Advertisement

Echocardiographic Image of a Cannula in the Inferior Vena Cava After Decannulation of Venoarterial Extracorporeal Membrane Oxygenation

      A 16-year-old boy presented to the ED with a 2-day history of dyspnea and hypotension. The patient experienced cardiogenic shock and was diagnosed with fulminant myocarditis. Maintaining hemodynamic stability was a challenge despite the IV administration of large doses of positive inotropic and vasoactive agents. Transthoracic echocardiography (TTE) revealed severe cardiac dysfunction. Femoral venoarterial extracorporeal membrane oxygenation (VA-ECMO) was initiated under ultrasound guidance. Cannulation was performed with use of a right femoral artery reinfusion cannula (15F) and a left femoral venous drainage cannula (21F). The patient's clinical status improved during VA-ECMO support with administration of anticoagulation therapy. VA-ECMO was discontinued after 7 days. TTE after decannulation detected a cannula-shaped structure in the inferior vena cava (IVC), where the femoral venous drainage cannula had been placed during VA-ECMO support. (Video 1, Fig 1A, C, D). Surprisingly, TTE detected blood flow signals in the cannula-shaped structure (Fig 1B).
      Figure thumbnail gr1
      Figure 1(A) Longitudinal ultrasound scan shows a cannula-shaped structure in the external iliac vein. (B) Longitudinal ultrasound scan of the external iliac vein shows blood flow signals in the cannula-shaped structure. (C) Longitudinal ultrasound scan of the IVC shows a cannula-shaped structure in the abdominal segment of the IVC. (D) Horizontal ultrasound scan of the IVC shows a cannula-shaped structure in the abdominal segment of the IVC. The short white arrow points to the cannula-shaped structure in the IVC. The long white arrows point to the external iliac vein and the IVC.
      Question: Based on the echocardiographic images included, what is the most likely diagnosis in this clinical context? After reviewing the clinical information and the echocardiographic images, what would you recommend as the next step?
      Answer: The most likely diagnosis in the clinical context we have described is long- segment caval thrombus. After the administration of anticoagulation therapy, repeat TTE assessments showed no residual thrombotic material 10 days later (Video 2). The patient did not experience symptoms or clinical signs suggestive of IVC syndrome (eg, lower extremity edema) or pulmonary embolism (eg, dyspnea)

