. ⁵ The patient was intubated and ventilated and admitted to ICU.

Fig 1.

CXR showing bilateral peripheral mid and lower zone patchy airspace opacifications. These are classic COVID‐19 findings according to the British Society of Thoracic Imaging.

The patient had a pulseless electrical activity cardiac arrest on ICU the next day. A bedside echocardiogram (ECG) showed dilated right cardiac chambers with intracardiac thrombus. Systemic thrombolysis was administered on ICU (100 mg of tissue plasminogen activator, tPA) failed to improve his haemodynamic status or blood oxygenation. At 20 h later, after further decline in both pulmonary gas exchange and haemodynamic stability, another ECG showed dilated right cardiac chambers and reduced radial function. He was discussed with haematology and interventional radiology, and a CT pulmonary angiogram (CTPA) was performed. This confirmed saddle embolus with large volume of clot burden within all the segmental pulmonary arteries and right heart strain (Fig 2). There were peripheral bilateral ground glass opacities, in keeping with classic COVD‐19. The patient attended the interventional radiology suite for catheter‐directed thrombectomy. At the start of the procedure his oxygen saturation was 93% on 100% oxygen and he had a tachycardia of 130 beats/min.

Fig 2.

(A) CTPA showed a saddle pulmonary embolus (arrows). Peripheral airspace consolidation is seen within the right upper lobe. There is a small right pleural effusion and some left lower lobe collapse. (B) There is bowing of the intraventricular septum (arrow heads) and a dilated right atrium, in keeping with right heart strain. (C) Complete resolution of saddle embolus. There is a small left pleural effusion and peripheral lung consolidation, in keeping with resolving NCP.

Ultrasound‐guided access to the right common femoral vein was achieved. A Berenstein II catheter (Cordis, Miami, FL, USA) and hydrophilic Zipwire™ (Boston Scientific, Marlborough, MA, USA) were manipulated into the pulmonary trunk. The wire was exchanged for a 260‐cm Amplatz superstiff wire and an 80‐cm 10‐F Flexor sheath (Cook Inc., Bloomington, IN, USA) was placed with its tip in the main pulmonary trunk. A pumped angiogram showed a saddle embolus with reduced enhancement of the peripheries of the lungs (Fig 3a). A 115‐cm Penumbra Indigo^® CAT8TORQ 8‐F mechanical thrombectomy catheter (Penumbra Inc, Alameda, CA, USA) was then introduced into the sheath and manipulated into the lower lobe pulmonary arteries bilaterally to engage thrombus. A 60‐ml luer lock syringe was attached to the end of the catheter and manual aspiration was performed several times to extract thrombus. During the procedure a bolus of 20 and 30 mg of tPA was administered into the right and left pulmonary arteries respectively.

Fig 3.

(A) Pumped pulmonary artery angiogram shows central filling defect (small arrows), in keeping with a saddle embolus. Decreased peripheral pulmonary enhancement, in keeping with distal thrombus burden. (B) Post‐thrombectomy angiogram shows resolution of saddle embolus with some improved peripheral enhancement.

During clot retrieval, the patient showed intermittent signs of improvement; at one point saturations improved to 95% but returned to pre‐procedure levels again. Given a good angiographic result (Fig 3b) but static haemodynamic stability a decision was made to stop after 80 min. He had been on a transport ventilator for >2 h by this point and once back on ICU he was placed on a departmental ventilator and nursed in a prone position. Over the following hours and days his oxygenation and haemodynamic stability improved. After a 14 day stay in ICU he was extubated and moved to ward‐based care. A subsequent CTPA done 16 days after the procedure showed almost complete resolution of the pulmonary embolus (Fig 2).

Discussion

This case highlights the diagnosis of COVID‐19 is not straightforward and confirmation cannot be relied on nasopharyngeal swab reverse transcriptase‐polymerase chain reaction (RT‐PCR) results alone. Although this test is very specific, the sensitivity has been reported as low as 60–70%, ⁶ adding to the diagnostic challenge of NCP. During the outbreak in China, Chinese authorities broadened the official diagnostic criteria to include patients with typical findings on CT chest, ⁷ indicating the importance of imaging as a diagnostic tool. We have adopted this in our institution due to the high number of false negatives.

Additionally, this case not only emphasises the increased VTE events associated with COVID‐19 but also the severity of them. In both haematology and interventional radiology, our routine workload has altered and we need to be prepared to treat the complications of severe NCP alongside our intensive care colleagues. This case highlights that even if oxygen requirement or vital signs do not improve immediately after thrombectomy, it does not mean it is a futile procedure; quality of mechanical ventilation and patient positioning also have an impact on oxygenation and haemodynamic stability. In this case, percutaneous pharmaco‐mechanical pulmonary artery embolectomy was a successful and life‐saving treatment. We believe it should be offered to other patients with NCP‐associated massive pulmonary emboli.

Take home points

• Relying solely on a positive PCR nasopharyngeal swab to diagnose COVID‐19 will lead to incorrect patient management and may put members of staff at risk. Clinical and radiological features must be taken into account when diagnosing COVID‐19 pneumonia.

• When standard therapies are unsuccessful in treating massive pulmonary emboli in haemodynamically unstable patients, catheter‐directed thrombolysis and pulmonary artery thrombectomy can be a successful treatment in deteriorating patients and should always be considered.

• During the fight against COVID‐19 clinicians need to be aware that managing these patients is complex and often requires input from multiple specialties; communication is key between intensivists, haematologists and interventional radiologists to ensure the COVID‐19‐positive patient is management appropriately.

Conflict of interest

The authors declare they have no conflict of interest.

Funding

This study was not supported by any funding.