Med. J. Cairo Univ., Vol. 88, No.

5, December: 2121-2129, 2020

www.medicaljournalofcairouniversity.net

Anemia in Critically Ill Patients; Prevalence and Prognostic

Implications
NAGLAA M. ALY, M.D.; MANAL M. KAMAL SHAMS ELDIN, M.D.;
WAEL S.A. ABD-ELGHAFAR, M.D. and KAREEM M. FAWZY, M.Sc.
The Department of Anesthesiology, Intensive Care and Pain Management, Faculty of Medicine, Ain Shams University

Abstract group 39.7% of them died with hemoglobind 7g/dl which was
chosen as a best predictor to outcome.
Background: Anemia is a commonly encountered clinical
problem in the critically ill. Ninety-five percent of critically Conclusion: It was found that there is no association
ill patients who stay in the Intensive Care Unit (ICU) for 72 between management of anemia (blood transfusion and iron
hours or greater suffer from anemia and approximately 40% supplementation) and the outcome of the patient and that
of them receive Packed Red Blood Cell (PRBC) transfusions. patient's hemoglobin at the end of ICU stay is of good prog-
In 2001, nearly 14 million units of packed red blood cells nostic value of anemic patients.
were transfused, but the physiologic basis for transfusion in
Key Words: Anemia in critically Ill patients – Prevalence and
the critically ill is not without controversy. In the last two
prognostic implications.
decades transfusion practices have become more restrictive
likely in response to prospective research.
Introduction
Aim of Study: The aim of this work is to evaluate the
prevalence of anemia in critically ill patients and to assess ANEMIA of critical illness is defined as anemia
the effect of anemia of critical illness on the patients' outcome. in the critically ill patient that cannot be explained
Patients and Methods: The study was conducted from by other causes and that is characterized by an
the existing data base in the Critical Care Department of inadequate response of endogenous erythro-
Elsahel Teaching Hospital from January 2015 to December poietin in relation to the degree of hemoglobin
2018. The collected data focused on fulfilling the following
measures on the selected patients: Comparison of age, length deficiency present [1] . The prevalence of anemia
of ICU stays of the patients and APACHE II score in survivors in the Intensive Care Unit (ICU) warrants a detailed
and non survivors groups, frequency of blood transfusion and evaluation and review of the available therapeutic
iron supplementation in the management of anemia of critical options [2,3].
illness and its association with the outcome, the relation of
follow-up complete blood count and the outcome. During an average ICU stay, a critically ill
Results: In this retrospective cohort study involving 165 patient can lose a total of 762ml of blood to labo-
patients 74 males and 91 females with mean age 55. 13±20.72, ratory tests [1]. A recent report from ICUs in West-
and mean length of stay 11, 87±12,04. The survivors were ern Europe demonstrated an average total phlebot-
103 (62.5%), and those who not survive were 62 (37.5%) the
–1
mean hemoglobin at admission was 8.28g dl (±1.96) for
omy volume of 41.1ml during a 24-hour observa-
survivors group and 8.27 (±2.36) for non survivors. reasons tion period. Another study in trauma patients sug-
of admission which were associated with higher frequency gested that laboratory testing is becoming more
of mortality among the others are post cardiac arrest, respiratory frequent with an increase in the number of blood
failure, neurologic problems, cardiogenic shock, septic shock, tests ordered [4] .
acute renal failure and obstetrics catastrophes with frequency
(85.7%, 77.8%, 66.7%, 60%, 55%, 45.5%, 60%) respectively.
Regarding blood transfusion as in management of anemia, in
Critically ill patients are exposed to blood loss
this study there were no significant difference in relation of that will inevitably participate in the onset or
blood transfusion and the outcome (our hemoglobin threshold worsening of anemia [5], but also potentially to a
was 7g/dl). In this study we had 78 patients admitted for more true iron deficiency [6]. A recent study in Australia
than fourteen days in the ICU they represent 25% of our study and New Zealand reported that bleeding was the
reason for transfusion in 46% of transfusion events
Correspondence to: Dr. Naglaa M. Aly, The Department of [7]. Blood loss is rarely the only explanation for
Anesthesiology, Intensive Care and Pain Management, anemia. During resuscitation with colloid and
Faculty of Medicine, Ain Shams University crystalloid solutions hemodilution contributes to

