SCCM is updating its SCCM Connect Community. Access to SCCM Connect may be limited until April 23.

Providing Extracorporeal Membrane Oxygenation to a Jehovah’s Witness

visual bubble
visual bubble
visual bubble
visual bubble

Ashley R. Hurst, JD, MDiv; Benjamin A. Moses, MD, MS

A 37-year-old man had a witnessed seizure while picking up his child from school. The seizure was complicated by closed head trauma and loss of consciousness. He subsequently had a cardiopulmonary arrest. Basic life support was initiated while emergency medical services (EMS) were en route to the emergency department (ED). On arrival at the ED, EMS identified pulseless ventricular tachycardia, progressing to asystole. Advanced cardiac life support was initiated, and three rounds of epinephrine were administered. The patient developed ventricular fibrillation (VF), with return of spontaneous circulation (ROSC) after electrical cardiac defibrillation. He was transported to a level 1 ED with persistent hypoxia (pulse oximetry 50%). After a second episode of VF after ROSC, he was intubated, with subsequent red frothy sputum emanating from his endotracheal tube, concerning for pulmonary edema. Electrocardiographic tracings were consistent with an anterior ST segment elevation myocardial infarction as the etiology of his cardiac arrest. Initial lactate level was 9.0 mmol/L. Head CT demonstrated no acute intracranial hemorrhage. Initial course was complicated by limited improvement in oxygenation due to patient dyssynchrony with the ventilator.

He underwent emergent cardiac catheterization with angioplasty, stenting of his left anterior descending artery, and placement of an intra-aortic balloon pump for acute cardiogenic shock. He was admitted to the coronary care unit, where he was cooled and paralyzed to assist in ventilation. He became increasingly unstable and difficult to ventilate, consistent with severe acute respiratory distress syndrome (ARDS). A pulmonology consultation helped manage his hypoxemia. Bronchoscopy did not reveal evidence of significant aspiration or mucus plugging, and extracorporeal membrane oxygenation (ECMO) was considered. Based on the patient’s condition, the medical team felt that he required venoarterial (VA) ECMO for any chance of survival. However, the family said that the patient was a Jehovah’s Witness and that his prior wishes were to not receive blood products. He had no advance directive and had not had a prior conversation with his family regarding end of life. Faced with his imminent death, his family agreed to VA ECMO but declined blood products, a routine component of VA ECMO therapy.

Despite initiation of ECMO, the patient’s lactic acidosis worsened persistently over the next 10 hours. Interval head CT demonstrated a catastrophic anoxic brain injury. His already poor prognosis and the new radiographic findings were discussed with his family in the presence of one of his church leaders. His family elected to discontinue life-sustaining therapies. He died within minutes of ceasing cardiopulmonary support. Consistent with his family’s declaration of his wishes, he never received blood products during his admission. Although his survival was not likely to have been more certain with administration blood products, some staff questioned the appropriateness of using VA ECMO without the ability to provide potentially necessary transfusions.

For decades, Jehovah’s Witnesses’ (JW) religious prohibition against certain blood products has presented an ethical quandary for clinicians. Indeed, JWs’ refusal of blood transfusions is a classic medical ethics case that highlights the tension between respect for autonomy and beneficence. However, what is often obscured by this ethical focus is the medical advancements clinicians have achieved by working with JWs to respect both their religious beliefs and their desire for the same level of lifesaving interventions offered to non-JWs. For example, the field of bloodless surgery, once seen as merely an accommodation for a non-mainstream religion, has informed surgical practice and benefited JWs as well as non-JWs expressing objections to blood transfusions. Yet, despite decades of accommodation and clinical advancement, the question remains: Are all interventions appropriate to offer JWs regardless of their associated need for blood transfusions? This VA ECMO case highlights this issue.

Experience exists with bloodless use of ECMO, particularly in isolated ARDS;1 however, VA ECMO generally requires blood transfusion due to the need for anticoagulation,2 the mechanical damage the circuit causes red blood cells, and the ongoing need to optimize oxygen delivery to damaged tissue by augmenting red cell mass.3 Given the high likelihood of blood transfusions while on VA ECMO, when, if ever, is it medically appropriate to use this intervention with JWs who refuse blood transfusions? Are clinicians offering a resource-intensive, aggressive intervention with too low a likelihood of success?

Few reported cases demonstrate the successful use of bloodless ECMO.1,4,5 Despite defying clinical expectations, these few cases alone cannot provide sufficient reason to use VA ECMO without regard for the complexities of a JW patient. This case, however, highlights additional considerations.

Here, the issue of blood transfusion on VA ECMO did not arise because the patient’s devastating neurologic injury was discovered shortly after VA ECMO was started. Unlike the two reported cases, this patient did not recover. Instead, VA ECMO allowed for further diagnostic tests that gave the family and clinicians a more informed picture of the patient’s condition. With this information, the family was able to make a decision to withdraw life-sustaining interventions, a decision they believed was consistent with the patient’s values. The use of VA ECMO in this case did not compromise their religious values or clinical standards of care.

Survival is only one measure of the successful use of VA ECMO (or any intervention). The unilateral decision not to offer VA ECMO deprives JW patients and families of options that others have. In this case and other reported cases, because the clinical team did not see the patient’s beliefs as an insurmountable barrier to VA ECMO, the intervention was used as it would have been used for non-JW patients: 1) to restore function and health and 2) to allow for crucial diagnostic tests to inform decision-making. If clinicians routinely and unilaterally decide not to offer VA ECMO based solely on blood transfusion refusal, then JW patients and families may be deprived of valuable clinical data they need to make sound decisions, some successful outcomes will be foreclosed, and an opportunity to advance practice will be lost. Ultimately, the questions to be asked when deciding whether to offer VA ECMO to JW patients are the same as with non-JW patients: what are the attainable outcomes, what are the risks and benefits, what harm is avoided or caused by the intervention? Asking these questions, instead of whether clinicians should offer VA ECMO to a JW, allows for different answers based on patients and circumstances, which is what JWs are asking from the medical community.6


  1. Preston TJ, Olshove VF Jr, Chase M. Bloodless extracorporeal membrane oxygenation in the Jehovah’s Witness patient. J Extra Corpor Technol. 2012 Mar;44(1):39-42.
  2. Agerstrand CL, Burkart KM, Abrams DC, Bacchetta MD, Brodie D. Blood conservation in extracorporeal membrane oxygenation for acute respiratory distress syndrome.. Ann Thorac Surg. 2015 Feb;99(2):590-595.
  3. Standards for hemoglobin levels in VA ECMO. In: Annich GM. ECMO: Extracorporeal Cardiopulmonary Support in Critical Care. 4th ed. Ann Arbor, MI: Extracorporeal Life Support Organization; 2012.
  4. Lindholm J, Palmer K, Frenckner B. Long-term ECMO treatment in Jehova’s Witness patient without transfusions. Perfusion.  2012 Jul;27(4):332-334.
  5. Russo MJ, Merlo A, Eton D, et al. Successful use of ECMO in a Jehova’s Witness after complicated re-heart transplant. ASAIO J. 2013 Sep-Oct;59(5):528-529.
  6. Doyle DJ. Blood transfusions and the Jehovah’s Witness patient. Am J Ther. 2002 Sep-Oct;9(5):417-424.