Transfusion in the Critically III (also
see Chapter 47
)
Anemia in the Intensive Care Unit
Anemia, a common problem in the ICU patient, is usually treated
with red blood cell (RBC) transfusions. Statistics show that more than 11 million
units of RBCs are transfused in more than 3 million patients per year.[174]
As many as 25% of the patients in the ICU eventually receive at least one unit of
packed red blood cells (PRBCs)[175]
in an attempt
to augment the delivery of oxygen and avoid the deleterious effects of oxygen debt.
[176]
Although there is significant variation in
critical care transfusion practices, many intensivists still adhere to the 10 g/dL
threshold.[175]
However, critically ill patients
may also be harmed by the immunosuppressive[177]
[178]
and microcirculatory complications of RBC
transfusions.
[179]
RBC transfusions also carry a finite risk
of transmission of infectious diseases (i.e., human immunodeficiency virus and hepatitis
B and C).[180]
Transfusions can also result in
volume overload and pulmonary edema. Current projections are that by the year 2030,
there will be an annual shortfall of 4 million units of PRBCs.[181]
All these issues have encouraged studies examining a more conservative approach
to blood transfusions and consideration of the use of alternative therapies such
as recombinant human erythropoietin.
Transfusion thresholds in critically ill patients have been compared
in several trials. Most of these trials have been limited by the small number of
patients enrolled and have been unable to demonstrate a difference in adverse outcomes
or mortality.[182]
[183]
[184]
Given the lack of equipoise in transfusion
thresholds, in 1999, Hebert and associates[185]
randomized 838 patients to a restrictive transfusion strategy (i.e., hemoglobin maintained
between 7 and 9 g/dL) or a liberal transfusion strategy (i.e., hemoglobin between
10 and 12 g/dL). Their results showed that the restrictive strategy decreased the
average number of RBC units transfused by 54% (2.6 ± 4.1
versus 5.6 ± 5.3 RBCs/patient, P
< 0.1). The overall 30-day mortality rate was similar for the two groups (18.7%
versus 23.3%, P = .11), but the mortality rate was
lower for the restrictive group among patients who were less acutely ill (8.7% versus
16.1%, P = .03) and among patients younger than 55
years (5.7% versus 13.0%, P = .02). The mortality
rate was also similar for the two groups in patients with clinically significant
cardiac disease (20.5% of the restricted group versus 22.9% of the liberal group,
P = .69). This study suggested that there was no
clear benefit to a higher transfusion threshold. Although there was a trend toward
a higher 30-day mortality rate for the liberal transfusion group, the in-hospital
mortality rate was significantly lower in the restrictive-strategy group (22.2% versus
28.1%, P = .05). These data support the adoption
of more restrictive transfusion triggers because the restrictive group used one half
of the number of transfusions without any adverse impact on mortality. For most
critically ill patients, the restrictive transfusion strategy is at least as effective
as and possibly superior to a liberal transfusion strategy. The effects of restricted
transfusion triggers on the functional status, morbidity, and mortality of patients
with coronary disease need to be studied in future clinical trials. It must be emphasized,
however, that patients with cardiac ischemia were excluded from these studies. In
a companion study, Hebert and coworkers[186]
reported
that the subset of patients found retrospectively to have severe cardiac ischemia
had a trend toward improved survival when treated with the liberal transfusion strategy.
This concept is further supported by the findings of Wu and colleagues,[187]
who performed a retrospective study of 78,974 patients hospitalized with acute myocardial
infarction. In these patients, anemia was an independent risk factor for mortality,
and transfusion reduced the mortality rate for patients up to a hemoglobin concentration
of 11 mg/dL.
Erythropoietin Therapy
The high transfusion requirements of critically ill patients have
many causes.[188]
Aside from phlebotomy for routine
laboratory testing, it appears that critically ill adults have a blunted erythropoietic
response because of the action of inflammatory mediators.[189]
Administration of recombinant human erythropoietin (rHuEPO) to patients with multiple
organ failure resulted in a productive erythropoietic response.[190]
Corwin and coworkers[191]
hypothesized that ICU
patients might have inadequately elevated erythropoietin levels or were unable to
respond
to endogenous erythropoietin. They randomized patients to receive rHuEPO (300 units/kg)
subcutaneously daily for 5 days and then every other day thereafter for a minimum
of 2 weeks or until ICU discharge versus placebo. They showed that the rHuEPO group
received 45% fewer units of RBCs (166 versus 305 units, P
< .002), and the final hematocrit was higher in the rHuEPO group than placebo
(35.1 ± 5.6 versus 31.6 ± 4.1,
P < .01). There were no significant differences
in mortality or frequency of adverse events between the two groups. In a larger
study, 1302 patients who had been in the ICU at least 2 days were randomized to EPO
(40,000 units) weekly or to placebo. Patients receiving EPO were less likely to
be transfused (60.4% versus 50.5%, P < .001) and
received fewer units of PRBCs.[192]
There was no
difference in mortality between the groups.
In summary, alternatives to RBC transfusions may prove to be beneficial,
and along with a more restrictive transfusion strategy, the number of RBC transfusions
may decline, leading to lower costs as well as improved patient safety. The cost-effectiveness
of routine EPO use in critically ill patients has not been established and will depend
on complex factors, including the relative cost of the drug and the cost and availability
of blood products. In the meantime, eliminating unnecessary laboratory testing,
using pediatric blood sampling tubes, and aggressive nutritional support can decrease
transfusion requirements.