      Discussion

      Cannula-associated DVT has been linked to the activation of the coagulation system induced by the contact of blood with nonbiological artificial materials.
      • Annich G.M.
      • Zaulan O.
      • Neufeld M.
      • et al.
      Thromboprophylaxis in extracorporeal circuits: current pharmacological strategies and future directions.
      The diagnosis of DVT is established based on incompressibility of peripheral veins, absence of blood flow, and visualization of thrombi, typically by echocardiography and CT. In some reports, cannula-associated DVT after discontinuation of venovenous ECMO (VV-ECMO) affected between 18.1% and 71.4% of surviving patients.
      • Cooper E.
      • Burns J.
      • Retter A.
      • et al.
      Prevalence of venous thrombosis following venovenous extracorporeal membrane oxygenation in patients with severe respiratory failure.
      • Parzy G.
      • Daviet F.
      • Persico N.
      • et al.
      Prevalence and risk factors for thrombotic complications following venovenous extracorporeal membrane oxygenation: a CT scan study.
      • Fisser C.
      • Reichenbächer C.
      • Müller T.
      • et al.
      Incidence and risk factors for cannula- related venous thrombosis after venovenous extracorporeal membrane oxygenation in adult patients with acute respiratory failure.
      Studies published in recent years have reported superior and inferior vena cava thrombosis rates as high as 51.7% at autopsy.
      • Trudzinski F.C.
      • Minko P.
      • Rapp D.
      • et al.
      Runtime and aPTT predict venous thrombosis and thromboembolism in patients on extracorporeal membrane oxygenation: a retrospective analysis.
      These findings suggest that cannula- associated DVT is a common complication after withdrawal of VV-ECMO. However, massive cannula- associated DVT after discontinuation of VV-ECMO has been reported
      infrequently.
      • Bein T.
      • Philipp A.
      • Pregler B.
      • et al.
      Long-segment caval thrombus after removal of ECMO cannula.
      The pathogenesis of cannula-associated DVT involves rheological changes caused by the blood flow around the cannulas, which may increase thrombogenic potential. The main challenge of anticoagulation therapy during ECMO is to maintain the delicate balance between a reduction in thrombotic risk and prevention of hemorrhage. The activated partial thromboplastin time (aPTT) test has been used to guide and monitor anticoagulation therapy with unfractionated heparin (UFH). However, aPTT is not a reliable measure of anticoagulant effects in practice. A higher level of aPTT has not been associated with a lower frequency of cannula-associated DVT.
      • Parzy G.
      • Daviet F.
      • Persico N.
      • et al.
      Prevalence and risk factors for thrombotic complications following venovenous extracorporeal membrane oxygenation: a CT scan study.
      Multiple variables, including baseline aPTT levels, which may vary between critically ill patients and normal control subjects, influence the reliability of the aPTT test for monitoring UFH therapeutic ranges. Additionally, studies about the predictive value of other blood coagulation indicators, such as prothrombin time, fibrinogen level, mean platelet count, and D-dimer value, for thrombotic events have reached discordant conclusions.
      • Fisser C.
      • Reichenbächer C.
      • Müller T.
      • et al.
      Incidence and risk factors for cannula- related venous thrombosis after venovenous extracorporeal membrane oxygenation in adult patients with acute respiratory failure.
      ,
      • Trudzinski F.C.
      • Minko P.
      • Rapp D.
      • et al.
      Runtime and aPTT predict venous thrombosis and thromboembolism in patients on extracorporeal membrane oxygenation: a retrospective analysis.
      The coagulation indicators that are commonly used during ECMO support are likely insufficient for accurate anticoagulation monitoring. Although effective anticoagulation is necessary to prevent cannula-associated DVT, no universally accepted standard for anticoagulation monitoring is currently available in clinical practice. Further research is needed to improve anticoagulation protocols and achieve reliable anticoagulation monitoring, to successfully prevent cannula-associated DVT.
      The heterogeneous distribution of cannula-associated DVT is determined by the different cannulation approaches. A greater diameter of the cannula often correlates
      with a higher risk of cannula-associated DVT.
      • Cooper E.
      • Burns J.
      • Retter A.
      • et al.
      Prevalence of venous thrombosis following venovenous extracorporeal membrane oxygenation in patients with severe respiratory failure.
      ,
      • Fisser C.
      • Reichenbächer C.
      • Müller T.
      • et al.
      Incidence and risk factors for cannula- related venous thrombosis after venovenous extracorporeal membrane oxygenation in adult patients with acute respiratory failure.
      Therefore, cannula-associated DVT occurs more frequently in the IVC, because the drainage cannula placed in the IVC is usually the largest one. In our center, cannula-associated DVT after VA-ECMO decannulation was detected by TTE in only a few patients. The low number of cases may be explained by our protocol and the characteristics of the patient population in our center. First, we implanted smaller drainage cannulas (19-21 F) in patients. A small cannula may have little effect on the thrombogenic potential of the blood surrounding it. Second, most of the patients in our center received VA-ECMO support for cardiogenic shock. Whether the particular hemodynamic changes associated with the two types of ECMO will have different effects on blood thrombogenicity around the cannula is unclear. Researchers have focused mostly on cannula-associated DVT during VV-ECMO support. Few reports have been published about cannula- associated DVT after discontinuation of VA-ECMO.
      A retrospective observational analysis showed that in-hospital mortality rates and death rates at discharge from the ICU and at 90 days did not differ significantly between patients with cannula-associated DVT and those who did not experience this complication.
      • Parzy G.
      • Daviet F.
      • Persico N.
      • et al.
      Prevalence and risk factors for thrombotic complications following venovenous extracorporeal membrane oxygenation: a CT scan study.
      However, despite the lack of statistical significance, the in-hospital mortality rate was higher in the cannula-associated DVT group, suggesting that thrombotic events related to VV-ECMO may have an impact on mortality outcomes.
      • Cooper E.
      • Burns J.
      • Retter A.
      • et al.
      Prevalence of venous thrombosis following venovenous extracorporeal membrane oxygenation in patients with severe respiratory failure.
      ,
      • Parzy G.
      • Daviet F.
      • Persico N.
      • et al.
      Prevalence and risk factors for thrombotic complications following venovenous extracorporeal membrane oxygenation: a CT scan study.
      Confirming the diagnosis of cannula-associated DVT is critical because DVT is associated with a potential risk of sudden death from pulmonary embolism.
      In theory, DVT can lead to IVC syndrome manifesting with edema of the lower limbs.
      However, this manifestation is not apparent in many patients with DVT, likely because, in most cases, the thrombus is too small to obstruct the blood flow in the IVC. Only a few patients, who have a long-segment DVT, develop IVC syndrome.
      • Bein T.
      • Philipp A.
      • Pregler B.
      • et al.
      Long-segment caval thrombus after removal of ECMO cannula.
      The patient in our case had a long-segment DVT but the blood continued to flow through the thrombus. IVC syndrome did not develop because the cannula-shaped thrombus was able to mitigate the passive venous congestion in the IVC.
      TTE has low sensitivity in detecting cannula-associated DVT when the cannulas are in situ during VA-ECMO support in this patient. Daily TTE was not able to confirm presence of the cannula-associated DVT while this patient was receiving VA-ECMO support. However, a long-segment thrombus was found immediately after removal of the cannula. The thrombus copied the shape of the cannula that had been removed. TTE can provide clear images of the cannula-associated DVT after removal of the cannula. The optimal treatment of long-segment IVC thrombosis after discontinuation of ECMO has yet to be determined. Options include anticoagulation as sole therapy, thrombolytic therapy, mechanical intervention, and surgical removal. In our case, implanting an IVC filter was challenging because the massive DVT had expanded above the level of the hepatic vein ostium. Surgical removal was not considered an appropriate first-line therapy because the patient was critically ill. Thrombolytic therapy carries a high risk of bleeding. Therefore, anticoagulation alone was considered the optimal choice for the treatment of the cannula-associated DVT in this patient. After the patient was treated solely with UFH, TTE assessment did not detect any residual thrombotic material after 10 days.
      Cannula-associated DVT is often underdiagnosed because most affected patients do not present with characteristic clinical features. Moreover, TTE does not show clear images of the thrombus during ECMO support. However, cannula-associated DVT can be detected by TTE immediately after removal of the cannula. UFH anticoagulation alone may be the optimal treatment choice in certain critically ill patients, such as the patient described in our case.