2121
2122 Anemia in Critically Ill Patients

the rapid decrease in hemoglobin concentration has not been proved either for outcome or for
seen early after ICU admission in many critically improving tissue hypoxia [16].
ill patients without altering red cell mass [8].
Aim of the study:
A major factor resulting in the development The aim of this work is to evaluate the preva-
and persistence of anemia is reduced new red blood lence of anemia in critically ill patients and to
cell production (erythropoiesis) which occurs in assess the effect of anemia of critical illness on
the bone marrow and is controlled by erythropoietin the patients' outcome.
hormone [9]. This appears to result from a combi-
nation of inappropriately low circulating erythro- Patients and Methods
poietin and hypo-reactive bone marrow. Inflamma-
tion is responsible for a transcriptional repression This retrospective historical cohort study was
of erythropoietin synthesis, mediated by many conducted from the existing data base in the Critical
proinflammatory cytokines (including TNF-α and Care Department of Elsahel Teaching Hospital
IL1). In addition, erythropoietin receptor expression from January 2015 to December 2018 and was
is suppressed by inflammatory cytokines [10] . conducted on one hundred and sixty five-critical
care patients.
Hemolysis may also cause anemia in critically
ill patients. This may be associated with several Sample size was calculated using PASS pro-
pathologic conditions, including bacterial infec- gram, setting alpha error at 5% and confidence
tions, malaria, trauma and conditions in which interval width at 0.15. Result from previous study
mechanical forces can lead to RBCs rupture, such (Vincent et al., 2002) showed that 60% of patients
as surgical procedures, hemodialysis and blood admitted to ICU get anemia [17].
transfusion. Hemolysis results in release of free Inclusion criteria:
plasma hemoglobin and heme, which are toxic to
The study was conducted on the patients admit-
the vascular endothelium [11] . Hypersplenism may
ted to the Critical Care Department with anemia
also lead to excessive RBCs destruction. As hemo-
or developed anemia during their ICU course.
lytic anemia occurs because of intrasplenic destruc-
tion of erythrocytes [12] . Exclusion criteria:
The pathophysiologic consequences of anemia 1- Hemorrhagic anemia due to surgical causes.
in critically ill patients include inadequate tissue 2- Congenital causes of anemia as sickle cell tha-
oxygenation and eventual ischemia of end organs. lassaemia.
Inadequate tissue oxygenation results from either
decreased oxygen delivery and/or increased tissue 3- Hemolytic anemia as a direct cause of ICU
oxygen consumption. Oxygen delivery is a function admission.
of arterial oxygen content and cardiac output. Data collection focused on fulfilling the follow-
Hemoglobin concentration and Oxygen saturation ing measures on the selected patients:
both affect arterial oxygen content [13]. Oxygen
consumption is the rate at which tissues take up • Comparison of age, length of ICU stays of the
oxygen and is a function of oxygen delivery and patients and APACHE II score in survivors and
the amount of oxygen that is extracted by tissues non survivors groups.
(i.e., the oxygen extraction ratio). • Frequency of the reasons of admission and its
association with the outcome.
The 'critical hemoglobin concentration' is usu-
ally defined as the concentration below which • Frequency of patients with anemia of critical
oxygen consumption is supply-dependent assuming illness mechanically ventilated and had vasoactive
normovolaemia is maintained [14,15] . This is un- drugs.
likely to be a fixed value, but varies between organs • Frequency of blood transfusion and iron supple-
and is dependent on the metabolic activity of the mentation in the management of anemia of critical
tissue and oxygen extraction capabilities. illness and its association with the outcome.
Despite these factors anemia is well tolerated • The relation of follow-up complete blood count
by critically ill patients and a hemoglobin of 7-9 and the outcome.
g/dl does not adversely affect outcome in compar-
–1
ison with maintaining a value >10g dl . The Clinical end point:
effectiveness of red blood cell transfusions in –1
All clinical events were reviewed and docu-
patients with hemoglobin concentration >7-8g dl mented. The clinical end point was the documented
Naglaa M. Aly, et al. 2123

end of ICU stay either by death of the patient or Results

improvement discharge or DAMA (discharge
Table (1): Demographic disruptions among the study.
against medical advises) document.
Variable Mean (± std. deviation)
Statistical methods:
• Data were verified and coded prior to analysis. Age 55.13 (±20.72)
Length of ICU stay in days 11.87 (±12.049)
• All quantitative data were expressed as mean ±
SD. Number (%)
• All qualitative data were expressed as frequency Males 74 (44.5%)
tables.
Females 91 (55.5%)
• Chi-square test to confirm the presence of asso- Survivors 103 (62.5%)
ciation between different categorical data. Student Non survivors 62 (37.5%)
t-test to compare between quantitative data.
Total 165 (100%)
• p-value <0.05 considered significant.
• Analysis has been performed using SPSS (statis- The survivors were 103 (62.5%), and non sur-
tical package for social science). vivors were 62 (37.5%).

Table (2): Comparison of age, length of ICU stays of the patients and APACHE II score in survivors and non survivors groups.

Survivors Non survivors

p-
t
Number Mean (std.deviation) Number Mean (std.deviation) value

Age 103 55.6809 (±19.79478) 62 54.2301 (±22.24218) 0.571 0.569

Length of ICU stay in days 103 11.7500 (±11.09301) 62 12.0708 (±13.54142) 0.223 0.832
APACHEII 103 15.617 (±5.80910) 62 20.8850 (±7.64707) 6.310- 0.0001

An independent t-test was conducted to compare groups. A significant higher value of APACHE
age, length of ICU stay, APACHE II score, and was found in non survivors (20.88±7.647 versus
blood transfusions in survivors and non survivors 12.61±5.809, p=.0001).
Table (3): The association of reasons of admission and out comes.

