Evolution of adverse changes in stored RBCs
Elliott Bennett-Guerrero*, Tim H. Veldman†, Allan Doctor‡, Marilyn J. Telen§, Thomas L. Ortel§, T. Scott Reid†¶,
Melissa A. Mulherin§, Hongmei Zhu§, Raymond D. Buck?, Robert M. Califf**, and Timothy J. McMahon§††‡‡
Departments of *Anesthesiology and§Medicine and **Duke Clinical Research Institute, Duke University Medical Center, Durham, NC 27710;†NITROX LLC,
Durham, NC 27701;‡Pediatric Critical Care, Washington University, St. Louis, MO 63110;¶Cato Research Ltd., Durham, NC 27713;?School of Nursing,
University of North Carolina, Greensboro, NC 27402; and††Durham Veterans Affairs Medical Center, Durham, NC 27705
Communicated by Irwin Fridovich, Duke University Medical Center, Durham, NC, August 28, 2007 (received for review June 13, 2007)
Recent studies have underscored questions about the balance of risk
and benefit of RBC transfusion. A better understanding of the nature
and timing of molecular and functional changes in stored RBCs may
provide strategies to improve the balance of benefit and risk of RBC
transfusion. We analyzed changes occurring during RBC storage
focusing on RBC deformability, RBC-dependent vasoregulatory func-
(Hb) O2desaturation is coupled to regional increases in blood flow in
healthy volunteers was processed into leukofiltered, additive solu-
tion 3-exposed RBCs and stored at 1–6°C according to AABB stan-
dards. Blood was subjected to 26 assays at 0, 3, 8, 24 and 96 h, and at
1, 2, 3, 4, and 6 weeks. RBC SNO-Hb decreased rapidly (1.2 ? 10?4at
3 h vs. 6.5 ? 10?4(fresh) mol S-nitrosothiol (SNO)/mol Hb tetramer
(P ? 0.032, mercuric-displaced photolysis-chemiluminescence assay),
and remained low over the 42-day period. The decline was corrobo-
rated by using the carbon monoxide-saturated copper-cysteine assay
[3.0 ? 10?5at 3 h vs. 9.0 ? 10?5(fresh) mol SNO/mol Hb]. In parallel,
ability assayed at a physiological shear stress decreased gradually
over the 42-day period (P < 0.001). Time courses vary for several
storage-induced defects that might account for recent observations
linking blood transfusion with adverse outcomes. Of clinical concern
is that SNO levels, and their physiological correlate, RBC-dependent
even ‘‘fresh’’ blood may have developed adverse biological
adenosine triphosphate ? hemoglobin ? nitric oxide ? S-nitrosothiols ?
licensed only on the basis of the procedures used for collection,
processing, and storage. Mandated, specific testing ensures safety
from infectious diseases and compatibility between blood product
specifying the clinical outcomes constituting an effective transfu-
sion. In addition, RBC transfusion has not been subjected to the
RBCs may be stored for up to 42 days under controlled condi-
tions before transfusion. However, numerous changes occur in
RBCs during storage (collectively referred to as the ‘‘storage
lesion’’) that may alter their biological function, including delivery
of oxygen to cells (2). Retrospective cohort studies have found a
correlation between RBC storage duration and morbidity and
mortality rates after transfusion (3–5), suggesting progressive stor-
age lesions may be responsible for adverse outcomes. Despite these
observational data, no large controlled clinical trials have been
conducted to evaluate the relationship between the age of stored
RBCs and clinical outcomes. RBC-transfused patients had worse
outcomes than nontransfused patients matched for clinical vari-
ables in several studies (6–10). Moreover, in randomized clinical
trials, a more liberal RBC transfusion strategy failed to benefit
pediatric or adult patients with anemia and critical illness (11, 12),
very year in the US, ?14 million units of blood are collected,
and could be improved. This is particularly relevant because, to
date, the development of and outcomes with blood substitutes have
been disappointing (14–16).
Although storage-induced changes in certain RBC molecular
function with storage (2, 17, 18). One of the RBC’s principal
functions is O2delivery, a product of changes in O2content and
blood flow. Increases in O2 affinity in stored RBCs, reflecting
progressive decreases in 2,3-diphosphoglycerate (2,3-DPG) over
by stored RBCs is deficient even early after processing and before
significant decline in 2,3-DPG (19). However, less is known of how
storage influences the role of the RBC in the O2-dependent
regulation of blood flow (‘‘hypoxic vasodilation’’), in part because
this RBC function was only recently appreciated (20, 21). The O2
sensor role of hemoglobin (Hb) subserves this RBC activity by
dispensing vasodilator S-nitrosothiol (SNO) equivalents in propor-
tion to the degree of hypoxia in the tissues it perfuses (20, 21). In
concert with the oxygenation-induced allosteric transition, S-
nitrosohemoglobin (SNO-Hb) forms in human blood when a nitric
of Hb (?1 in 1,000–10,000 Hb tetramers bind NO there). Con-
versely, RBCs perfusing tissues release limited fluxes of vasodilator
SNO equivalents in proportion to Hb O2desaturation (22), match-
ing regional blood flow with metabolic demand.
