Drug Sequestration in the Management of Patients with Extracorporeal Membrane Oxygenation Support

Dosing recommendations for patients requiring ECMO support are unlikely to be supported by high-quality evidence, but are usually extrapolated from the physicochemical characteristics of individual drugs in conjunction with data from pharmacokinetic studies and case reports.

Drug sequestration by an extracorporeal membrane oxygenation (ECMO) circuit is a well-recognized complication in critically ill patients who require ECMO for advanced hemodynamic and respiratory support. This drug phenomenon depends on the circuit component materials and the drug physicochemical properties.

Critical care pharmacists play a key role in recognizing this phenomenon and can make significant dosing and monitoring adjustments. Participating in these modifications may potentially minimize the likelihood of therapeutic failure, drug toxicity, and/or antimicrobial resistance.

This review will describe pharmacokinetic considerations, factors that enhance drug-related sequestration within the ECMO circuit, optimal drug dosing strategies, and demonstrate the information using a unique patient case.

Drug sequestration or removal in the circuit depends on certain factors, including the potential interaction of drugs to various plastic components of the ECMO system. The type of circuit may also be a variable predictor influencing the level of drug sequestration.

The tubing of the ECMO circuit, made of polyvinyl chloride, accounts for most of the drug sequestration. In terms of drug properties, lipophilicity, molecular size, ionization, and protein-binding enhance the degree of drug sequestration in the ECMO circuit.

Typically, lipophilic, and highly protein-bound medications are significantly sequestered in the circuit. Lipophilicity is defined by the LogP value, and in general, the higher the logP value, the more lipophilic the drug is, and thus, more likely to be sequestered. Despite efforts to better categorize drugs sequestration, this phenomenon remains clinically challenging and unpredictable.1

Dosing recommendations for patients requiring ECMO support are unlikely to be supported by high-quality evidence, but are usually extrapolated from the physicochemical characteristics of individual drugs in conjunction with data from pharmacokinetic studies and case reports. In general, the pharmacokinetic effects of ECMO are less significant for drugs that can be titrated to clinical endpoints, such as vasopressors or inotropes. However, in drugs for which this is not possible, dosing should be guided by therapeutic drug monitoring to ensure efficacy while minimizing toxicity.1

A 61-year-old man with a history of chronic myeloid leukemia undergoing long-term therapy with imatinib, chronic obstructive pulmonary disease, untreated obstructive sleep apnea, stage 3 chronic kidney disease, type 2 diabetes mellitus, and history of polysubstance use, including cocaine and tobacco, presented to an outside hospital with 3 days of fevers, myalgias, cough, and dyspnea on exertion.

On initial presentation, he was febrile to 102.5° F, not requiring supplemental oxygenation, and was euvolemic. Imaging was suggestive of a right lower lobe pneumonia, and the patient was admitted to the medical floors and initiated on ceftriaxone and azithromycin for presumed community-acquired pneumonia.

On day 2 of hospitalization, the patient experienced increased respiratory distress and developed hypoxemia requiring nasal cannula up to 5 L, and was persistently febrile up to 103° F. A computed tomography (CT) of the chest was performed showing a dense consolidation in the right lower lobe along with new, smaller areas of consolidation in the left lower and upper lobes, and antibiotics were changed to ceftriaxone and vancomycin.

The patient was transferred to the intensive care unit and intubated in the setting of worsening hypoxia by day 3 of hospitalization. Antibiotics were broadened further to vancomycin, cefepime, and azithromycin. Vancomycin was later changed to linezolid in the setting of acute kidney injury with decreased urine output and serum creatinine increasing from 2.4 on admission to 4.1 on day 3 of hospitalization.

On day 4 of hospitalization, the patient was intubated as his respiratory status continued to decline. Throughout the day, he became progressively more difficult to ventilate, and concern for acute respiratory distress syndrome (ARDS) developed.

The patient also became hypotensive requiring initiation of a norepinephrine infusion. Given concern for worsening ARDS and academia, the patient was transferred to our medical ICU for advanced care. At the time of transfer, norepinephrine reached maximum dosing, requiring the addition of vasopressin.