Survivors Non survivors Total n (%) p-

χ
2
N (% in reason) Outcome % N (% in reason) Outcome % within total value

Post cardiac arrest 1 (14.3%) .5% 2 (85.7%) 5.3% 3 (2.3%) 52.073 .0001
Acute coronary syndrome 31 (80.6%) 30.9% 8 (19.4%) 12.4% 39 (23.9%)
Acute heart failure 14 (81.3%) 8.6% 4 (18.8%) 2% 18 (10.6%)
Arrhythmias 8 (83.3%) 8% 2 (16.7%) 2.7% 10 (6%)
Other cardio-vascular 5 (66.7%) 4.3% 2 (33.3%) 3.5% 7 (4%)
Respiratory failure 3 (22.2%) 2.1% 7 (77.8%) 12.4% 10 (6%)
Other respiratory 2 (50%) 1.6% 2 (50%) 2.7% 4 (2%)
Neurologic problems 4 (33.3%) 3.2% 6 (66.7%) 10.6% 10 (6%)
Hypovolumic shock 17 (66%) 16.5% 9 (34%) 14.2% 26 (15.6%)
Cardiogenic shock 2 (40%) 2.1% 4 (60%) 5.3% 6 (3.3%)
Septic shock 5 (45%) 4.8% 6 (55%) 9.7% 11 (6.6%)
Acute renal failure 4 (54.5%) 3.2% 2 (45.5%) 4.4% 6 (3.7%)
Obstetric catastrophes 1 (40%) 2.1% 3 (60%) 5.3% 4 (3.3%)
Others 6 (65%) 6.9% 5 (35%) 6.2% 11 (6.6%)

Total 103 (62.5%) 100% 62 (37.5%) 100% 165 (100%)

A chi-square test of independence was calcu- failure and obstetrics catastrophes were associated
lated comparing the frequency of reasons of ad- with higher frequency of mortality among the total
missions named before and the outcomes. Post number of each reason; (85.7%, 77.8%, 66.7%,
cardiac arrest, respiratory failure, neurologic prob- 60%, 55%, 45.5%, 60%) respectively.
2
A significant
lems, cardiogenic shock, septic shock, acute renal inter action was found χ [13]=52.073 p<.05.
2124 Anemia in Critically Ill Patients

Table (4): Frequency of patients mechanically ventilated and while 42 (25.4%) where on both inotropics and
had vasoactive drugs. vasopressors.
Variables Number (%)
A non significant interaction was found com-
MV 77 (46.8%) paring the frequency of blood transfusions, and
No vasoactive drugs 79 (47.9%) iron supplementation in management of anemia in
Both vasoactive drugs 42 (25.4) critical illness and the outcome using chi-square
2
Inotropic drugs only 7 (4.3%) test of independence. (χ [6]=10.044, p=.123, χ 2
Vasopressor drugs only 37 (22.4%) [1]=.563, p=.453) respectively.
Total 165 (100%)
An independent-sample t-test was conducted
to compare complete blood count and follow-up
Out of 165 cases 77 (46.8%) cases were me- of hemoglobin and HCT in survivors and non
chanically ventilated, 86 (52.5%) cases had taken survivors. A significant difference were in hemo-
vasoactive drugs, patients had inotropic drugs globin in 14th day, hemoglobin at the end of ICU
only where 7 (4.3%), 37 (22.4%) cases had taken stay, WBCs at admission, platelets at admission
vasopressors drugs only as hemodynamic support, p<.05.

Table (5): The association between management of anemia of critical illness and the outcome.

Chi square/fisher exact

Non Total n (%) within
Survivors
survivors total patients χ2 p-value

No blood transfusion 40 (24.2%) 23 (14%) 63 (38.2%) 10.044 .123

Blood transfusion 63 (38.3%) 39 (23.5%) 102 (61.8%)
Total 103 (62.5%) 62 (37.5%) 165 (100%)
No iron supplementations 70 (42.5%) 45 (26.6%) 114 (69.1%) .563 .453
Iron supplementation 33 (20%) 18 (10.9%) 51 (30.9%)
Total 103 (62.5%) 62 (37.5%) 165 (100%)

Table (6): The relation of follow-up complete blood count and the outcome.

