Adverse Effects of RBC Transfusions: A Unifying Hypothesis

Transfusion of red blood cells (RBCs) is often effective at preventing morbidity and
mortality in anemic patients. In contrast, recent studies indicate that some RBC components
may have functional defects ("RBC storage lesions") that actually cause morbidity and
mortality when transfused. For example, patients transfused with RBCs stored >14 days have
statistically worse outcomes than those receiving "fresher" RBC units. In addition to the
age of stored RBCs, the volume transfused may be important. The TRICC study showed that
specific patients whose transfusions were limited by a restrictive trigger (RBCs
transfusions only when hemoglobin [Hb] < 7 g/dL) had significantly better outcomes than
those transfused with a more liberal trigger ([Hb] < 10 g/dL Hb). This finding has been
particularly difficult to understand since conventional wisdom suggests that an elevated
[Hb] should be beneficial because it supports increased O2 delivery. Recipient-specific
factors may also contribute to the occurrence of these adverse events. Unfortunately, these
events have been difficult to investigate because up to now they have existed only as
"statistical occurrences" of increased morbidity and mortality in large data sets. There
are currently no clinical or laboratory methods to detect or study them in individual
patients.

The microcirculation is composed of a continuum of small vessels including small arterioles,
capillaries, and post-capillary venules. The microcirculation represents an
actively-adjusting vascular circuit that matches blood flow (and O2 delivery) to local
tissue oxygen demands. While the physiologic mechanisms that match O2 delivery to local
requirements are incompletely understood, endothelium-derived nitric oxide (NO) clearly
plays an important role. Interestingly, recent work has revealed that in addition to
transporting O2 and CO2, the RBC also controls local NO concentrations and thus may also
play a surprisingly important role in regulating blood flow in the microcirculation.

Herein, we bring together previously unconnected data to propose a unifying hypothesis,
centered on insufficient NO bioavailability (INOBA), to explain the increased morbidity and
mortality observed in some patients following RBC transfusion. In this model, variables
associated with RBC units (storage time; 2,3-DPG concentration) and transfusion recipients
(endothelial dysfunction; hematocrit [Hct]) collectively lead to changes in NO levels in
vascular beds. Under certain circumstances, these variables are "aligned" such that NO
concentrations are markedly reduced, leading to vasoconstriction, decreased local blood flow
and insufficient O2 delivery to end organs. Under these circumstances, the likelihood of
morbidity and mortality escalates. The INOBA hypothesis is attractive because of its
explanatory power and because it leads to a number of readily testable predictions, which
will be investigated in the following 3 Aims:

Aim 1: To investigate the effects of blood processing and storage (using standard
FDA-approved conditions) on NO production and scavenging by human RBCs/Hb in vitro. Using
sensitive biochemical assays (electron spin resonance [ESR]) and a rat aortic ring in vitro
bioassay, we will test the effects of RBC storage time, leukoreduction, and irradiation on
NO synthesis and/or scavenging by intact RBCs and free Hb. Modifications such as washing
and rejuvenation will be investigated as possible approaches to correct abnormalities in NO
bioavailability.

Aim 2: To transfuse healthy volunteers and investigate the effects of storage-related RBC
changes on blood flow, tissue oxygenation, and biomarkers of cardiovascular function. We
will determine whether RBCs prepared and stored under conditions that alter NO
bioavailability in vitro (Aim 1) inhibit NO-mediated vasodilation, reduce tissue perfusion,
and decrease tissue O2 delivery in healthy transfusion recipients in vivo. The role of
2,3-DPG depletion as well as exercise-induced O2 demand will also be investigated with these
specialized experimental systems.

Aim 3: To determine the effects of transfused RBCs in patients with endothelial dysfunction
due to cardiovascular disease. A non-invasive ultrasound assay will be used to test whether
patients with cardiovascular disease and endothelial dysfunction (who have intrinsic defects
in NO synthesis) are more susceptible to adverse effects from stored/processed RBCs
(impaired in NO bioavailability) than fresh RBC units. Vasodilation and tissue oxygenation
in response to transfusion will be monitored, and Framingham risk scores and cardiovascular
disease biomarkers will be tested as potential predictive factors to identify patients most
at risk from adverse effects of RBC transfusions.

The proposed studies will not only carefully evaluate each tenet of this unifying hypothesis
of the RBC storage lesion, but will also lead to potential remediations including altered
RBC storage/processing conditions and media, novel transfusion recipient screening, and
improved RBC/recipient matching.

Inclusion Criteria:

- Healthy male or female volunteers (age 21-60 years) and subjects with greater than
two cardiovascular risk factors (hypercholesterolemia, diabetes, hypertension, and
tobacco smoking) or known cardiovascular disease will be invited to participate in
the study.

- Healthy subjects: Healthy male or female volunteers (aged 21-60 years).

- Hypercholesterolemia: Defined as serum low density lipoprotein cholesterol > 140
mg/dL if not currently on lipid-lowering therapy or > 100 mg/dL if on lipid-lowering
therapy.

- Diabetes: Defined as having fasting blood glucose sample of > 126 mg/dL or a
hemoglobin A1c of > 7% or being treated with diabetes medications such as oral
hypoglycemic agents, insulin sensitizing agents, or subcutaneous insulin.

Patients with diabetes will be asked to hold their oral diabetes medications starting the
evening before the study and withhold insulin injections starting the morning of the
study.

- Smoking: active tobacco use, 20 cigarettes per day for the past year.

- Hypertension: Blood pressure of > 140/90 or currently on anti-hypertensive
medications

- Cardiovascular disease: known coronary artery disease by angiogram or documented
myocardial infarction.

Exclusion Criteria:

- Presence of intercurrent illness or other chronic diseases

- Age <21 or > 80 years

- Pregnancy or menopause

- Renal failure (creatinine>1.4 mg/dl)

- Allergies to aspirin

- Bleeding disorders

- Uncontrolled hypertension with BP > 180 mmHg systolic and > 120 mmHg diastolic

- Acute infection in previous 4 weeks

- History of substance abuse

- Liver failure (Liver enzymes >2x normal)

- Inability to give informed consent Inability to return to Emory for follow-up