Previous studies have looked at particular aspects of storage
alone or have not collected, processed, and stored RBCs consistent
with standards of the AABB. Therefore, we conducted this study
RBC storage lesion and related storage-induced changes in RBC
physiologic functions critical to O2delivery, particularly deform-
ability and RBC-dependent vasoactivity. We reasoned that an
impairment in RBC-dependent hypoxic vasodilation might under-
Author contributions: E.B.-G., T.H.V., and T.J.M. designed research; E.B.-G., T.H.V., A.D.,
M.J.T., T.L.O., T.S.R., M.A.M., H.Z., R.M.C., and T.J.M. performed research; E.B.-G., A.D.,
M.J.T., T.L.O., T.S.R., M.A.M., H.Z., R.D.B., R.M.C., and T.J.M. analyzed data; and E.B.-G.,
T.H.V., A.D., and T.J.M. wrote the paper.
from NITROX LLC (www.nitrox.com) to perform this study; A.D. received less than $10,000
grant support from iNO Therapeutics. T.H.V. and T.S.R. (formerly an employee of Cato
Inc., which was contracted by NITROX LLC to perform the statistical analyses; R.M.C. is a
founder of and has a significant equity interest in NITROX LLC; T.J.M. is coinventor of U.S.
Patent 6,916,471, 2005 ‘‘Red blood cells loaded with S-nitrosothiols and uses therefore.’’
Freely available online through the PNAS open access option.
Abbreviations: AABB, organization formerly known as the American Association of Blood
Banks; NO, nitric oxide; PS, phosphatidyl serine; SNOs, S-nitrosothiols; Hb, hemoglobin;
(M)PC, (mercuric-coupled) photolysis-chemiluminescence; 3C, CO (carbon monoxide)-
saturated cuprous chloride/cysteine assay; 2,3-DPG, 2,3-diphosphoglycerate; CP2D, citrate-
phosphate-dextrose-dextrose anticoagulant solution; AS-3, additive solution-3.
‡‡To whom correspondence should be addressed at: Duke University Medical Center (Box
103003), Durham, NC 27710. E-mail: email@example.com.
This article contains supporting information online at www.pnas.org/cgi/content/full/
© 2007 by The National Academy of Sciences of the USA
October 23, 2007 ?
vol. 104 ?
no. 43 ?
lie the functional RBC storage lesion. In particular, we tested the
hypothesis that the processing and storage of RBCs for transfusion
may disturb SNO-Hb stability (e.g., by oxidation, degradation, or
release into the storage medium) and thus compromise RBC-
Demographic data from the 15 principal subjects are shown in
supporting information (SI) Table 1. To allow comparison of our
results with previously published studies, we measured several
variables that have been shown to change during processing and
was essentially unchanged between 3 h and 14 days, whereas Hb O2
saturation increased steadily during this period, possibly reflecting
(Fig. 1; 98% decline by 2 weeks, P ? 0.001). MetHb did not change
significantly over time, remaining below 0.3% throughout (SI Fig.
5). After the first week of RBC storage, pO2increased, consistent
Hb O2saturation reached ?99%. Potassium levels increased by
376% (P ? 0.001) and exceeded the maximum level of instrument
in the storage medium (indicating RBC hemolysis) increased
throughout the 6 weeks but remained below allowable levels, as in
previous reports (23, 24). Consistent with previous studies of
leukodepleted blood, RBC adhesion to endothelial cells (data not
shown) and RBC exposure of phosphatidyl serine (Fig. 1) did not
change significantly during storage (25, 26).
No changes occurred in calcium, magnesium, or chloride levels
(data not shown). Median (25th–75th percentile) serum glucose
addition of the glucose-containing CP2D (citrate, phosphate, dou-
ble dextrose; see formula in SI Methods) solution and with pro-
cessing for AS-3 RBCs, and remained above 500 mg/dl throughout.
Consistent with prior studies (27), RBC ATP content decreased by
55% from initial levels during storage (P ? 0.004, data not shown);
initial levels were also depressed, possibly reflecting an artifact
introduced by sample freezing.