Number Mean (std. deviation) Number Mean (std. deviation) t p-value

Hemoglobin at admission 103 8.28 (±1.967) 62 8.27 (±2.368) .039 .969

Hemoglobin in 3 rd day 93 8.98 (±4.51) 49 9.09 (±7.82) .140- .889
Hemoglobin in 7th day 61 9.4 (±7.710) 33 9.40 (±8.741) .040 .968
Hemoglobin in14th day 22 8.56 (±1.223) 16 7.706 (±.915) 3.445 .001
Hemoglobin at discharge 94 9.26 (±1.075) 56 8.11 (±1.820) 6.112 .0001
WBCs at admission 99 11.53 (±6.294) 66 14.34 (±11.025) 2.476- .014
Platelets counts at admission 95 230.04 (±133.72) 70 182.44 (±125.45) 3.060 .002

Discussion cohort study of 3534 patients admitted to 146

Western European ICUs with varying case mix
Anemia is highly prevalent in critically ill and (the ABC study; anemia and blood transfusion in
injured patients. Approximately two-thirds present critical care trial) found that the mean hemo-
with a hemoglobin concentration less than 12g/dl globin concentration at ICU admission was 11.3g
on admission, and 97% become anemic by Day 8 –1
dl [17].
[17,18]. Optimal management of the anemia of crit-
ical illness is an area of much controversy and A similarly designed study in the USA examined
ongoing research. 4892 admissions to ICUs (the CRIT study; Anemia
The prevalence of anemia among critically ill and blood transfusion in the critically ill-current
patients is influenced by factors that include patient clinical practice in the United States) [3] . In this
case mix, illness severity and pre-existing co- study the mean hemoglobin–1
concentration at ICU
morbidity. admission was 11.0g dl .

Several recent studies have documented the A cohort study of 1023 sequential admissions
prevalence of anemia on admission to ICU. A to 10 Scottish ICUs found that the median hemo-
Naglaa M. Aly, et al. 2125

globin
–1
concentration at ICU admission –1
was 10.5g decrease in hemoglobin values was also found in
dl (inter quartile range 9.0-12.4g dl ) [19]. The a prospective observational single center cohort
authors also showed that the patients studied rep- study of patients receiving >24h of intensive
resented 44% of all general adult ICU admissions care [20].
nationally over the study period. At ICU admission,
25% of–1patients had a hemoglobin concentration Also, in the CRIT study in USA ICUs, the mean
<9g dl (range 16-34% across the 10 ICUs). hemoglobin concentration in a cohort of non- –1
transfused patients decreased
–1
from ~12g dl at
Among ICU survivors and non-survivors 21 admission to 11g dl by days 3-4, after which
and 29%, respectively, of patients had an
–1
admis- values reached a plateau among patients remaining
sion hemoglobin concentration <9g dl at ICU in the study [3] .
admission.
Patients among this study how were admitted
The prevalence of anemia at ICU admission because of cardiac diseases represented 50.1%of
varies, but it appears that 20-30% of patients have the total patients among the study and the non
moderate to severe
–1
anemia (hemoglobin concen- survivors group of them represent 31.2% of the
tration <9g dl ). Only 10-15% had documented total mortality outcome while those who were
pre-existing anemia. admitted due to respiratory causes represent 8%
of the total and 15.1% of the total mortality out-
In this retrospective cohort study involving 165 come, 6% of the total admitted due to neurogenic
patients 74 males and 91 females with mean age problems, 15.6% of total had hypovolemic shock,
55.13±20.72, and mean length of stay 11,87±12,04. 6.6% had septic shock, and 6.6% had obstetric
The survivors were 103 (62.5%), and those catastrophes.
who not survive were 62 (37.5%) the –1
mean hemo- In this study, reasons of admission which were
globin at admission was 8.28g dl (±1.96) for associated with higher frequency of mortality
survivors group and 8.27 (±2.36) for non survivors, among the others are post cardiac arrest, respiratory
which is lower than the mean hemoglobin in the failure, neurologic problems, cardiogenic shock,
ABC study which was (11.3g/dl) [49], and in the septic shock, acute renal failure and obstetrics
CRIT study (11.0g/dl) [3] , but correlate with the catastrophes with frequency (85.7%, 77.8%, 66.7%,
25% of the patients among the multicenter Scottish 60%, 55%, 45.5%, 60%) respectively.
study who had hemoglobin concentrations <9g/dl.
Numerous recent studies have shown anemia
The prevalence and severity of anemia during to be associated with worse outcomes in patients
ICU admission is clearly linked closely with the with coronary artery disease. After reviewing the
transfusion practice used. The evolution of anemia data on nearly 40,000 patients enrolled in trials on
among non-transfused, non-bleeding, critically ill Acute Coronary Syndrome (ACS) Sabatine et al.,
patients is difficult to study both ethically and in found anemia to be associated with a greater like-
practice. lihood of death in patients with ST-Segment Ele-
Fifty-two per cent of patients in the TRICC vation MI (STEMI) and Non-ST-Segment Elevation
study (Transfusion Requirements in Critical Care MI (NSTEMI) [22]. These investigators also found
trial) had a hemoglobin concentration ≤9g dl–1 on an increased association of anemia with recurrent
the first day of ICU care, increasing to 77% by ischemia or acute MI in patients with NSTEMI.
second day [20], this result support our study in Aronson et al., found that lower nadir hemoglobin
which the mean hemoglobin concentration in the in hospitalized patients following MI were strongly
3rd day after admission were 8.98 (±4.51) for associated with increased mortality [23] and those
survivors group and 9.09 (±7.82) for non survivors results correlate with this study.
group which was a little higher than the mean in According to respiratory failure and mechani-
the first day of admission. cally ventilated patients, Nevins and Epstein found
However, Nguyen and colleagues found that anemia (mean hematocrit 36) to be associated with
among non-bleeding ICU patients who did not poor outcomes in a retrospective study of 166
receive red cell transfusions hemoglobin concen- patients with Chronic Obstructive Pulmonary Dis-
–1 –1
trations decreased by a mean 0.52g dl day [21]. ease (COPD) receiving mechanical ventilation
(type 2 respiratory failure) [24] . Although, data
On average, hemoglobin
–1 –1
concentrations de- exists showing anemia to be associated with poorer
creased by 0.66g
–1
dl –1
day for the first 3 days outcomes in mechanically ventilated patients, no
and by 0.12g dl day thereafter. This early rapid significant literature supports the transfusion of
2126 Anemia in Critically Ill Patients

