Clinical,Ethical,and Financial Considerations With Red Blood Cell Transfusion Support at the End of Life in Hematologic Malignancies

RBC Transfusion Support Alleviates Symptoms at the EOL

There is significant evidence suggesting the benefits of RBC transfusions at the EOL. ¹³ However, the level of supporting evidence is low given the absence of randomized studies, which are admittedly very difficult due to ethical concerns, hence their absence in the literature. ¹⁴ Additionally, objective measures correlating with patient-reported improvements in symptoms with RBC transfusion are lacking, which can also make the benefits of RBC transfusions difficult to appreciate.^15,16 Irrespective of this, the subjective improvement in symptoms with RBC transfusions has been noted in multiple studies in up to 70% of patients, though these benefits may not be durable.^14,15,17,18

Orme et al. explored the experiences of hematology patients in the UK receiving RBC transfusions in a day hospice setting, who reported improvements in fatigue and dyspnea with transfusions. ¹³ Similarly, a prospective cohort study of 97 patients with advanced malignancies, including hematologic malignancies, evaluated the impact of transfusions on dyspnea, weakness, and overall sense of well-being using visual analog scales with total scores ranging from 0 to 10. Scores reported by patients post-transfusion reflected improvements in dyspnea, weakness, and overall well-being. ¹⁹

Despite the time commitment and risk of complications, patients and caregivers alike endorse that transfusions improve energy, allow patients to engage in desired activities, and ultimately enhance QOL during their remaining time. ²⁰ With this in mind, the 2020 American Society of Hematology guidelines for AML in older patients included a conditional recommendation for RBC transfusions for older adults with AML not receiving anti-leukemic therapy (EOL or hospice care). ²¹

There is a clear benefit from a symptom perspective with transfusions at the EOL, but it is worth noting that providing such services can be intensive. For patients and caregivers, transfusion services are not routinely provided at home, so they must travel back and forth to the clinic, introducing the concept of time toxicity.^22,23 Transfusions at the EOL also have the potential to 1) increase strain on transfusion services for cross-matching blood, as this can become more challenging with more transfusions over time, and 2) require greater manpower from a nursing perspective to administer transfusions, in some cases outside of regular hours, and monitor transfusion reactions. ²⁴

Limited Access to Transfusions Negatively Impacts EOL Care

Limited access to RBC transfusions at EOL delays hospice enrollment, increases the rates of hospital utilization, chemotherapy use, and death in the hospital away from home.^3,18,25 In a cross-sectional survey of hospices in the US, only 2.7% reported always providing RBC transfusions, 40.7% sometimes offered transfusions, and 54.9% never offered transfusions. ⁹ For those who ultimately enroll in hospice, the limited access to transfusions can result in a shortened length of stay of 6 vs 11 days for transfusion-dependent vs transfusion-independent patients, respectively. ²⁶

Hospice utilization improves EOL by reducing aggressive care. Among Medicare beneficiaries enrolled in hospice, 96% were less likely to die in an inpatient setting, 44% were less likely to have an ICU stay in the last 30 days of life, and 62% were less likely to receive chemotherapy in the last 14 days of life. There was also a 38% decrease in mean Medicare spending at the EOL for those on hospice. ²⁵

In contrast, in a retrospective study of patients from the Surveillance, Epidemiology, and End Results (SEER) cancer registry with myelodysplastic syndrome from 2006-2012 who died, transfusion-dependent patients were more likely to be admitted to the ICU within the last 30 days of life, receive chemotherapy within the last 14 days of life, and were less likely to enroll in hospice. ²⁷ Similarly, in a retrospective study of AML patients using SEER-Medicare linked data, fewer than half of the population enrolled in hospice, and the increase was driven by late enrollment, with 47.3% and 28.8% enrolling in the last 7 and 3 days of life, respectively. ²⁸

Transfusions Are Finite Resources

Donors altruistically provide RBC units with an unpredictable supply from year to year, so blood may be perceived as a finite resource. ¹⁵ For example, during the COVID-19 pandemic, mandated quarantines and widespread fear of visiting healthcare facilities led to a decline in blood donations and blood product shortages. ²⁹

The ethical principles of autonomy, non-maleficence, and beneficence at the individual patient level have been noted to be important to balance with justice for all patients when considering transfusions at the EOL, given the finite nature of RBCs. ¹⁵ Some also raise concerns about futility in allocating RBCs at the EOL, given their scarcity. Futility is a challenging concept, as perspectives of all parties involved may differ; thus, in clinical practice, it may be necessary to involve an ethics committee to help make decisions. ³⁰

Given the finite nature of blood and the futility argument raised, some have proposed guidelines to assist in decision-making regarding RBC transfusions at the EOL for patients with anemia-related symptoms.^30-32 Smith et al. proposed the following guidelines: limit transfusions to the minimum necessary to provide symptom relief, review cases requiring >2 transfusions per week with transfusion medicine, and defer transfusions for patients at the EOL if products are scarce. ³⁰ Overall, careful thought and consideration should be given when administering RBC transfusions, given the finite and scarce nature of blood.

Transfusions Are Associated With Risks and Are Contradictory to Hospice Care

Transfusions can be associated with mild allergic reactions or life-threatening complications such as transfusion-related acute lung injury, circulatory overload, and anaphylaxis. These potential complications of varying severity and some of which can be life-threatening, in a population for which a traditional risk assessment may not be accurate, given medical complexity often with poor nutrition, co-morbidities, frailty, and deconditioning, are of concern.^13,15 As a result, some hospice agencies decide that these risks outweigh the benefits and choose not to offer transfusions, but the frequency of these transfusion reactions is debatable, and pilot studies of transfusions at home have not noted such reactions.^33,34

Secondly, given the scarcity of high-quality evidence on the benefits of EOL transfusions and the subjectivity of what constitutes a good death, some hospice services believe the provision of transfusions is inconsistent with hospice care.^4,17