We used two complementary assays to measure bioactive forms
of RBC NO (Fig. 2; individual data are shown in SI Fig. 6). Total
Hb-bound NO and SNO-Hb decreased markedly from 0 h (fresh
RBCs) to 3 h in unprocessed samples (i.e., samples unexposed to
AS-3 and not leukofiltered, Fig. 2). Specifically, RBC SNO-Hb was
1.2 ? 10?4mol of SNO per mole of Hb tetramer (1 SNO per 7,915
Hb tetramers) at 3 h vs. 6.5 ? 10?4mol of SNO per mole of Hb (1
SNO per 1,527 Hb tetramers) in fresh RBCs. Total Hb-NO and
serine (PS) expression (H) as a function of storage time. Data are median with 25th and 75th percentiles. P values represent significance for change over time.
RBC 2,3-DPG (A), potassium (B), pH (C), lactate (D), pO2(E), Hb O2saturation (SO2) (F), cell-free Hb in storage medium (G), and RBC surface phosphatidyl
www.pnas.org?cgi?doi?10.1073?pnas.0708160104Bennett-Guerrero et al.
SNO-Hb were similarly depressed in processed samples at 3 h (the
earliest postprocessing point), and remained markedly depressed
for the 6 weeks of the study. Hb[Fe]NO (iron-nitrosyl hemoglobin,
in which NO binds to Hb’s heme iron) levels did not change
significantly over time (Fig. 2). Total RBC SNO content, measured
by the 3C assay, was also depressed in stored RBCs (3.01 ? 10?5
mol of SNO per mole of Hb, or 1 SNO per 33,223 Hb tetramers)
at 3 h relative to that in both fresh RBC controls (9.0 ? 10?5mol
of SNO per mole of Hb, or 1 SNO per 11,106 Hb tetramers) and
published normal values (Fig. 2) (22). There were nonsignificant
trends toward later increases in total Hb-NO (Fig. 2A) and
SNO(Hb) (Fig. 2 C and D) values.
The vasoactivity of fresh, air-exposed venous RBCs was similar
in magnitude to that reported previously by us and others (21,
28–30). RBC vasoactivity was depressed by 3 h (Fig. 2; individual
was seen in time-control samples held for 3 h (Fig. 2). There was a
nonsignificant trend toward recovery of RBC vasoactivity, peaking
at a time (?1 week) similar to the nonsignificant resurgence in
SNO-Hb and total Hb-bound NO levels (Fig. 2). In separate
experiments to probe the possible role of RBC-derived ATP in
these responses, we investigated the influence of a nitric oxide
synthase (NOS) inhibitor (L-NAME) on RBC-dependent hypoxic
vasodilator responses (Fig. 3). The results show that L-NAME
Nitrite elicited no significant vasodilation (in hypoxia) at a concen-
with recent findings (29–32); moreover, there was no unmasking of
a response to nitrite by prior exposure of vessels to RBCs (Fig. 3).
(Fig. 4; individual data are shown in SI Fig. 7). Deformability
deteriorated over time at all sheer stresses tested (data not shown).
Blood cultures on all units during storage showed no evidence of
bacterial contamination. Consistent with this finding and with
cytokine assays in units of uncontaminated, leukodepleted RBCs
(33), levels of the proinflammatory cytokines IL-6, IL-8 (SI Fig. 8),
IL-1?, and tumor necrosis factor (TNF)-? (data not shown) in all
samples were extremely low and did not rise during storage.
Accordingly, only the first 76 were analyzed for cytokine content.
The findings of this study are that, in blood that has been collected,
processed, and stored by using blood-banking industry standard
to total Hb-NO minus Hb[Fe]NO), and RBC membrane SNO (E) were determined by the PC assay. (D) RBC (total) SNO was determined by the 3C assay. (F)
Vasoactivity represents the percentage decrease in tension induced by RBCs in the bioassay (percentage of vasorelaxation). Because of the complexity of the
membrane SNO assay, samples were assayed only at selected time points. Data are median with 25th and 75th percentiles. Unprocessed samples (open circles)
were assayed immediately (0 h) and, for some parameters, after a 3-h delay in addition to assays at the indicated times after processing was begun (filled circles,
beginning at 3 h). P values represent comparison between values in RBCs assayed immediately (0 h) vs. 3 h later (in unprocessed samples for A, B, C, and F). No
significant change from 3 h to 6 weeks was observed for any of these variables in processed samples.
SNO-Hb, related NO adducts, and vasoactivity of stored RBCs. (A–C and E) Total Hb-bound NO (A), Hb[Fe]NO (B), SNO-Hb (C, a calculated value equal
RBC-dependent hypoxic vasodilator responses are NOS-independent. Fresh
washed human RBCs [0.4% hematocrit (Hct) or ATP (10?6M] were added to
the percentage of vasorelaxation was measured.*, P ? 0.05. (B) Minimal nitrite-
to preconstricted rabbit aortic rings at 1% O2(PO27 mmHg) in the absence or
presence of RBCs (0.4% Hct). Data are mean ? SD from four experiments each.