Packed Red Blood Cells (PRBC) to facilitate wean- difference in relation of blood transfusion and the
ing patients from mechanical ventilation [25]. outcome (our hemoglobin threshold was 7g/dl).
Khamiees et al., in a prospective observational In the study there were 63 patients out of 165
study found anemia (Hgb <10g/dL) to be associated had no blood transfusion, 23 patients of them not
with extubation failure in a mixed medical-surgical survive represent 14% of the total, while there
ICU population [26] . were 102 patients had blood transfusion during
their ICU sitting, 40 patients of them not survive
Analyses of the medical arm of the National
represent 23.6% of the total patients. This result
Emphysema Treatment Trial [27] and of patients
had an agreement with the appropriate RBC trans-
in the French ANTADIR database with severe O2-
fusion thresholds. In the Transfusion Requirements
requiring COPD [28] identified anemia as an inde-
in Critical Care (TRICC) where [34], 838 euvolemic
pendent predictor of death.
patients without chronic anemia, myocardial
Anemia is associated with worse outcomes in ischemia, or on-going bleeding were randomized
nontraumatic subarachnoid hemorrhage (ruptured to either a restrictive or liberal transfusion strategy
brain aneurysm) [29] . Preventing brain hypoxia (threshold hemoglobin, 7 vs. 10g/dl). No difference
might be important to reduce the incidence and in the primary outcome of all-cause 30-day mor-
severity of cerebral infarction from vasospasm, tality was observed between treatment arms. Sub-
and P-RBC transfusion in that setting leads to group analyses identified patients less than 55
improved markers of brain tissue function on pos- years old and with APACHE II scores less than 20
itron emission tomography [30]. as having decreased 30-day mortality with a re-
strictive strategy. Although results of this trial have
The BOOST2 study is planned to assess if brain affected both guidelines and common practice,
oxygen tension-guided therapy improves outcomes; controversy still exists regarding specific patient
Intracerebral hemorrhage does not lead to vasos- groups: The elderly, and those with cardiovascular
pasm, but cerebral infarction can be found on disease, with difficulty being liberated from me-
magnetic resonance imaging scans [31] and this chanical ventilation, and in the early phase of septic
may impact outcomes. There is probably not hy- shock.
poxia around the clot [32], but there may be altered
metabolism for a period of several days [33] . Several groups have examined the association
of transfusions and mortality in large data sets of
Anemia is also a common occurrence in the patients with acute coronary syndromes. Of these,
setting of sepsis [34]. This is in part because medi-
one found transfusions to have a beneficial effect
ators of sepsis (e.g., TNF-α and IL-1β) decrease
on survival when the hematocrit was less than 33%
expression of the erythropoietin gene and protein [38]. In contrast, four studies found transfusions to
[35] . Goal-directed therapy during early severe
be an independent predictor of greater short-term
sepsis discuss the higher mortality rate of anemic
mortality [39,40]; one identified a threshold hema-
patients with sever sepsis and the improvement of
tocrit of 25%, above which transfusions were
mortality in a well-performed single center rand-
associated with increased risk of death [41] . It is
omized trial [36]. The intervention algorithm used
difficult to exclude confounding in such studies,
central venous oxygen saturation <70% as a trigger
and further trials of transfusion thresholds among
for interventions to increase global oxygen delivery.
patients with ischemic heart disease are needed.
Part of this algorithm was blood transfusion to
maintain a haematocrit ≥30% (haemoglobin ≥10
–1 Regarding effect of blood transfusion as a
g dl ), but it is unclear how important this com-
management of anemia on the ventilated patients
ponent was to improving mortality [37].
in our study there was no significant in mortality
In the TRICC trial, subgroup analyses of older between ventilated patients who received blood
or more severely ill patients showed no difference transfusion and those who did not received, al-
in 30-day mortality between the restrictive-strategy though ventilation affect the outcome and had a
and liberal strategy groups. Younger and less ill high mortality blood transfusion did not improve
patients had better outcomes if transfused in a the outcome.
more restrictive manner (i.e., at lower hemoglo-
bin) [34]. This result had a disagreement with Schönhofer
and colleagues found that transfusion to goal he-
Regarding blood transfusion as in management moglobin greater than 11g/dl decreased ventilation
of anemia, in this study there were no significant and work of breathing [42].
Naglaa M. Aly, et al. 2127

