Blood Transfusion Volumes Linked to Infection in Cardiac Surgery Patients

Author: Naveed Saleh MD, MS

Anesthesiology News

Blood transfusion rates remain high, despite higher transfusion volumes being linked to infection in patients undergoing cardiac surgery.

In research presented at the 2019 annual meeting of the Society of Cardiovascular Anesthesiologists (SCA), investigators examined blood transfusion rates and other risk factors to assess the correlation between infectious and other adverse outcomes in cardiac surgery.

In the study, the Society of Thoracic Surgery database was mined and yielded 2,458 patients who underwent coronary artery bypass graft surgery or valvular surgery at North Shore University Hospital, in Manhasset, N.Y., between mid-2015 and 2017. The primary outcome was a composite of infection conditions. Secondary outcomes included a composite of morbidity (renal, myocardial and pulmonary events), infectious outcomes, and death.

“The most common postoperative infections include pneumonia, sternal wound infection, sepsis and conduit harvest site infection,” said study author Linda Shore-Lesserson, MD, the director of cardiac anesthesiology at North Shore and past president of the SCA. “Our composite infectious outcome was defined as having any one of these infectious complications. The incidence of each of these alone is rather low. Thus, in order to statistically increase our chances of detecting an infectious outcome, we defined the composite endpoint as having any one of the [aforementioned] infectious outcomes.”

Initially, a univariate analysis was performed using demographic, preoperative and intraoperative parameters (including transfusion) to identify associations with the composite infectious outcome, Dr. Shore-Lesserson explained. Based on the literature, a transfusion of four units or more of red blood cells (RBCs) was defined as large. Next, a multivariable analysis was performed that included the significant univariate parameters (P<0.1) to identify independent associations with infection.

Overall, four units of RBCs or more were transfused in 23% (n=568) of patients. Of these patients receiving a large transfusion, the composite infectious outcome was present in 3.5% (n=85) and the composite of all adverse events occurred in 17%.

Dr. Shore-Lesserson noted that her team’s finding of an association between infectious outcomes and increased transfusion volumes is consistent with what is already documented in the cardiac and orthopedic literature. “Transfusion is known to be immune-modulating and can increase the risk of infectious outcomes. In a multivariate analysis, we demonstrated in our own patient population that transfusion of four or more RBC units was independently associated with infection outcomes.”

Specifically, patients who were given four units of RBCs or more were five times more likely to experience an infectious outcome than those who received less.

In the study, anemia was not associated with composite infectious outcomes. Dr. Shore-Lesserson pointed out that this finding is often confusing in the literature. “The question is whether anemia is itself a risk factor for adverse events or whether it is the transfusion that is used to treat anemia that is the morbid event. Anemia itself has been shown to be an independent marker for morbid outcomes, such as central nervous system dysfunction and renal failure. However, since our investigation centered around infection, our findings make sense,” she said.

Of note, interventions can be made to potentially reduce the risk for an infectious complication, according to Dr. Shore-Lesserson. “Two of the risk factors associated with infectious outcomes in our study can be modified. Transfusion can be modified by institution of patient blood management strategies. Preoperative anemia can be addressed and treated, surgical technique can be modified to include hemostatic techniques, and prophylactic medication such as antifibrinolytic agents can be administered. Transfusion thresholds and algorithms can be created that follow medical society standards and thus withhold transfusion until those thresholds are reached. Finally, the low temperature risk factor can be modified by avoiding extremes of hypothermia during cardiopulmonary bypass,” she said.

Of note, hypothermia increases the potential need for a transfusion via two mechanisms. “One is that hypothermia itself impairs coagulation enzyme function and contributes to coagulopathy. The second is that hypothermic patients require additional time on cardiopulmonary bypass for warming. Additional time on cardiopulmonary bypass increases the risk for coagulopathy,” Dr. Shore-Lesserson explained.

Future prospective studies should evaluate the effect of patient blood management on outcomes in cardiac surgery patients, particularly b ecause blood conservation metrics may soon be reported in national quality database registries.

In a separate interview with Anesthesiology News, Marie-Louise Meng, MD, an assistant professor of anesthesiology at NewYork-Presbyterian Hospital/Columbia University Medical Center, in New York City, commented on the importance of the study. “The increased risk of infection in patients who receive multiple transfusions is clear. These authors clearly highlight the risk factors—advanced age, diabetes and hypothermia—that increase the risk of infection in patients undergoing cardiac surgery. It behooves the practitioner to avoid unnecessary blood transfusions when possible by employing point-of-care testing and blood conservation strategies.”