The Use of Transesophageal Echocardiography to Demonstrate Obstruction of Venous Drainage Cannula During ECMO

Case Presentation A 25 year old man presented to his local hospital with a 1 month history of hemoptysis and hematuria. A provisional diagnosis of vasculitis was made, and steroid therapy was commenced. C-antineutrophil cytoplasmic antibody was strongly positive, renal biopsy confirmed vasculitis, and a diagnosis of Wegener’s granulomatosis was reached. He was commenced on steroids and cyclophosphamide. Over the next 48 hours he developed pulmonary hemorrhage, and his respiratory function deteriorated, requiring ventilatory support. Ventilatory requirements increased, and he needed high pressure/high FiO2 ventilation. Ventilator settings reached peak pressures of 50 cmH2O with a positive end-expiratory pressure of 20 cmH2O and an inverse ratio ventilation of 2:1 with 100% O2, giving maximum O2 saturations of 66%. He became hemodynamically unstable and required inotropic support. He was referred to our institution for consideration for ECMO. Technique At the time of referral, the patient had a Murray score of 3.25 (Figure 1).1 He required hand ventilation throughout transfer, which achieved maximum saturations of 40–60%. PreECMO blood gases showed pH 6.96, PaCO2 92 mm Hg, PaO2 60 mm Hg, and Base Excess (BE)−11.8. He was immediately cannulated for venovenous ECMO with a short 28 Fr DLP® drainage cannula in the right jugular vein and a 23 Fr (25 cm) Biomedicus return cannula in the left femoral vein. An extracorporeal flow of 4.2 L/min was obtained but was insufficient to allow the patient to be weaned from the ventilator settings. Therefore, a third cannula was inserted—a long 28 Fr DLP® cannula into the inferior vena cava via the right femoral vein. The return flow was changed to the new (right femoral) cannula, and the left femoral (short) cannula was converted to drain.2 After this, flows dropped to 1.6 L/min and O2 saturations fell to 60%, with hemodynamic instability. Bilateral intercostal drains were placed to exclude pneumothorax, with little effect. Emergent transthoracic echocardiography showed a possible pericardial effusion but with poor windows. Therefore, a transesophageal echocardiogram (TEE) was performed, which demonstrated poor flow and a mobile thrombus within the jugular venous drainage cannula. Guidewire exchange of this cannula verified its obstruction with the clot and resulted in restoration of flow to 5.4 L/min. Repeated TEE confirmed good flows through the superior vena cava cannula. Thrombophilia screen was negative, but in view of the thrombotic problems, the activated clotting time was kept between 200 and 220 seconds for the first 24 hours. The patient received pulsed cyclophosphamide therapy and steroids while on ECMO. He was decannulated after 235 hours and made a good recovery.Figure 1.: Murray score for grading of acute respiratory distress syndrome.Presumably, this patient had a thrombus in the right iliofemoral system, which was displaced by the third cannula into the jugular cannula. The utility of TEE in diagnosing this unusual problem is amply demonstrated here. Discussion Problems with gaining adequate venous drainage while on ECMO are not uncommon. Obvious causes for poor drainage, such as hypovolemia, tension pneumothorax, cardiac tamponade, and poor cannula position, should be investigated and treated. Once these reversible causes have been managed, then placement of an additional venous drainage cannula should be considered. In this case, flows deteriorated with placement of an additional cannula. This led us to consider all of the aforementioned causes and investigate them. The use of TEE allowed us to demonstrate obstruction within the venous drainage cannula secondary to thrombus. TEE is an imaging modality that is being used more commonly. The diagnostic accuracy of TEE within the intensive therapy unit setting has already been reported,3 and despite good data from Nixdorff et al.,4 it is still an uncommonly used modality in the diagnosis of pulmonary embolus. A prospective trial5 has demonstrated TEE to be an extremely useful bedside tool in the diagnosis of suspected massive pulmonary embolism. The authors were able to visualize the pulmonary trunk, right pulmonary artery, and proximal left pulmonary artery, as well as gain information regarding ventricular function. This allowed them to proceed to intervention with pulmonary thrombolysis. The bedside benefits of TEE have obvious attractions in the care of critically ill patients and particularly in ECMO patients, for whom transfer to the radiology department for investigations such as pulmonary angiography or spiral computed tomography scan would not be feasible. Conclusion The possibility of displaced thrombus within a venous drainage cannula should be considered (along with more common causes) in patients with sudden deterioration in ECMO flows. TEE provides excellent information not only on the presence or absence of thrombus but also on cardiac filling status, cannula position, ventricular function, and the presence of pericardial fluid. We would therefore recommend the early use of TEE in patients on ECMO with problems with venous drainage. Acknowledgment We would like to express our thanks to Dr. David Chinn and the Department of Echocardiography for their assistance with providing images for this case.