In a study of five anemic patients with COPD 2- It is associated with higher health care resource
(mean hemoglobin, 8.7g/dl) who were unable to use.
be liberated from mechanical ventilation (28-d
3- It may be associated with poor patient outcomes;
mean duration of ventilation; range, 13 to 49d)
and
[43], all were successfully extubated within 4 days
of being transfused to a mean hemoglobin level of 4- There is no currently available therapy without
12.4g/dl. shortcomings.
Also, the TRICC trial included 713 patients on According to data collected we recommended
mechanical ventilation, of whom 219 were venti- the following;
lated for greater than 7 days. In these subgroups, a- As much as possible decrease unnecessary blood
there was no difference in duration of mechanical sample tests; the introduction of blood conser-
ventilation or mortality between the two transfusion vation sampling devices should be considered
strategies [44] . This analysis had power only to to reduce phlebotomy-associated blood loss as
detect 25% differences in duration of mechanical much as possible decrease unnecessary blood
ventilation, so a clinically important difference sample tests.
may have been missed. Transfusion is most likely
to be beneficial in patients with the most severe b- Activate non invasive hemodynamic monitoring.
ventilatory impairment and respiratory muscle c- As much as possible shorten the duration spent
weakness, and it remains to be determined whether in ICUs.
and when a more liberal transfusion strategy is
warranted in these patients.
References
In this study, there were 51 patients had iron
supplementation; 17 of them not survived represent 1- SMOLLER B.R. and KRUSKALL M.S.: Phlebotomy for
diagnostic laboratory tests in adults: Pattern of use and
28.3% within the outcome and 10.6% within the effect on transfusion requirements. N. Engl. J. Med., 314:
total, with p=.453, this result agree with few studies 1233-5, 1986.
have examined iron supplementation in the critical
2- SHAPIRO M.J., GETTINGER A., CORWIN H.L., et al.:
care population. Anemia and blood transfusion in trauma patients admitted
to the intensive care unit. J. Trauma, 55: 269-74, 2003.
One retrospective analysis in surgery patients
identified 27 who received intravenous iron therapy, 3- CORWIN H.L., GETTINGER A., LEVY M.M., et al.:
The CRIT study: Anemia and blood transfusion in the
and found that compared with matched control
critically ill. Current clinical practice in the United States.
subjects; these patients did not have higher rates Crit. Care Med., 32 (1): 39-52, 2004.
of bacteremia [45].
4- BRANCO B.C., INABA K., DOUGHTY R., BROOKS
However intravenous iron supplementation may J., BARMPARAS G., et al.: The increasing burden of
phlebotomy in the development of anaemia and need for
have better efficacy than enteral administration blood transfusion amongst trauma patients. Injury, 43:
because of the block of intestinal absorption by 78-83, 2012.
hepcidin. It has been shown that iron may be useful
5- LYON A.W., CHIN A.C., SLOTSVE G.A. and LYON
in heart failure and pulmonary hypertension [46,47].
M.E.: Simulation of repetitive diagnostic blood loss and
onset of iatrogenicanemia in critical care patients with a
In this study we had 78 patients admitted for mathematical model. Comput. Biol. Med., 43: 84-90,
more than fourteen days in the ICU they represent 2013.
25% of our study group 39.7% of them died with
6- HEMING N., MONTRAVERS P. and LASOCKI S.: Iron
hemoglobind 7g/dl which was chosen as a best deficiency in critically ill patients: Highlighting the role
predictor to outcome. of hepcidin. Crit. Care, 15: 210, 2011.
This agreed with a large epidemiologic study, 7- WESTBROOK A., PETTILA V., NICHOL A., BAILEY
conducted in European ICUs, reported that lower M.J., SYRES G., et al.: Transfusion practice and guide-
lines in Australian and New Zealand intensive care units.
hemoglobin levels are associated with longer ICU Intensive Care Med., 36: 1138-46, 2010.
length of stay and increased in-hospital morta-
lity [17] . 8- VAN P.Y., RIHA G.M., CHO S.D., UNDERWOOD S.J.,
HAMILTON G.J., et al.: Blood volume analysis can
Conclusion: distinguish true anemia from hemodilution in critically
ill patients. J. Trauma, 70: 646-51, 2011.
A growing body of literature on anemia of
9- SINCLAIR A.M.: Erythropoiesis stimulating agents: A
critical illness points to four conclusions: pproaches to modulate activity. Biologics, 7: 161-74,
1- Anemia is highly prevalent in the critically ill. 2013.
2128 Anemia in Critically Ill Patients

