Full text of DOI:10.1007/PL00000229

© Birkhäuser Verlag, Basel, 2001
Inflamm. res. 50 (2001) 523–527
1023-3830/01/100523-05 $ 1.50+0.20/0
Inflammation Research
Differences in normal values for murine white blood cell counts
and other hematological parameters based on sampling site
J.A. Nemzek, G. L. Bolgos, B.A. Williams and D. G. Remick
Department of Pathology, University of Michigan, M2210 Medical Science I, 1301 Catherine Road, Ann Arbor, Michigan 48109-0602, USA,
Fax: ++1 734 763 6476, e-mail: remickd@umich.edu
Received 8 December 2000; returned for revision 27 June 2001; accepted by L.G. Letts 5 July 2001
Although the methods to quantify cell numbers are readily
available, the investigator must be aware that intrinsic vari-
ability may arise in cell counts in some species. It has been
reported that white blood cell counts and differentials can
vary substantially between strains, sexes, and ages of mice
[3], illustrating the importance of consistency during an
experiment. Of even greater significance, the white blood
cell counts from an individual mouse, under certain circum-
stances, can also vary greatly when the blood is obtained
from different sampling sites. It has been established that the
total WBC counts obtained from the heart blood of unanes-
thetized mice and of rats anesthetized with Nembutal are sig-
nificantly lower than those from tail [4] or retroorbital blood
[5]. However, these differences do not appear in rodents that
are anesthetized with ether [4, 6]. The effects of sample site
on cell counts in mice anesthetized with other anesthetic
agents such as ketamine and xylazine have not been evaluat-
ed.
As with cell counts, quantification of cell adhesion mole-
cules may be important to the evaluation of inflammatory
responses but also can be subject to variability. Several fac-
tors are known to modulate the presence of the b-integrins
CD11b and CD18 including cytokines, endotoxin [7], oxy-
gen radicals [8], purification techniques [9], and anesthetic
agents [10, 11]. However, it is unknown if the presence of
neutrophil adhesion molecules varies between sample sites,
as do the white blood cell counts.
Abstract. Objective and design: The effect of blood sam-
pling site on the hemogram and neutrophil adhesion mole-
cules was examined in BALB/c mice.
Materials and methods: Blood samples were drawn from the
tail, eye, and heart during anesthesia with ketamine and
xylazine. Cell numbers were quantified with an automated
counter and flow cytometry was used to quantify CD11b and
CD18.
Results: Total white blood cell (WBC) counts were highest
from tail, lower from eye, and significantly lower from heart
blood. In general, differences between tail and heart counts
reflected changes in all cell types. RBCs, platelets and hema-
tocrits were significantly increased in tail compared to heart
blood. Although CD18 levels were not different, CD11b was
significantly higher on neutrophils from tail compared to
heart blood.
Conclusions: In anesthetized BALB/c mice, sampling site
readily influences blood counts and neutrophil CD11b. The
findings underscore the need to standardize sampling site
when measuring these parameters.
Key words: WBC counts – Differentials – CD11b – RBC
count
Introduction
The classic studies examining the effects of sample sites
were performed a number of years ago on several different
strains of mice without the benefit of automated cell counters
or currently used anesthetic regimens. In order to appropri-
ately control our own in vivo murine studies, we sought to
further investigate the effects of blood sample site on fre-
quently measured parameters of inflammation. The first aim
of this study was to examine the effects of sampling site
(heart, tail, eye) and order on the hemogram of a commonly
used strain of mouse, the BALB/c, and to quantify these
effects with an automated system. The second aim was to
determine the effects of sampling site and order on the pres-
ence of the integrins CD11b and CD18 on neutrophils in
these mice. Finally, we would determine if the effects seen in
In studies of complex inflammatory responses, the evalua-
tion of multiple, interrelated parameters including peripheral
blood cell counts, WBC differentials, and the presence of cell
adhesion molecules has become an integral part of many in
vivo experiments. In the past, repeated evaluation of these
parameters in murine models was hampered by the limited
blood volume of the subjects and the labor involved in pro-
cessing large numbers of the samples. However, technical
advances including automated cell counters and flow cytom-
etry now allow rapid evaluation of complete blood counts [1,
2] and other factors from very small amounts of blood.
Correspondence to: D. G. Remick

