ARTICLE IN PRESS
Anemia and blood transfusion practices
in the critically ill: A prospective cohort review
Jissy Thomas, RN, MN,a Louise Jensen, RN, PhD,a Susan Nahirniak, MD, FRCP (C),b and
R.T. Noel Gibney, MB, FRCP (C)c
BACKGROUND: Nearly 75% of critically ill patients develop anemia in the intensive care unit (ICU).
Anemia can be treated with red blood cell (RBC) transfusions, although evidence suggests that lower
hemoglobin levels are tolerated in the critically ill. Despite such recommendations, variation exists in clinical practice.
METHODS: A prospective cohort was assessed for anemia and RBC transfusion practices in 100 consecutive adults admitted to our General Systems ICU.
RESULTS: The prevalence of anemia in this cohort was 98%. Mean blood loss via phlebotomy was
2510.3 mL per patient per day. The RBC transfusion rate for the ICU stay was 40%, increasing to 70%
in patients whose ICU stay was >7 days. The mean pretransfusion level of hemoglobin was 7.350.47
mg/dL for the total cohort, and 8.20.65 mg/dL for those with a history of cardiovascular disease.
CONCLUSION: Anemia was common in this critically ill cohort, with hemoglobin levels continuing to
drop with ICU stay. Pretransfusion hemoglobin levels were lower than reported by others, yet the RBC
transfusion rate was comparable. There was no association between anemia and phlebotomy practices
in our ICU. (Heart LungÒ 2009;-:1–9.)
N
early 75% of critically ill patients develop
anemia in the intensive care unit (ICU).1
Among the causes of anemia in the critically
ill,2 phlebotomy for diagnostic testing is a major
source of blood loss.3 Technical improvements in
diagnostic measurements and restraints in blood
sampling have resulted in less phlebotomy blood
loss and resultant iatrogenic anemia.4,5 However,
volumes of iatrogenic blood loss and rates of anemia
vary along with institutional practices. More than
50% of ICU patients receive at least one unit of red
blood cells (RBCs) to maintain hemoglobin and
hematocrit levels at >10 mg/dL and 30%, respectively.6 This 10/30 rule, first proposed in 1942 by
From the aFaculty of Nursing, University of Alberta, Edmonton,
Alberta, Canada; bDepartment of Laboratory Medicine and Pathology, University of Alberta Hospital, Alberta, Edmonton, Alberta,
Canada; and cDivision of Critical Care Medicine, Faculty of
Medicine and Dentistry, University of Alberta, Edmonton, Alberta,
Canada.
Corresponding author: Louise Jensen, RN, PhD, Faculty of Nursing, University of Alberta, Edmonton, Alberta T6G 2G3, Canada;
E-mail: louise.jensen@ualberta.ca
0147-9563/$ – see front matter
Copyright Ó 2009 by Mosby, Inc.
doi:10.1016/j.hrtlng.2009.07.002
HEART & LUNG VOL. -, NO. -
Adams and Lundy, was accepted as the transfusion
trigger.7 Historically, anemia was assumed risky, particularly among patients with cardiovascular disease, and RBC transfusion was deemed beneficial.
However, the [TRICC] trial suggests that lower hemoglobin levels are well-tolerated in most critically ill
patients, including those with cardiovascular disease.8,9 Therefore, blood should be transfused considering factors such as age, disease severity, and
specific diagnoses.6,7,10,11 Yet in a United Kingdom
survey of transfusion practices with the critically ill,
Boralessa et al12 reported RBC transfusion triggers
of 9 to 10 mg/dL. Another survey of European ICU
transfusion practices found that the mean pretransfusion hemoglobin level of patients not actively
bleeding was 8.5 mg/dL.2 This was similar to the
finding of Corwin et al.13 Thus despite the recommendations of the TRICC trial,9 there is variation in
clinical practice.6,7,11,14 In a contemporary Canadian
survey, Hebert et al15 reported that critical-care physicians have adopted a lower transfusion threshold
of 7 mg/dL. However, this change in practice was
not uniform. Furthermore, Turgeon et al16 found significant variation in transfusion practices among
Canadian anesthesiologists within specific patient
www.heartandlung.org
1
ARTICLE IN PRESS
Anemia and Blood Transfusion in the Critically Ill
Thomas et al
Table I
Subject characteristics (n=100)
Characteristics
Frequency
Age
18-30 years
31-50 years
51-70 years
71-90 years
7
25
49
19
Gender
Male
Female
57
43
Admission diagnosis
Respiratory disorders
Gastrointestinal disorders
Neurologic disorders
Cardiovascular disorders
Genitourinary disorders
Multisystem disorders
Other
41
17
8
7
3
9
15
subgroups. Concerns regarding this and the limited
blood supply led to an evaluation of blood transfusion practices in our General Systems ICU.