      Reverberations

      • 1.
        Cannula-associated DVT is a complication that occurs frequently after removal of VV-ECMO. However, a long-segment cannula-associated DVT is rare, especially in patients who receive VA-ECMO.
      • 2.
        TTE has low sensitivity for detecting cannula-associated DVT during ECMO support, but TEE can show clear images of DVT after removal of the cannula.
      • 3.
        UFH anticoagulation alone may be the optimal treatment for cannula- associated DVT confirmed by echocardiography.

      Acknowledgments

      Financial/nonfinancial disclosures: None declared.
      Other contributions: CHEST worked with the authors to ensure that the Journal policies on patient consent to report information were met.
      Additional information: To analyze this case with the videos, see the online article.

      Supplementary Data

      • Loading ...

      References

        • Annich G.M.
        • Zaulan O.
        • Neufeld M.
        • et al.
        Thromboprophylaxis in extracorporeal circuits: current pharmacological strategies and future directions.
        Am J Cardiovasc Drugs. 2017; 17: 425-439
        • Cooper E.
        • Burns J.
        • Retter A.
        • et al.
        Prevalence of venous thrombosis following venovenous extracorporeal membrane oxygenation in patients with severe respiratory failure.
        Crit Care Med. 2015; 43: e581-e584
        • Parzy G.
        • Daviet F.
        • Persico N.
        • et al.
        Prevalence and risk factors for thrombotic complications following venovenous extracorporeal membrane oxygenation: a CT scan study.
        Crit Care Med. 2020; 48: 192-199
        • Fisser C.
        • Reichenbächer C.
        • Müller T.
        • et al.
        Incidence and risk factors for cannula- related venous thrombosis after venovenous extracorporeal membrane oxygenation in adult patients with acute respiratory failure.
        Crit Care Med. 2019; 47: e332-e339
        • Trudzinski F.C.
        • Minko P.
        • Rapp D.
        • et al.
        Runtime and aPTT predict venous thrombosis and thromboembolism in patients on extracorporeal membrane oxygenation: a retrospective analysis.
        Ann Intensive Care. 2016; 6: 66
        • Bein T.
        • Philipp A.
        • Pregler B.
        • et al.
        Long-segment caval thrombus after removal of ECMO cannula.
        Intensive Care Med. 2015; 41: 1967-1968