10- NAIRZ M., SONNWEBER T., SCHROLL A., THEURL 26- KHAMIEES M., RAJU P., DeGIROLAMO A., AMOAT-
I. and WEISS G.: The pleiotropic effects of erythropoietin ENG-ADJEPONG Y. and MANTHOUS C.A.: Predictors
in infection and inflammation. Microbes Infect, 14: 238- of extubation outcome in patients who have successfully
46, 2012. completed a spontaneous breathing trial. Chest, 120 (4):
1262-70, 2001.
11- VINCHI F. and TOLOSANO E.: Therapeutic approaches
to limit hemolysis-driven endothelial dysfunction: Scav- 27- MARTINEZ F.J., FOSTER G., CURTIS J.L., CRINER
enging free heme to preserve vasculature homeostasis. G., WEINMANN G., FISHMAN A., DeCAMP M.M.,
Oxid Med. Cell Longev., 396527, 2013. BENDITT J., SCIURBA F., MAKE B., et al.: NETT
Research Group. Predictors of mortality in patients with
12- GONZALEZ-CASAS R., JONES E.A. and MORENO-
emphysema and severe airflow obstruction. Am. J. Respir.
OTERO R.: Spectrum of anemia associated with chronic
Crit. Care Med., 173: 1326-34, 2006.
liver disease. World J. Gastroenterol., 15: 4653-8, 2009.
13- MAIZES J.S., MURTUZA M. and KVETAN V.: Oxygen 28- CHAMBELLAN A., CHAILLEUX T. and SIMILOWSKI
transport and utilization. Respir. Care Clin. N. Am., 6: T.: Prognostic value of the hematocrit in patients with
473-500, 2000. severe COPD receiving long term oxygen therapy. Chest,
128: 1201-8, 2005.
14- Is a restrictive transfusion strategy safe for resuscitated
and critically ill trauma patients? J. Trauma, 57: 563-8, 29- NAIDECH A.M., JOVANOVIC B., WARTENBERG
2004. K.E., PARRA A., OSTAPKOVICH N., CONNOLLY
E.S., MAYER S.A. and COMMICHAU C.: Higher hemo-
15- Eliminating blood transfusions: New aspects and perspec- globin is associated with improved outcome after sub-
tives. Anesthesiology, 93: 242-55, 2000. arachnoid hemorrhage. Crit. Care Med., 35: 2383-9, 2007.
16- Should we demand fresh red blood cells for perioperative 30- DHAR R., ZAZULIA A.R., VIDEEN T.O., ZIPFEL G.J.,
and critically ill patients? Br. J. Anaesth, 9: 537-40, 2002. DERDEYN C.P. and DIRINGER M.N.: Red blood cell
17- VINCENT J.L., BARON J.F., REINHART K., et al.: transfusion increases cerebral oxygen delivery in anemic
Anemia and blood transfusion in critically ill patients. patients with subarachnoidhemorrhage. Stroke, 40: 3039-
JAMA, 288: 1499-507, 2002. Piagnerelli M., Boudjeltia 44, 2009.
K.Z., Brohee D., et al.: Alterations of red blood cell shape 31- PRABHAKARAN S., GUPTA R., OUYANG B., JOHN
and sialic acid membrane content in septic patients. Crit. S., TEMES R.E., MOHAMMAD Y., LEE V.H. and
Care Med., 31: 2156-62, 2003. BLECK T.P.: Acute brain infarcts after spontaneous
18- THOMAS J., JENSEN L., NAHIRNIAK S. and GIBNEY intracerebral hemorrhage: A diffusion-weighted imaging
R.T.: Anemia and blood transfusion practices in the study. Stroke, 41: 89-94, 2010.
critically ill: A prospective cohort review. Heart Lung, 32- ZAZULIA A.R., DIRINGER M.N., VIDEEN T.O., AD-
39: 217-25, 2010. AMS R.E., YUNDT K., AIYAGARI V., GRUBB R.L.
19- WALSH T.S., LEE R.J., MACIVER C., et al.: Anaemia and POWERS W.J.: Hypoperfusion without ischemia
during and at discharge from intensive care: The impact surrounding acute intracerebral hemorrhage. J. Cereb.
of restrictive blood transfusion practice. Intensive Care Blood Flow Metab., 21: 804-10, 2001.
Med., 32: 100-9, 2005.
33- ZAZULIA A.R., VIDEEN T.O. and POWERS W.J.: Tran-
20- NGUYEN B.V., BOTA D.P., MELOT C. and VINCENT sient focal increase in perihematomal glucose metabolism
J.L.: Red cell transfusion practice following the transfusion after acute human intracerebral hemorrhage. Stroke, 40:
requirements in critical care (TRICC) study: Prospective 1638-43, 2009.
observational cohort study in a large UK intensive care
34- HÉBERT P.C., WELLS G., BLAJCHMAN M.A., et al.:
unit. Vox Sang., 84: 211-18, 2003.
A multicenter, randomized, controlled clinical trial of
21- NGUYEN B.V., BOTA D.P., MELOT C. and VINCENT transfusion requirements in critical care. New England
J.L.: Time course of hemoglobin concentrations in non- Journal of Medicine, 340 (6): 409-17, 1999.
bleeding intensive care unit patients. Crit. Care Med., 31:
406-10, 2003. 35- JELKMANN W.: Proinflammatory cytokines lowering
erythropoietin production. Journal of Interferon and
22- SABATINE M.S., MORROW D.A., GIUGLIANO R.P., Cytokine Research, 18 (8): 555-9, 1998.
et al.: Association of hemoglobin levels with clinical
outcomes in acute coronary syndromes. Circulation, 111 36- RIVERS E., NGUYEN B., HAVSTAD S., et al.: Early
(16): 2042-9, 2005. goal-directed therapy in the treatment of severe sepsis
and septic shock. New England Journal of Medicine, 345
23- ARONSON D., SULEIMAN M., AGMON Y., et al.: (19): 1368-77, 2001.
Changes in haemoglobin levels during hospital course
and long-term outcome after acute myocardial infarction. 37- WALSH T.S., LEE R.J., MACIVER C., et al.: Br. J.
European Heart Journal, 28 (11): 1289-96, 2007. Anaesth. (2006) 97 (3): 278-91. doi: 10.1093/bja/ael189
First published online: July 27, 2006.
24- NEVINS M.L. and EPSTEIN S.K.: Predictors of outcome
for patients with COPD requiring invasive mechanical 38- WU W.C., RATHORE S.S., WANG Y., RADFORD M.J.
ventilation. Chest, 119 (6): 1840-9, 2001. and KRUMHOLZ H.M.: Blood transfusion in elderly
patients with acute myocardial infarction. N. Engl. J.
25- GOULD S., CIMINO M.J. and GERBER D.R.: Packed Med., 345: 1230-6, 2001.
red blood cell transfusion in the intensive care unit:
Limitations and consequences. American Journal of Crit- 39- JOLICOEUR E.M., O'NEILL W.W., HELLKAMP A.,
ical Care, 16 (1): 39-49, 2007. HAMM C.W., HOLMES D.R., AL-KHALIDI H.R., PA-
Naglaa M. Aly, et al. 2129