BACKGROUND
Two main factors contribute to anemia in the critically ill: insufficient production of red blood cells,1,6,17
and blood loss.4,18 Underproduction anemia is commonly referred to as the anemia of chronic inflammatory disease. It is attributable to abnormal serum
erythropoietin (EPO) concentrations, with a minimal
reticulocyte response to endogenous EPO because
of inhibition of the EPO gene by inflammatory mediators such as interleukin-1a, tumor necrosis factor-a,
tumor growth factor-b, and interleukin-1b.1,18 Second,
iatrogenic anemia occurs because of blood loss
through blood draws, occult gastrointestinal bleeding, and invasive techniques. Tarpey and Lawler19
investigated ICU patients, and found a mean blood
loss per day of 66 mL, and a total blood loss of 336
mL. Dale and Pruett20 studied patients in a medical
ICU and in a medical unit, and found that ICU patients
2 www.heartandlung.org
Characteristics
Comorbidities
Coronary Artery Disease history
Hypertension
Liver insufficiency
Diabetes mellitus
COPD
Chronic renal failure
CHF
Other
Admission hemoglobin
6.0-8.0 mg/dL
8.1-10.0 mg/dL
10.1-12.0 mg/dL
12.1-14.0 mg/dL
14.1-18.2 mg/dL
Admission APACHE II cores
1-20
21-30
>30
Length of ICU stay
1-3 days
4-7 days
8-12 days
13-20 days
21-28 days
Frequency
16
14
12
8
5
4
3
32
7
32
32
22
7
75
21
4
26
39
19
9
7
had more blood draws per day (3 vs. 1.5), and a higher
mean total blood loss (550 mL vs. 208 mL). More
recently, von Ahsen et al4 found that diagnostic blood
loss ranged from 20 to 41 mL, accounting for 17% of
total blood loss. Similarly, Ba et al21 reported a phlebotomy blood loss of 40 mL per day, and Vincent et
al22 in the Anemia and Blood Transfusion in Critical
Care (ABC) Trial documented an average daily phlebotomy volume of 41 mL.
A primary treatment of anemia in the critically ill
has been RBC transfusions. However, studies reviewing RBC transfusion practices, based on a variety of
criteria, estimate a proportion of unnecessary RBC
transfusions ranging from 4% to 66%.2,23 Vincent et
al2 reported RBC transfusion rates of 37% in European
ICUs, and 73% with ICU length of stay >1 week, with
a mean pretransfusion hemoglobin level of 8.5 mg/
dL. Corwin et al24 reported that 85% of patients with
an ICU stay of >1 week received at least one unit of
RBCs during their stay. In a later study,25 they found
that most RBC transfusions in the ICU were not associated with acute blood loss, but rather a constant
transfusion rate of 2 to 4units/week. In the Anemia
--/-- 2009
HEART & LUNG
ARTICLE IN PRESS
Thomas et al
Anemia and Blood Transfusion in the Critically Ill
120
10
100
8
80
6
60
4
40
2
20
0
Number of Patients
Mean Hemoglobin (mg/dL)
Mean Hemoglobin Per Patient Per Day (mg/dL)
12
0
1
3
5
7
9
11 13 15 17 19 21 23 25 27
ICU Days
Mean Hemoglobin (mg/dL)
Number of Patients
Fig 1 Mean hemoglobin per patient per day (mg/dL).