Alternative mediators of RBC-dependent hypoxic vasodilation. (A)
Bennett-Guerrero et al.
October 23, 2007 ?
vol. 104 ?
no. 43 ?
operating procedures, RBC SNO-Hb levels and RBC-dependent
vasodilation are profoundly depressed immediately, whereas RBC
courses of individual functional changes differ markedly from one
relevant, and further study is needed to determine how these
transfusion (7–9, 12). In addition, we confirm numerous biochem-
ical and functional changes previously reported during RBC stor-
age, including some changes that may be related to our findings.
Our results bear directly on several key questions regarding the
safety and efficacy of RBC transfusion. Although RBC transfusion
evidence links transfusion of RBCs with increased mortality in
certain high-risk patients. The emphasis over the last several
decades has been on lesions that develop over the storage period.
linking the administration of older blood to increased mortality
(3–5), support the concern that transfusion of older blood may be
detrimental. Evidence also exists that transfusion is detrimental in
some settings regardless of RBC storage duration. Several studies
have assessed the relationship between RBC transfusion and out-
come, not accounting for the duration of storage of the blood. For
example, an association was found between perioperative RBC
transfusion and long-term mortality after coronary artery bypass
graft surgery (7, 8). Consistent with these studies, among 24,112
patients with acute coronary syndrome, those who received a RBC
transfusion had a significantly higher unadjusted rate of 30-day
death (8.00% vs. 3.08%; P ? 0.001) (9). This association persisted
in an analysis that took into account other known risk factors. A
clinical trial in critically ill adult patients demonstrated diminished
survival (approaching statistical significance, P ? 0.1) and signifi-
cantly higher in-hospital mortality in patients randomized to ‘‘lib-
eral’’ administration of RBC transfusions (12). In addition, mor-
tality was greater in those subgroups of liberally transfused patients
and other studies argue that any RBC transfusion may in fact be
deleterious rather than beneficial in some patient populations (6).
Recognition of the limited benefit or potential harm from RBC
transfusion is suggested by the adoption of more restrictive, sce-
(34). The failure of RBC transfusion to provide a clinical benefit in
other groups of pediatric or adult patients with anemia and critical
illness raises the broader concern that RBC storage is problematic
and could be improved (2, 6, 11, 12).
In support of the possibility that even fresh, processed RBCs are
dysfunctional in some respects, we show that RBC SNOs are
depressed by a factor of ?4 at 3 h after collection. In earlier work,
we showed that in venous RBCs in PBS (pH 7.4) held ex vivo for
30 min before assay (25°C, native PO2), SNO-Hb levels declined
markedly (30); similarly, under conditions mimicking blood bank-
ing (pH ?7.0), RBC SNO-Hb and RBC vasoactivity also decline
early [see companion article (35)]. Together, these two observa-
tions motivated the current study. Our finding of early and marked
RBC SNO-Hb depletion is strengthened by the use of two mech-
chloride-cysteine) assays]; indeed, the extent of SNO decline in
RBCs (time point 0 h vs. postprocessing, AS-3-exposed RBCs, at
3 h) was similar by both methods (77% and 67% declines by MPC
Hb-NO (measured by the MPC assay) and RBC bioactivity were
likewise depressed at 3 h in unprocessed samples suggests that time
to produce the deficiencies. RBC SNO-Hb (by MPC assay) and
total RBC SNO content (measured by the 3C assay, and repre-
senting the sum of Hb-bound SNO, other protein-bound SNO, and
S-nitrosoglutathione) are different but overlapping pools of SNO
species. RBC SNO-Hb represents the major RBC SNO (21, 36),
the distinct mechanisms of the two assays. Indeed, recoveries of
concentrations (as in RBC lysates) by 3C and MPC are essentially
complete and therefore equivalent between the two assays. But
herein and previously, basal levels of total RBC SNO content
measured by 3C are several-fold lower than basal RBC SNO-Hb
measured by MPC. The exact reason for the difference in basal
values of RBC SNO(Hb) using the two techniques is unknown, but
likely reflects the known differences in sample preparation. In
addition, the ability of 3C to measure membrane vs. cytosolic
SNO-Hb, as well as different subpopulations of SNO-Hb species
(37), is currently unknown.