TEL M.R., VAN De WERF F.J., PIEPER K., ARM- transfusion facilitating difficult weaning from the venti-
STRONG P.W., et al.: APEX-AMI Investigators. Trans- lator. Anaesthesia, 53: 181-4, 1998.
fusion and mortality in patients with ST-segment elevation
myocardial infarction treated with primary percutaneous 44- HÉBERT P.C., BLAJCHMAN M.A., COOK D.J., YETI-
coronary intervention. Eur. Heart J., 30: 2575-83, 2009. SIR E., WELLS G., MARSHALL J. and SCHWEITZER
I.: Transfusion Requirements in Critical Care Investigators
40- YANG X., ALEXANDER K.P., CHEN A.Y., ROE M.T., for the Canadian Critical Care Trials Group. Do blood
BRINDIS R.G., RAO S.V., GIBLER W.B., OHMAN E.M. transfusions improve outcomes related to mechanical
and PETERSON E.D.: CRUSADE Investigators. The ventilation? Chest, 119: 1850-7, 2001.
implications of blood transfusions for patients with non-
ST-segment elevation acute coronary syndromes: Results 45- SWOBODA S.M. and LIPSETT P.A.: Intravenous iron
from the CRUSADE National Quality Improvement Ini- as a risk factor for bacteremia in the surgical intensive
tiative. J. Am. Coll. Cardiol., 46: 1490-5, 2005. care unit patient. Surg. Infect. (Larchmt), 6: 158, 2005.

41- RAO S.V., JOLLIS J.G., HARRINGTON R.A., GRANG- 46- ANKER S.D., COMIN COLET J., FILIPPATOS G.,
ER C.B., NEWBY L.K., ARMSTRONG P.W., MOLIT- WILLENHEIMER R., DICKSTEIN K., DREXLER H.,
ERNO D.J., LINDBLAD L., PIEPER K., TOPOL E.J., LÜSCHER T.F., BART B., BANASIAK W., NIEGOWS-
et al.: Relationship of blood transfusion and clinical KA J., et al.: FAIR-HF Trial Investigators. Ferric carbox-
outcomes in patients with acute coronary syndromes. ymaltose in patients with heart failure and iron deficiency.
JAMA, 292: 1555-62, 2004. N. Engl. J. Med., 361: 2436-48, 2009.

42- SCHÖNHOFER B., WENZEL M., GEIBEL M. and KÖH- 47- SMITH T.G., TALBOT N.P., PRIVAT C., RIVERA-CH
LER D.: Blood transfusion and lung function in chronically M., NICKOL A.H., RATCLIFFE P.J., DORRINGTON
anemic patients with severe chronic obstructive pulmonary K.L., LEÓN-VELARDE F. and ROBBINS P.A.: Effects
disease. Crit. Care Med., 26: 1824-8, 1998. of iron supplementation and depletion on hypoxic pulmo-
nary hypertension: Two randomized controlled trials.
43- SCHÖNHOFER B., BÖHRER H. and KÖHLER D.: Blood JAMA, 302: 1444-50, 2009.