Table II
Phlebotomy practices
Mode
Median
Phlebotomy rate per patient
Per day
5.3
Total LOS
15.0
Blood loss per patient (mL)
Per day
18.25
Total LOS
117.0
Phlebotomy discard volume per patient (mL)
Per day
7.7
Total LOS
21.0
3.5
19.5
Mean (SD)
3.7 (1.8)
29.6 (29.3)
Range
2-15
3-155
22.7
156.5
24.7 (10.3)
223.6 (183.4)
8-59
32-944
12.4
68.0
13.2 (6.3)
106.0 (103.5)
3-54
0-528
LOS, length of stay.
and Blood Transfusion in Critical Care (CRIT) trial,26
70% of ICU patients developed anemia within 48 hours
of admission, and 44% received at least one unit of
RBCs during their ICU stay, with a mean pretransfusion hemoglobin level of 8.6 mg/dL.
The decision to transfuse is usually based on
patient factors such as volume status, acuteness of
anemia, severity of symptoms, age, and presence of
comorbid conditions, and particularly cardiovascular
disease.27 Hebert et al28 found that critically ill
patients with cardiac disease exhibited a trend toward
increased mortality when hemoglobin levels were
<9.5 mg/dL, compared with other diagnoses. Rivers
HEART & LUNG VOL. -, NO. -
et al29 found that early intervention in severe sepsis
and septic shock, by transfusing RBCs to a hematocrit
level $30%, led to reduced in-hospital mortality,
a higher central venous oxygen saturation, a lower lactate level, and a lower base deficit. Precise indications
for RBC transfusion in other situations remain controversial in the critically ill because of a lack of data from
clinical trials. Existing guidelines for RBC transfusion
and other blood components are derived from animal
experiments, and from observations of patients undergoing resuscitation, surgery, and anesthesia.30,31
Adam and Lundy32 first recommended the administration of RBCs for hemoglobin levels of 8 to 10 mg/dL in
www.heartandlung.org
3
ARTICLE IN PRESS
Anemia and Blood Transfusion in the Critically Ill
Thomas et al
Mean Number of Blood Draws Per Patient Per Day
120
7
100
5
80
4
60
3
40
Number of Patients
Mean Number of Blood Draws
6
2
20
1
0
0
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28
ICU Days
Mean Number of Blood Draws
Number of Patients
Fig 2 Mean number of blood draws per patient per day.
60
120
50
100
40
80
30
60
20
40
10
20
Number of Patients
Mean Blood Loss (mls)
Mean Phlebotomy Blood Loss Per Patient Per Day
0
0
1
2
3
4
5
6
7
8
9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28
ICU Days
Mean Blood Loss (mls)
Number of Patients
Fig 3 Mean phlebotomy blood loss per patient per day.
the perioperative period, when surgical patients with
hemoglobin levels of <10 mg/dL were not observed
to do well. In the 1980s, a majority of surveyed anesthesiologists believed a preoperative hemoglobin
4
www.heartandlung.org
level of 9 to 10 mg/dL to be mandatory.33,34 Surgenor
et al7 recommended maintaining a hemoglobin level
of 10 to 12 mg/dL in patients with cardiovascular
disorders.
--/-- 2009
HEART & LUNG
ARTICLE IN PRESS
Thomas et al
Anemia and Blood Transfusion in the Critically Ill
Mean RBC Transfusions Per Patient Per Day (Units)
3.5
120
100
2.5
80
2
60
1.5
40
Number of Patients
Mean RBC Transfusions (Units)
3
1
20
0.5
0
0
1 2
3
4
5
6
7
8
9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28
ICU Days
Mean RBC Transfusion (Units)
Num b er of Patients
Fig 4 Mean RBC transfusions per patient per day (units).