Taken together, our findings that total RBC SNO, RBC SNO-
are consistent with the loss of SNOs. Hb[FeII]NO, which serves as
a precursor to SNO-Hb (21) but is biologically inactive itself (30),
is equivalent to the sum of these two species). The SNO loss from
RBCs may be either through export (e.g., to thiol SNO acceptors,
the red cells were removed) or degradation to an inactive NO
metabolite and cannot be accounted for by accumulation of
Hb[Fe]NO, which did not change. A functionally active endothelial
NO could serve as a substrate in formation of RBC SNO-Hb, thus
providing RBC-eNOS-derived NO equivalents with a pathway for
export and activity. However, the loss of RBC SNO(Hb) with
storage is not a direct result of limitation of RBC eNOS, because
in fresh RBCs, SNOs were abundant despite the absence of
required for the reported RBC eNOS activity) and despite the
presence of the calcium-chelator DTPA. The mechanistic basis for
Data are median with 25th and 75th percentiles. P values represent significance for change over time.
RBC deformability as elongation index for two representative shear stress levels as a function of storage time. Values at 0 h are from unprocessed RBCs.
www.pnas.org?cgi?doi?10.1073?pnas.0708160104Bennett-Guerrero et al.
under study by our group.
In contrast to some other RBC disorders involving SNOs, in
which the loss of SNO-Hb was accounted for by a gain in inactive
Hb[Fe]NO, the RBC storage lesion entails depression of SNO-Hb
with no change in Hb[Fe]NO levels, so that total Hb-bound NO
levels (the sum of Hb[Fe]NO and SNO-Hb) were also depressed.
In addition, total RBC SNO (measured by the 3C assay, the sum of
Hb-SNO, other protein-bound SNO, low-mass SNO) and levels of
membrane SNOs, which are essential for RBC export of SNO
bioactivity, were substantially depressed compared with levels in
fresh RBCs and published values (22, 28, 39).
The initial vascular response to RBCs at low pO2is relaxation,
requires minutes and is therefore of little or no biological signifi-
cance because RBCs typically traverse capillaries within seconds
in RBC (Hb-bound) SNO were similar, whereas decreases in ATP
were more gradual and less profound, as in previous studies (40).
Further evidence that the loss of RBC-derived ATP during storage
does not account for the weakening of hypoxic vasodilation to
stored RBCs came from experiments in which a NOS inhibitor
The weak vasodilator nitrite cannot account for these responses
because, even in the presence of RBCs, the levels required for
vasoactivity (micromolar) by nitrite exceed the values present in
blood [Fig. 3 and Crawford et al. (29)]. By contrast, a SNO synthase
function of RBC Hb can convert nitrite to bioactive SNO-Hb (41,
42), and our data do not exclude a role for this activity in
maintaining the low, residual SNO-Hb levels seen during RBC
storage or in the later, nonsignificant trend toward small increases
in SNO(Hb) (Fig. 2 and SI Fig. 5).
RBC SNO-Hb levels correlate with the ability of RBCs to relax
blood vessels in hypoxia (‘‘hypoxic vasodilation’’), a functional
‘‘bioassay’’ for the recently described role of the RBC in blood flow
regulation in vivo (20–22). Derangements in RBC SNO-Hb mirror
derangements in RBC vasodilator activity in several diseases stud-
ied to date, with the specific lesion varying from one disorder to
nitrosylation is increased because glycosylation of Hb favors the R
structure, but RBC vasoactivity is depressed because the R-state
diabetes (20). In sickle cell disease, RBC SNO-Hb and vasoactivity
are depressed, and the degree of depression correlates with disease
of differences in heme redox potential between sickle and normal
human Hb. Hypoxic vasodilation to sickle RBCs is impaired
because of both the SNO-Hb deficiency and abnormal transfer of
SNO from Hb to thiols in the membrane protein anion exchanger
1 (AE1), an essential step in RBC-SNO-dependent vasorelaxation
(28, 39). In patients with pulmonary arterial hypertension and
hypoxemia, RBC SNO-Hb formation was deficient, and both the
improved when RBC SNO-Hb was replenished in vivo (30). Thus,
the in vitro RBC bioassay model recapitulates in vivo O2-sensitive
function of the RBC in blood flow regulation in humans in health
(21, 37), disease (44), and in corrective therapy.
We showed that, whereas changes in RBC-dependent vasodila-
tion and SNO-Hb take place early after storage (within 3 h),
changes in RBC deformability take place more gradually (days to
weeks). We measured RBC deformability using the laser-assisted
optical rotational cell analyzer (LORCA) assay, in which RBCs are
subjected to increasing shear stress over several minutes (45, 46).
Our prespecified major shear stress of interest was 3 Pa, a level that
may be encountered in the microcirculation of humans (47–49).
