Researchers identify transfusion threshold that confers lower mortality risk in VV ECMO
Click Here to Manage Email Alerts
Packed red blood cell transfusion improved survival among adults with acute respiratory distress syndrome on venovenous extracorporeal membrane oxygenation when using a hemoglobin threshold of less than 7 g/dL, according to study results.
“In everyday life now there is enough evidence to say that transfusions should be administered just when hemoglobin drops to a value lower than 7 g/dL,” Gennaro Martucci, PhD, from the department of anesthesia and intensive care at Istituto Mediterraneo per i Trapianti e Terapie ad Alta Specializzazione (ISMETT) in Palermo, Italy, told Healio. “At the same time, a relevant therapeutic target that should be reached daily is the adequate blood flow to oxygenate the patient. This goal should not be reached at the price of more positive fluid balance.”
In PROTECMO, a multicenter, prospective observational cohort study published in The Lancet Respiratory Medicine, Martucci and colleagues evaluated 604 adults (mean age, 50 years; 71% men) with ARDS receiving venovenous (VV) ECMO between Dec. 1, 2018, and Feb. 22, 2021, to evaluate practices used regarding hemoglobin thresholds for packed red blood cell (PRBC) transfusion.
Patients came from 41 ECMO centers across Europe, North America, Asia and Australia, for which researchers classified as having low volume (one to 11 runs/year), medium volume (12 to 20 runs/year) and high volume (more than 20 runs/year) of ECMO activity a year.
Researchers used marginal structural models with inverse probability weighting to evaluate the effect of PRBC transfusion at certain thresholds with adjustments for baseline and time varying confounding factors.
Results
At cannulation, the mean hemoglobin concentration was 10.9 g/dL, decreasing to 9.1 g/dL over 7,944 days on ECMO.
After stratification, researchers found lower average hemoglobin concentrations in North America than in Europe (difference, –0.66 g/dL; P < .0001) and in high-volume vs. low-volume centers (difference, – 0.26 g/dL; P = .022).
Of the total ECMO days, PRBC transfusion occurred on 2,432 days (31%), with 504 (83%) patients receiving at least one unit. The median amount of PRBC transfused per day was 115 mL (interquartile range [IQR], 48-223), increasing to 425 mL (IQR, 350-556) during days with a transfusion.
Researchers found that more than half of the transfusions occurred due to low hemoglobin (63%). Other reasons included bleeding (23%), hemodynamic impairment (9%), low ECMO blood flow rate (3%) and other (3%).
Researchers identified a mean pretransfusion hemoglobin concentration of 8.1 g/dL but noted that this concentration differed based on clinical rationale prompting the transfusion.
They also found that more transfusions were given on days when concentration was less than 7 g/dL than on days when concentration was greater than 10 g/dL (71% vs. 21%), with higher thresholds more commonly used for older patients, patients with new-onset kidney failure and those with reduced extracorporeal blood flow rate to estimated cardiac output ratio.
Analyses showed that a higher recurrence rate of transfusions was linked to a positive fluid balance, higher cardiovascular item of the sequential organ failure assessment score, higher pH, bleeding and ECMO circuit change.
Mortality risk
Overall, 59% of patients survived to ICU discharge and 58% survived to hospital discharge.
Researchers observed an increased risk for ICU mortality at 28 days with daily hemoglobin concentrations less than 7 g/dL (HR = 2.99; 95% CI, 1.95-4.6) compared with higher concentrations in a time-dependent Cox model.
Conversely, researchers noted a decreased risk for ICU death when blood was transfused with a hemoglobin concentration less than 7 g/dL (HR = 0.15; 95% CI, 0.03-0.74). Researchers identified 7.2 g/dL as the highest hemoglobin concentration used as a threshold to prompt PRBC transfusion that reduced risk for death.
“The study findings were quite unexpected since we could imagine that, just as other critically ill patients, a higher trigger was associated with higher mortality, but we would not imagine the ‘normal’ trigger is valid also for ECMO patients,” Martucci told Healio. “Moreover, this was demonstrated with several different statistical methods.
“Future interventional studies can compare different strategies and will have to consider also the other determinants of transfusions,” Martucci added.
In an accompanying editorial, Joseph E. Tonna, MD, MS, FCCM, FACEP, FAAEM, associate professor in the divisions of cardiothoracic surgery and emergency medicine at University of Utah, wrote that these findings suggest a hemoglobin threshold of 7 g/dL should be used to trigger transfusion during routine practice, but there may be cases on an individual patient level that should trigger transfusion at higher thresholds, such as refractory hypoxemia and ischemia.
“Predicting how well or poorly individual patients will respond to blood transfusion remains imperfect and is the art of critical care,” Tonna wrote. “Individual situations require individualized assessments and thoughtful, responsive interventions. The PROTECMO study by Martucci and colleagues should certainly inform routine care for patients receiving VV ECMO. It is a high-quality analysis and suggests that there is little benefit to routine blood transfusion when the hemoglobin concentration is more than 7 g/dL, even for adults on VV ECMO.”
For more information:
Gennaro Martucci, PhD, can be reached at gmartucci@ismett.edu.