To address the RBC transfusion needs of the critically ill, Hebert et al conducted a series of trials to
evaluate the effects of restrictive and liberal RBC
transfusion strategies on mortality and morbidity.9,28,35 The largest of these was the TRICC Trial,9
with 838 critically ill patients whose hemoglobin
levels were <9 mg/dL within 72 hours after admission
to the ICU: 418 were randomly assigned to a restrictive
strategy of RBC transfusion (hemoglobin <7 mg/dL),
and 420 were assigned to a liberal strategy (RBC transfusion if hemoglobin levels were <10 mg/dL). The
30-day mortality was similar in both groups, although
the rates were significantly lower with the restrictive
transfusion strategy among patients with Acute Physiology and Chronic Health Evaluation (APACHE) II
score of #20, and among patients <55 years of age,
but not among patients with clinically significant cardiac disease. They concluded that a restrictive RBC
transfusion strategy was at least as effective as, and
possibly superior to, a liberal strategy in the critically
ill, with the exception of acute myocardial infarction
and unstable angina patients. Although that was one
of the largest randomized, controlled trials to evaluate
outcomes in critically ill patients treated with a restrictive transfusion approach, blood administration
practices have not changed much in the last decade.
Furthermore, the recent Sepsis Occurrence in
Acutely Ill Patients Study36 did not find that
blood transfusions were associated with increased
HEART & LUNG VOL. -, NO. -
mortality in critically ill patients. Clearly, evidence is
controversial regarding transfusion safety or a specific
transfusion trigger.
PURPOSE OF STUDY
Our primary objective was to examine current
RBC transfusion practices for critically ill adults
admitted to our General Systems ICU. We addressed
the specific questions: 1) What is the prevalence of
anemia? 2) What is the rate of RBC transfusion? 3)
What are the reported indications for RBC transfusions? 4) What are the factors associated with
anemia and RBC transfusions?
METHODS
A cohort was used to examine RBC transfusion
practices in our ICU, and to assess the relationship
between anemia and RBC transfusion practices in
critically ill adults. Consecutive patients (n = 100)
admitted to the ICU over a 2-month period were included. Excluded were patients: 1) less than 18 years
of age, 2) who remained in the ICU for less than 24
hours, 3) with chronic endstage renal disease receiving exogenous erythropoietin, 4) of the Jehovah’s
Witnesses, who declined blood products, and 5)
with confirmed primary hematologic disease, including recent bone marrow transplantation or pancytopenia after chemotherapy. Ours is a 30-bed adult
www.heartandlung.org
5
ARTICLE IN PRESS
Anemia and Blood Transfusion in the Critically Ill
Thomas et al
Table III
RBC transfusion practices
RBC transfusions per patient (units)
Mode
Per day
Total length of stay
1.4
2
Median
1.5
2
Mean (SD)
Range
1.5 (.4)
1.66 (3.25)
1-2
0-19
RBCs units per transfusion order
Per day
Total length of stay
Mode
Median
1.25
2
1.4
2
Mean (SD)
Range
1.38 (.34)
2 (.5)
1-2
2-2
Pretransfusion hemoglobin (mg/dL) and number (percentage) of transfusions
<6.0
6.0-7.0
7.1-8.0
8.1-9.0
9.1-10.0
3 (3.6%)
29 (35%)
32 (38.5%)
15 (18%)
4 (5%)
Units per order and units transfused
Units per order
1
2
3
4
5
Total
Frequency (%)
26 (31)
52 (63)
3 (4)
1 (1)
1 (1)
83 (100)
General Systems ICU in a large Canadian academic
health center that admits approximately 1200
patients per year with a mean APACHE II score of
22. It is staffed by intensivists, medical residents,
nurses, respiratory therapists, physiotherapists, dieticians, pharmacists, social workers, service attendants, and students. The ICU does not have RBC
transfusion protocols.
Data-collection sheets were added to the
patient’s health record for nurses to prospectively
record date/time of blood collection, discard volumes, volumes of blood drawn for diagnostic purposes, and blood-transfusion events. Variables
included rate of anemia (hemoglobin <12 mg/dL
for women, and <14 mg/dL for men, per our laboratory), frequency of phlebotomy per patient per day,
associated volume of blood loss via phlebotomy
per patient per day, rate of RBC transfusions, number
of RBC units (volumes ranging from 250 to 350 mL,
with hematocrit levels at 50% to 80%) per transfusion, admission hemoglobin level, pretransfusion
hemoglobin level (hemoglobin level within 3 hours
before transfusion), and primary indication for RBC
6
www.heartandlung.org
Number of units
26
114
9
4
5
158
transfusion, as reported by the attending physician.