Deformability also decreased significantly over the 42-day storage
period at 30 Pa, suggesting that the membrane defect induced by
storage is pronounced enough to moderately resist the ability to
deform even at a very high level of shear stress. Less deformable
RBCs could exacerbate organ ischemia in surgical and other
wider than capillaries, so they must deform to traverse the micro-
circulation. RBCs that are less deformable can either block and
obstruct capillaries or, more commonly, traverse the microcircula-
tion at a significantly increased transit time, resulting in overall
diminished O2delivery to organs (50, 51). Impaired RBC deform-
ability was shown in patients with sepsis, a condition of microcir-
of changes in RBC deformability from septic patients is similar to
those we observed in stored RBCs from healthy volunteers. The
clinical relevance of this and other storage lesions requires further
The lack of RBC adhesiveness to endothelial cells or RBC
study suggests that stored, leukodepleted RBCs do not contribute
to diminished O2delivery to tissues by either adhesion, as seen in
sickle cell disease, or by direct activation of coagulation pathways
through exposure of PS, as is postulated to occur in other RBC
The storage-related deficiency of SNO-Hb and impairments in
in the microcirculation, predisposing to the excess morbidity and
mortality associated with RBC transfusion in some patient groups
(7–9, 12). Strategies to replenish SNOs in RBCs have been shown
in other diseases to be effective and partially reverse the related
impairment in RBC-dependent vasodilation and correlated phys-
iologic derangements (28, 30). Restoring SNOs might also improve
further study is needed to determine whether strategies that re-
plenish SNOs in stored RBCs, or prevent their loss, improve
storage-induced rheological lesions. It is rational to test, in future
clinical trials, whether replenishing SNOs in stored RBCs, or
preventing their loss, will improve patient outcomes with RBC
Materials and Methods
After written informed consent, healthy volunteers meeting eligi-
bility criteria were enrolled at Duke University Medical Center
(DUMC), whose Institutional Review Board approved the study.
Blood Donation and Processing. Blood was collected and processed
by American Red Cross-certified technicians using techniques
consistent with the Technical Manual of the AABB (55). Venous
blood was collected in CP2D anticoagulant, leukofiltered, and
stored in a monitored refrigerator. RBC aliquots were removed at
the indicated intervals. For certain assays, blood was also collected
into 10-ml Vacutainers (containing CP2D anticoagulant) and im-
mediately analyzed (0 h, see SI Methods), avoiding the ?3-h delay
inherent in the large-volume blood collection, leukofiltration, and
processing required to prepare additive-solution RBCs. For some
parameters, (2,3-DPG, RBC ATP, extraerythrocytic (free) Hb, 3C
SNO, and cytokines), aliquots were taken and stored at ?80°C for
later batch analysis. To keep all personnel performing assays
a unique identification code based on a computer-generated
randomization list. See SI Methods for more details.
Assays. RBC-dependent vasoactivity was determined as changes in
isometric tension in phenylephrine-preconstricted rabbit aortic
RBC SNO-Hb, Hb[Fe]NO, and membrane SNO assay. Photolysis-
chemiluminescence (PC) was described previously (21) and is
detailed in SI Methods. Washed RBCs were lysed and Hb desalted
by using a G-25 fine Sephadex column. Hb-bound NO was mea-
Bennett-Guerrero et al.
October 23, 2007 ?
vol. 104 ?
no. 43 ?
sured by PC assay in the presence (Hb[Fe]NO) and absence (total Download full-text
Hb-bound NO) of HgCl2, which selectively cleaves SNO bonds.
SNO-Hb is the difference between total Hb-NO and Hb[Fe]NO.
additional thiols are present).
Total RBC SNO: Reduction in carbon monoxide (CO)-saturated copper/
cysteine (3C). SNOs are selectively reduced to NO in a cuprous
chloride- and CO-saturated cysteine solution that is replaced
before the injection of each sample; released NO is detected as
the chemiluminescent product of its reaction with ozone (22).
RBC deformability. RBC deformability was measured by using a
laser-assisted optical rotational cell analyzer (LORCA) as de-
two main levels of shear stress, 3 and 30 Pa (49), with 3 Pa
representing a clinically relevant level of shear stress in the
microcirculation (47, 48).
Cytokine assays. Cytokines were determined in the storage medium
as described in SI Methods.
RBC adhesion. RBC adhesion was determined as described (56), by
using a variable-height flow chamber and TNF?-treated primary
human umbilical vein endothelial cells as the adhesive substrate.
RBC ATP and 2,3-DPG. RBC ATP and 2,3-DPG content were
determined after deproteinization by using standard techniques
described in SI Methods.
Extraerythrocytic (free) Hb assay. Apublishedtechnique(57)wasused
in modified form (SI Methods).