Phenylephrine was also briefly required for a MAP < 65 and he was given 100 mg hydrocortisone in transit. Imatinib was continued for the duration of hospitalization at the outside hospital.

The patient arrived intubated and sedated and was admitted to the BIDMC Medical Intensive Care Unit, where broad-spectrum antibiotics were continued. Ventilator settings were optimized using lung protective strategies for ARDS, the patient was initiated on inhaled epoprostenol, paralysis was induced with cisatracurium, and prone positioning was utilized.

Despite these interventions, no improvement in his ARDS was noted. Imatinib was discontinued and corticosteroids were initiated out of concern for imatinib-induced pneumonitis.

On day 6 of hospitalization, the patient was initiated on veno-venous extracorporeal membrane oxygenation (VV-ECMO). When hospitalization reached day 7, respiratory viral antigen screen returned negative for all viruses, but positive for adenovirus, with a viral load greater than 2 million copies/mL.

Upon identification of adenovirus, a decision was made to treat with cidofovir, and a dose of 5 mg/kg was given on day 8 of hospitalization.3 A day later, the decision was made to give a dose of intravenous immunoglobulin (IVIG) at a dose of 0.5 g/kg.4

Probenecid was administered at the standard 2 grams, 3 hours prior to cidofovir administration, and two 1-gram doses at 2 and 8 hours following cidofovir infusion. However, as the patient was on CRRT, additional hydration was not given, and CRRT was run even for the duration of cidofovir infusion.

Given the fact that the patient was anuric prior to initiating treatment with cidofovir, assessing the degree of nephrotoxicity attributable to cidofovir was difficult. Clinical improvement was seen after the first dose of cidofovir with discontinuation of paralysis on day 8 and a complete wean of inhaled epoprostenol on day 9. The patient’s white blood cell counts also decreased from 19.0 to 15.0 x109/L.

On day 15 of hospitalization, the patient was decannulated from VV-ECMO, the WBC further decreased to 9.5 x109/L, and the adenovirus viral load decreased to less than 500 copies/mL. A second infusion of cidofovir was given at a reduced dose of 3 mg/kg on day 16 of hospitalization.

One week later, the patient began to have signs of renal recovery, given his residual urine output, and CRRT was then stopped. By day 25, the patient was extubated, and his viral load remained undetectable.

This case highlights the potential role for cidofovir in ECMO patients who are refractory to conventional management and sheds light on optimal dosing strategies. Cidofovir is a hydrophilic drug with a LogP value of -3.6 with limited protein binding (<6% protein bound).

Given this pharmacokinetic profile, it was anticipated that the degree of sequestration in the ECMO circuit would be low and thus, used the standard dosing recommended in non-ECMO patients. Although no optimal cidofovir serum level for adenovirus management has been defined, the endpoint of an undetectable DNA PCR and symptom resolution were used as markers of treatment efficacy.

Based on this case and the patient’s response, it was concluded that no dosing adjustment may be necessary. The standard dose may be like non-ECMO patients and should be guided by patient response and changes in renal function during therapy.

About the Author

George T. Abdallah, PharmD, BCCCP, BCCP, Clinical Pharmacist IV, Critical CareCardiac and Cardiovascular Intensive CareDepartment of Pharmacy, Beth Israel Deaconess Medical Center.


  1. Cheng V, Abdul-Aziz MH, Roberts JA, Shekar K. Optimising drug dosing in patients receiving extracorporeal membrane oxygenation. J Thorac Dis. 2018 Mar;10(Suppl 5):S629-S641.
  2. Lynch JP, Kajon AE. Adenovirus: epidemiology, global spread of novel serotypes, and advances in treatment and prevention. Semin Respir Crit Care Med. 2016; 37:586-602.
  3. Lindermans CA, Leen AM, Boelens JJ. How I treat adenovirus in hematopoietic stem cell transplant recipients. Blood. 2010; 116(25): 5476-5485.
  4. Permaplung N, Mahoney MV, Alonso CD. Adjunctive use of cidofovir and intravenous immunoglobulin to treat invasive adenovirus disease in solid organ transplant patients. Pharmacotherapy. 2018: 38(12):1260-1266.