Each patient’s age, gender, admission diagnosis,
comorbidities, admission APACHE II scores, and
length of ICU stay were also recorded. Approval for
this study was obtained from our Health Research
Ethics Board.
Descriptive statistics were computed for all variables. To determine if a relationship existed between
phlebotomy blood loss, anemia, and RBC transfusions, and to examine relationships between age,
gender, comorbidities, illness severity (APACHE II
score), and length of ICU stay in the rate of anemia
and RBC transfusions, Pearson’s correlations were
used.
RESULTS
Patient characteristics
The mean age of patients was 56.5 14.8 (SD)
years (range, 18 to 85 years), with 19% (n = 19) above
70 years of age. Men accounted for 57% of this
cohort. The most frequent admitting diagnosis was
of complications related to the respiratory system
--/-- 2009
HEART & LUNG
ARTICLE IN PRESS
Thomas et al
Anemia and Blood Transfusion in the Critically Ill
Table IV
Indications for RBC transfusions
Indication
for transfusion
Table V
Number of RBC units transfused, based on
admission diagnosis
Frequency Percentage
Anemia
Acute bleeding
Cardiac history
Hypoxemia
APACHE II score
38
33
7
3
2
46
40
8
4
2
(41%). Of the comorbidities documented, a cardiac
history accounted for 16% of patients. The mean admission APACHE II score was 15 7, (SD) with
a range of 3 to 36. The mean ICU length of stay was
7.7 6.6 (SD) days, with a range of 1 to 28 days;
35% stayed longer than 7 days (Table I).
Admission
diagnosis
Respiratory
disorder
Multisystem
disorder
Gastrointestinal
disorder
Cardiovascular
disorder
Neurologic
disorder
Genitourinary
disorder
Other
Number of
RBC units Percentage
54
34
32
20.25
32
20.25
18
11.5
6
4
2
1
14
9
Course of anemia
The mean admission hemoglobin level was 10.8 0.22 (SD) mg/dL (range, 6.3-18.2 mg/dL), with 71%
having an admission hemoglobin level <12 mg/dL,
and 39% having an admission hemoglobin level
<10 mg/dL (Table I). The majority of patients (86%)
were anemic on admission to the ICU. The incidence
of anemia during their stay in the ICU was 98%. By
ICU day 8, 97% of the patients were anemic, and by
ICU day 13, 100% were anemic (Fig 1).
The number and volume of blood samples drawn
varied widely (Table II). The median number of phlebotomies per patient was 19.5 for total ICU stay, with
a mean of 3.5 1.04 (SD) per patient per day (Fig 2).
The mean blood loss per phlebotomy was 7.6 6.3
(SD) mL, with a mean of 25 10.3 (SD) mL per patient per day, and 223.6 183.4 (SD) mL per patient
for total ICU stay. The mean blood loss volume was
greatest on the day of admission, averaging 51.4 17.1 (SD) mL, and the lowest on the day of discharge,
averaging 16.6 11.3 (SD) mL (Fig 3).
Blood transfusion practices
The mean number of RBC units transfused per
patient per day was 1.5 0.4 (SD), for a mean of
1.66 0.5 (SD) during the total ICU stay, peaking on
ICU days 5 and 27 (Fig 4). The mean number of RBC
units per transfusion order per day was 1.38 0.34
(SD), and per total ICU stay, 2 0.5 (SD). There were
83 RBC transfusions during the study period, and in
total, 158 RBC units were transfused (Table III). The
majority (63%) of RBC transfusion orders were for 2
units per transfusion. The mean pretransfusion
HEART & LUNG VOL. -, NO. -
hemoglobin level was 7.3 0.47 (SD) mg/dL for this
cohort. The mean pretransfusion hemoglobin level
was 7.5 4.1 (SD) mg/dL for those with acute bleeding, and 7.2 0.54 (SD) mg/dL for those without acute
bleeding. Some (39%) RBC transfusions occurred in
the pretransfusion hemoglobin range of 7.1 to 8.0
mg/dL, with pretransfusion hemoglobin levels exceeding 9.0 mg/dL in only 5% of cases.