PS exposure. PS exposure on the RBC membrane was measured
as described in SI Methods.
Blood gases. Blood gases were performed by using a CCX 8 blood
gas analyzer (Nova Biomedical, Waltham, MA).
Blood culture. Blood culture was performed at the DUMC clinical
microbiology laboratory by using standard methods.
Statistical methods and data integrity. Statistical methods and data
integrity are described in SI Methods.
We are grateful to StatWorks, Inc. (Carrboro, NC) for providing statistical
support and to Jerry L. Kirchner for expert technical assistance. This study
involved in the design and conduct of the study; collection, management,
analysis, and interpretation of the data; and preparation of the manuscript.
Under their agreement with Duke University, NITROX had no right to
‘‘approve’’ or not of this manuscript. Additional support was provided by
1. Whitaker BI, Henry R (2005) 2005 Nationwide Blood Collection and Utilization
Survey Report (National Blood Data Resource Center, US Department of
Health and Human Services, Washington, DC).
3. Leal-Noval SR, Jara-Lopez I, Garcia-Garmendia JL, Marin-Niebla A, Herruzo-
Aviles A, Camacho-Larana P, Loscertales J (2003) Anesthesiology 98:815–822.
4. Purdy FR, Tweeddale MG, Merrick PM (1997) Can J Anaesth 44:1256–1261.
5. Zallen G, Offner PJ, Moore EE, Blackwell J, Ciesla DJ, Gabriel J, Denny C,
Silliman CC (1999) Am J Surg 178:570–572.
6. Vincent JL, Baron JF, Reinhart K, Gattinoni L, Thijs L, Webb A, Meier-
Hellmann A, Nollet G, Peres-Bota D (2002) J Am Med Assoc 288:1499–1507.
7. Kuduvalli M, Oo AY, Newall N, Grayson AD, Jackson M, Desmond MJ, Fabri
BM, Rashid A (2005) Eur J Cardiothorac Surg 27:592–598.
8. Koch CG, Li L, Duncan AI, Mihaljevic T, Loop FD, Starr NJ, Blackstone EH
(2006) Ann Thorac Surg 81:1650–1657.
9. Rao SV, Jollis JG, Harrington RA, Granger CB, Newby LK, Armstrong PW,
Moliterno DJ, Lindblad L, Pieper K, Topol EJ, et al. (2004) J Am Med Assoc
10. Hill SR, Carless PA, Henry DA, Carson JL, Hebert PC, McClelland DB,
Henderson KM (2002) Cochrane Database Syst Rev, CD002042.
11. Lacroix J, Hebert PC, Hutchison JS, Hume HA, Tucci M, Ducruet T, Gauvin
12. Hebert PC, Wells G, Blajchman MA, Marshall J, Martin C, Pagliarello G,
Tweeddale M, Schweitzer I, Yetisir E (1999) N Engl J Med 340:409–417.
13. Hogman CF, Meryman HT (2006) Transfusion 46:137–142.
14. Kerner T, Ahlers O, Veit S, Riou B, Saunders M, Pison U (2003) Intensive Care
15. Sloan EP, Koenigsberg M, Gens D, Cipolle M, Runge J, Mallory MN, Rodman
G, Jr (1999) J Am Med Assoc 282:1857–1864.
16. Winslow RM (2006) Vox Sang 91:102–110.
17. Valeri CR, Hirsch NM (1969) J Lab Clin Med 73:722–733.
18. Valtis DJ (1954) Lancet 266:119–124.
19. Tsai AG, Cabrales P, Intaglietta M (2004) Transfusion 44:1626–1634.
20. James PE, Lang D, Tufnell-Barret T, Milsom AB, Frenneaux MP (2004) Circ
21. McMahon TJ, Moon RE, Luschinger BP, Carraway MS, Stone AE, Stolp BW,
Gow AJ, Pawloski JR, Watke P, Singel DJ, et al. (2002) Nat Med 8:711–717.
22. Doctor A, Platt R, Sheram ML, Eischeid A, McMahon T, Maxey T, Doherty J,
Axelrod M, Kline J, Gurka M, et al. (2005) Proc Natl Acad Sci USA 102:5709–
23. Hess JR, Kagen LR, van der Meer PF, Simon T, Cardigan R, Greenwalt TJ,
AuBuchon JP, Brand A, Lockwood W, Zanella A, et al. (2005) Vox Sang
24. Sowemimo-Coker SO (2002) Transfus Med Rev 16:46–60.
25. Bratosin D, Leszczynski S, Sartiaux C, Fontaine O, Descamps J, Huart JJ,
Poplineau J, Goudaliez F, Aminoff D, Montreuil J (2001) Cytometry 46:351–356.