Patients with a cardiac history were transfused at
a hemoglobin level of 8.2 0.65 (SD) mg/dL, for
a transfusion rate of 8%. Table IV summarizes the indications for RBC transfusion. Patients admitted to
the ICU with respiratory disorders received up to
34% of transfusions, whereas those with cardiovascular disorders received 11.5% of transfusions (Table V).
The RBC transfusion rate during the total ICU stay
was 40%, with a daily transfusion rate of 13.6%. Patients whose length of stay was >7 days had an overall RBC transfusion rate of 70%. Transfusion rates
were consistent across age groups, with 37.5% in patients aged <50 years, and 42.5% in those aged >50
years. There was no difference in RBC transfusion
rate based on gender. Over 39% of patients received
their first RBC transfusion within their first 3 days
in ICU. The RBC transfusion rate was 11% at admission, and peaked on ICU day 19 at 71%. The mean
length of stay for those who received RBC transfusions was 11.45 days, and for those who did not receive RBC transfusions, 5.2 days. Other blood
products transfused were plasma (45%), albumin
(35%), platelets, and cryoprecipitate during the study
period. The rate of transfusion of other blood
www.heartandlung.org
7
ARTICLE IN PRESS
Anemia and Blood Transfusion in the Critically Ill
products was 18% on ICU day 1, which peaked on ICU
days 18 and 19 at 43%.
DISCUSSION
The prevalence of anemia in this ICU cohort was
similar to rates reported in other studies.2,4 Furthermore, patients with lower admission hemoglobin
levels became more anemic as their length of ICU
stay progressed (r = .66, P < .01). Patients with lower
admission hemoglobin levels had a longer ICU stay
(r = .16, P < .01) and higher admission APACHE II
scores (r = .17, P < .01). Gender, age, admission diagnoses, comorbidities, and phlebotomy practices did
not influence the rate of anemia. Pearl and Pohlman15
and Parillo1 documented the impact of nutritional deficiencies, decreased renal function, sepsis, and inappropriately low erythropoietin levels in the critically ill
on anemia. It is not known whether an inadequate
erythropoietin response contributed to the anemia
in this cohort. Only 2% of patients who were anemic received oral iron therapy, whereas none received folate,
vitamin B12, or exogenous erythropoietin. Anemia in
the ICU setting was also attributed to frequent phlebotomies.24 In our ICU, the mean phlebotomy blood
loss per patient per day was 25 10.3 (SD) mL, and
anemia increased with length of ICU stay.
The RBC transfusion rate in our ICU was 40%. As
length of stay in the ICU increased, the RBC transfusion rate increased (r =.78, P < .01). Patients with an
ICU stay >7 days had an overall transfusion rate of
70%, with a peak on ICU day 19. This is similar to
rates reported in other studies.2,26 Severity of illness
on ICU admission influenced RBC transfusion rates
(r =.24, P < .05), with 62% of RBC transfusions occurring in patients with APACHE II scores of 13 to 24,
and 19% in patients with APACHE II scores >25.
Phlebotomy practices also influenced RBC transfusion rates. As the volume of blood loss increased
through phlebotomies, the total number of RBC
transfusions also increased (r =.80, P < .01). Of those
patients who required RBC transfusions, 70% had
a phlebotomy-related blood loss of >300 mL for their
total ICU stay, which is equivalent to one unit of
RBCs. Patients who received >5 units of RBCs during
their ICU stay had a mean phlebotomy-related blood
loss of >700 mL. The mean pretransfusion hemoglobin level was 7.35 .47 (SD) mg/dL for this cohort.
The pretransfusion level of hemoglobin for patients
with a cardiac history was 8.2 .65 (SD) mg/dL.
These levels are lower than reported in other studies.2,7,11,26 This may be attributable to the findings
of the TRICC trial.9 In our ICU, the pre-RBC transfusion hemoglobin levels were similar for those with
8
www.heartandlung.org
Thomas et al
and without acute bleeding, suggesting that RBC
transfusion decisions were determined by hemoglobin levels rather than clinical diagnosis per se.
Gender, age, and comorbidities did not play a significant role in reported RBC transfusion decisions.
Limitations
Physicians in the ICU were aware that phlebotomy
and RBC transfusion practices were being recorded,
which could have led to temporary modifications in
behavior. There was no validation of the accuracy of
the reported volume of blood drawn for phlebotomy.
Moreover, blood loss via hemodialysis filters and dialyzers was not documented, and RBC transfusions in
surgical patients did not include transfusion events
that may have occurred in the operating room. Clinical
outcomes of patients transfused at lower pretransfusion levels of hemoglobin were not assessed.
CONCLUSION
Anemia occurred in 98% of critically ill patients,
and hemoglobin levels continued to drop with ICU
stay. Despite phlebotomy-related blood loss, there
was no correlation between anemia and phlebotomy
practices. However, this finding should not preclude
strategies to reduce phlebotomy-related blood loss.
These include using multichannel microchemistry
instruments for point-of-care testing, pediatric tubes
for blood collection, inline blood-conservation devices to eliminate discard volumes, the timing and
batching of phlebotomies, elimination of routine
ICU standing orders for phlebotomies, and a daily
assessment of frequency of phlebotomies. In this
cohort, the pretransfusion hemoglobin level was
7.35 .47 mg/dL, a value lower than reported in
other studies. However, the RBC transfusion rate
was 40%, a rate similar to that in other studies.2,4,22
The management of anemia in the ICU may need to
emphasize the complementary roles of decreasing
phlebotomy-related blood loss, increasing RBC production with exogenous hematopoietic agents, and
decreasing RBC transfusions, by using lower hemoglobin transfusion triggers.
REFERENCES
1. Parillo J. Anemia management, and evidence-based critical
care medicine. Crit Care Med 2001;29(Suppl):S139.
2. Vincent JL, Baron J, Reinhart K, et al. Anemia and blood transfusion in critically ill patients. JAMA 2002;288:1499-507.
3. Brown M, Whalen P. Red blood cell transfusion in critically ill
patients: emerging risks and alternatives. Crit Care Nurse
2000;20(Suppl):1-14.
4. von Ahsen N, Muller C, Serke S, Frei U, Eckardt K. Important
role of nondiagnostic blood loss and blunted erythropoietic
--/-- 2009
HEART & LUNG
ARTICLE IN PRESS
Thomas et al
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
response in the anemia of medical intensive care patients.
Crit Care Med 2001;29:141-59.
Zimmerman JE, Seneff MG, Sun X, Wagner DP, Knaus WA.
Evaluating laboratory usage in the intensive care unit: patient
and institutional characteristics that influence frequency of
blood sampling. Crit Care Med 1997;25:737-48.
Pohlman AS, Carven JH, Lindsay K. Conserving blood in the
intensive care unit. Crit Care Nurse 2001;21(Suppl):1-14.
Surgenor SD, Hampers MJ, Corwin HL. Is blood transfusion
good for the heart? Crit Care Med 2001;29:442-4.
Hebert PC, Yetisir E, Martin C, et al. Is a low transfusion
threshold safe in critically ill patients with cardiovascular
diseases? Crit Care Med 2001;29:227-34.
Hebert PC, Wells G, Blajchman MA, et al. A multicenter,
randomized, controlled clinical trial of transfusion requirements in critical care. N Engl J Med 1999;340:409-17.
Canadian Medical Association. Expert working group:
guidelines for red blood cell and plasma transfusion for
adults and children. Can Med Assoc J 1997;156(Suppl 11):
S1-24.
Hebert PC, Wells G, Martin C, et al. Canadian survey of transfusion
practices in critically ill patients. Crit Care Med 1998;26:482-7.
Boralessa H, Rao M, Morgan C, et al. A survey of physicians’
attitudes to transfusion practice in critically ill patients in
the UK. Anaesthesia 2002;57:584-8.