26. Luk CS, Gray-Statchuk LA, Cepinkas G, Chin-Yee IH (2003) Transfusion
27. Fagiolo E, Mores N, Pelliccetti A, Gozzo ML, Zuppi C, Littarru GP (1986) Folia
Haematol Int Mag Klin Morphol Blutforsch 113:783–789.
28. Pawloski JR, Hess DT, Stamler JS (2005) Proc Natl Acad Sci USA 102:2531–
29. Crawford JH, Isbell TS, Huang Z, Shiva S, Chacko BK, Schechter AN,
Darley-Usmar VM, Kerby JD, Lang JD, Jr, Kraus D, et al. (2006) Blood
30. McMahon TJ, Ahearn GS, Moya MP, Gow AJ, Huang YC, Luchsinger BP,
Nudelman R, Yan Y, Krichman AD, Bashore TM, et al. (2005) Proc Natl Acad
Sci USA 102:14801–14806.
31. Dalsgaard T, Simonsen U, Fago A (2007) Am J Physiol 292:H3072–H3078
32. Deem S, Min JH, Moulding JD, Eveland R, Swenson ER (2007) Am J Physiol
33. Shanwell A, Kristiansson M, Remberger M, Ringden O (1997) Transfusion
34. Hebert PC, Fergusson DA, Stather D, McIntyre L, Martin C, Doucette S,
Blajchman M, Graham ID (2005) Crit Care Med 33:7–12.
35. Reynolds JD, Ahearn GS, Angelo M, Zhang J, Cobb F, Stamler JS (2007) Proc
Natl Acad Sci USA 104:17058–17062.
36. Liu L, Yan Y, Zeng M, Zhang J, Hanes MA, Ahearn G, McMahon TJ, Dickfeld
T, Marshall HE, Que LG, Stamler JS (2004) Cell 116:617–628.
37. Singel DJ, Stamler JS (2005) Annu Rev Physiol 67:99–145.
38. Kleinbongard P, Schulz R, Rassaf T, Lauer T, Dejam A, Jax T, Kumara I,
Gharini P, Kabanova S, Ozuyaman B, et al. (2006) Blood 107:2943–2951.
39. Pawloski JR, Hess DT, Stamler JS (2001) Nature 409:622–626.
40. Hess JR, Rugg N, Knapp AD, Gormas JF, Hill HR, Oliver CK, Lippert LE,
Greenwalt TJ (2001) Transfusion 41:1045–1051.
41. Angelo M, Singel DJ, Stamler JS (2006) Proc Natl Acad Sci USA 103:8366–8371.
42. Luchsinger BP, Rich EN, Gow AJ, Williams EM, Stamler JS, Singel DJ (2003)
Proc Natl Acad Sci USA 100:461–466.
43. Crawford JH, Chacko BK, Pruitt HM, Piknova B, Hogg N, Patel RP (2004)
44. Sonveaux P, Lobysheva II, Feron O, McMahon TJ (2007) Physiology 22:97–112.
45. Baskurt OK, Gelmont D, Meiselman HJ (1998) Am J Respir Crit Care Med
46. Hardeman MR, Besselink GA, Ebbing I, de Korte D, Ince C, Verhoeven AJ
(2003) Transfusion 43:1533–1537.
47. Lipowsky HH, Firrell JC (1986) Am J Physiol 250:H908–H922.
48. Lipowsky HH, Sheikh NU, Katz DM (1987) J Clin Invest 80:117–127.
49. van Bommel J, de Korte D, Lind A, Siegemund M, Trouwborst A, Verhoeven
AJ, Ince C, Henny CP (2001) Transfusion 41:1515–1523.
50. Lipowsky HH (1982) Physiologist 25:357–363.
51. Secomb TW, Hsu R, Pries AR (2001) Am J Physiol 281:H629–H636.
52. Vincent JL, De Backer D (2005) Crit Care 9 Suppl 4:S9–S12.
53. Piagnerelli M, Boudjeltia KZ, Vanhaeverbeek M, Vincent JL (2003) Intensive
Care Med 29:1052–1061.
54. Bor-Kucukatay M, Wenby RB, Meiselman HJ, Baskurt OK (2003) Am J Physiol
55. Brecher ME (2005) AABB Technical Manual (AABB Press, Bethesda).
56. Zennadi R, Hines PC, De Castro LM, Cartron JP, Parise LV, Telen MJ (2004)
57. McMahon TJ, Stamler JS (1999) Methods Enzymol 301:99–114.
www.pnas.org?cgi?doi?10.1073?pnas.0708160104Bennett-Guerrero et al.