Corwin HL, Gettinger A, Pearl RG, et al. Efficacy of recombinant human erythropoietin in critically ill patients: a randomized controlled trial. JAMA 2002;288:2827-35.
Carson JL. Should patients in intensive care units receive
erythropoietin [editorial]? JAMA 2002;288:2884-6.
Pearl RG, Pohlman A. Understanding and managing anemia in
critically ill patients. Crit Care Nurse 2002;26(Suppl):1-12.
Hebert PL, Fergusson DA, Stather D, et al. Revisiting transfusion
practices in critically ill patients. Crit Care Med 2005;33:7-12.
Turgeon AF, Fergusson DA, Doucette S, et al. Red blood cell
transfusion practices amongst Canadian anesthesiologists:
a survey. Can J Anesth 2006;53:344-52.
Corwin HL. Anemia of the critically ill: ‘‘acute’’ anemia of
chronic disease. Crit Care Med 2000;28:3098-9.
Tarpey J, Lawler PG. Iatrogenic anemia: a survey of venesection in patients in the intensive therapy unit. Anaesthesia
1990;45:396-8.
Dale JC, Pruett SK. Phlebotomy: a minimalist approach. Mayo
Clin Proc 1993;68:249-55.
HEART & LUNG VOL. -, NO. -
Anemia and Blood Transfusion in the Critically Ill
21. Ba VN, Bota DP, Melot C, Vincent J. Time course of hemoglobin
concentrations in nonbleeding intensive care unit patients.
Crit Care Med 2003;31:406-10.
22. Vincent JL, Baron J-F, Reinhart K, et al. Anemia and blood
transfusion in critically ill patients. JAMA 2002;288:
1499-507.
23. Hebert PC, Schweitzer I, Calder L, Blajchman M, Giulivi A.
Review of the clinical practice literature on allogenic red
blood cell transfusion. Can Med Assoc J 1997;56(Suppl):S9-26.
24. Corwin HL, Parsonnet KC, Gettinger A. RBC transfusion in the
ICU: is there a reason? Chest 1995;108:767-71.
25. Corwin HL. Anemia in the critically ill: the role of erythropoietin. Semin Hematol 2001;38(Suppl 7):24-32.
26. Corwin HL, Gettinger A, Pearl RG, et al. The CRIT Study:
anemia and blood transfusion in the critically ill: current clinical practice in the United States. Crit Care Med 2004;32:39-52.
27. Sherk PA, Granton JT, Kapral MK. Red blood cell transfusion in
the intensive care unit. Intensive Care Med 2000;26:344-6.
28. Hebert PC, Wells G, Tweeddale M, et al. Does transfusion practice affect mortality in critically ill patients? Am J Respir Crit
Care Med 1997;155:1618-23.
29. Rivers E, Nguyen B, Havstad S, et al. Early goal-directed therapy in the treatment of severe sepsis and septic shock. N Engl
J Med 2001;345:1368-77.
30. American College of Physicians. Practice strategies for elective
red blood cell transfusion. Ann Intern Med 1992;116:403-6.
31. American Society of Anesthesiologists Task Force on Blood
Component Therapy. Practice guidelines for blood component
therapy. Anesthesiology 1996;84:732-47.
32. Adam RC, Lundy JS. Anesthesia in cases of poor risk: some
suggestions for decreasing the risk. Surg Gynecol Obstet
1942;74:1011-9.
33. Stehling LC, Esposito B. An analysis of the appropriateness of
intraoperative transfusion. Anesth Analg 1989;68(Suppl):
S278-9.
34. Stehling LC, Ellison N, Faust RJ, Grotta AW, Moyers JR. A survey of transfusion practices among anesthesiologists. Vox
Sang 1987;52:60-2.
35. Hebert PC, Wells G, Marshall J, et al. Transfusion requirements
in critical care: a pilot study. JAMA 1995;273:1439-44.
36. Vincent JL, Sakr Y, Sprung C, et al. Are blood transfusions
associated with greater mortality rates? Results of the Sepsis
Occurrence in Acutely Ill Patients Study. Anesthesiology 2008;
108:31-9.
www.heartandlung.org
9