524
J. A. Nemzek et al.
Inflamm. res.
ately performed by the core facility at the University of Michigan, Ann
Arbor, MI. The results were expressed as mean fluorescence intensity of
neutrophils.
these mice, anesthetized with ketamine and xylazine, would
be comparable to the historical data taken from mice
restrained by other means.
Statistics
Materials and methods
Summary statistics were expressed as the mean SEM. The differences
between sample sites were compared with an ANOVA. Tukey’s test for
pairwise comparisons was performed when indicated. Student’s t-test
was used to compare the order of sample collection from a particular
sampling site between different individuals.
Study design
With institutional animal care and use approval, blood samples were
drawn from anesthetized mice at three sampling sites, tail, eye, and
heart. In Group 1 (n = 6), the order of blood sampling sites used was tail,
eye, heart. In Group 2 (n = 6), the order of sampling was eye, tail, heart.
Because the heart blood collection was a terminal procedure and last in
the order of sampling in these two groups, a third group of mice (n = 3)
was used for sampling of heart blood only.
Results
Since the heart blood samples were obtained last in the series
in Groups 1 and 2, these samples were compared to values
from heart blood obtained without prior bleeding. There was
no significant difference based on the order of blood draw in
these samples. Therefore, the two groups were combined for
further comparison between sampling sites.
Animals
Female, BALB/c mice (Charles River Laboratories, Wilmington, MA)
weighing19–20 grams were used for sample collection. The animals
were kept under standard laboratory conditions with a 12:12 hour
dark/light cycle for at least one week prior to sample collection.
The cell counts obtained from the first blood sample
drawn in each group were compared (Fig. 1). Overall, the
Anesthesia
Animals were anesthetized with ketamine (83 mg/g) and xylazine
(13 mg/g) administered intraperitoneally for all of the blood drawing
procedures including tail blood samples.
Blood collection
Tail: The distal one-half centimeter of the tail was clipped and a capil-
lary pipette containing anticoagulants (EDTA for cell counting, heparin
for flow cytometry) were used to collect two 20 ml samples from the
bleeding surface. The first sample was used for cell counting and the
second for flow cytometry. Immediately after collection, the cut surface
of the tail was cauterized with styptic powder (Kwik-stop, ARC Labo-
ratories, Atlanta, GA) [12].
Eye: As previously described [12], a capillary pipette containing
anticoagulant (EDTA for cell counting, heparin for flow cytometry) was
inserted in the lateral canthus and blood collected from the retroorbital
sinus. A total of 40 ml of blood was removed form this site. After col-
lection, the pipette was removed and bleeding stopped when the eye
returned to a normal position.
Heart: With the animal positioned on its back, the abdomen was
opened and the heart visualized through the membranous part of the
diaphragm. A 21 gauge needle and 3cc syringe were used to obtain
0.5 ml of blood from the left ventricle. The blood was then immediate-
ly transferred to two 20 ml capillary pipettes containing the appropriate
anticoagulants for cell counting and flow cytometry.
Sample processing
Blood counts: EDTA anti-coagulated blood samples were used to obtain
a complete blood count with a Hemavet Mascot Multispecies Hematol-
ogy System Counter 1500R (CDC Technologies, Inc., Oxford, CT).
Samples were counted no longer than five minutes after blood was
drawn.
Flow cytometry: A 20 ml aliquot of heparinized blood was mixed
with phycoerythrin-labeled anti-CD11b and fluorescein isothiocyanate-
labeled anti-CD18 (Pharmingen, San Diego, CA). The samples were
incubated on ice for one half-hour. Flow cytometry was then immedi-
Fig. 1. Effect of Sample Site on WBC Counts. BALB/c mice were
anesthetized with ketamine and xylazine. Blood was drawn from either
the tail (n = 6), eye (n = 5), or heart (n = 15) and cells were counted with
an automated counter. Total WBC counts were highest in blood taken
from the tail and significantly lower in blood from the heart. The lower
counts reflected decreases in all of the types of WBCs. Cell counts from
the eye were, in general, intermediate when compared to the other sites.
Values are expressed as the mean SEM. * = p < 0.05 as compared to
the Tail group. † = p < 0.05 as compared to the Eye group.

Vol. 50, 2001
WBC variation by sample site
525
Table 1. Hematological data taken from unanesthetized, adult female
BALB/c Mice*.
Table 2. Hematological data from blood samples drawn from three
different sites.
Parameter
Mean SEM
Range
Parameter
Tail Blood
Eye Blood
637 56
Heart Blood
344 73*
Total WBCs (¥103/ul)
Neutrophils (¥103/ul)
Lymphocytes (¥103/ul)
Monocytes (¥103/ul)
Eosinophils (¥103/ul)
Platelets (¥103/ul)
Hematocrit (%)
17.7 1.034
5.7 0.4
9.1–28.7
1.9–11.5
6.7–15.7
0.3–1.4
0.05–0.51
668–1543
39.7–71.8
8.48–15.15
47.4–54.4
14.8–18.3
Platelets (¥103/ul) 882 95
Hematocrit (%)
RBC (¥106/ul)
MCV (fL)
53.15 2.27
10.94 0.54
48.85 1.94
46.12 1.90* 41.16 1.11*
10.5 0.5
0.7 0.06
0.2 0.03
934.7 34.5
51.2 1.0
10.2 0.2
50.7 0.4
16.9 0.2
10.14 0.42
45.5 0.62
14.28 0.59
9.00 0.22*†
45.5 0.33
13.04 0.34
Hemoglobin (g/dL) 15.5 1.15
* = p < 0.05 as compared to the tail blood sample; † = p < 0.05 as com-
pared to eye blood samples.
RBC (¥106/ul)
MCV (fL)
Hemoglobin (g/dL)
cence between the tail and heart blood samples. The amount
of CD18 did not differ between the sampling sites.
* Blood obtained from the tail and analyzed by Hemavet‚.
In addition to the WBC parameters, several other factors
were examined (Table 2). Platelet counts were significantly
higher in tail blood samples when compared to heart blood.
The hematocrit was significantly higher in the tail blood than
in the heart blood samples. This was probably due to a sig-
nificant decrease in RBC count in the heart blood samples
since the hematocrit presented by the automated counter is a
value derived from the total RBC count and the mean cor-
puscular volume (MCV). Hemoglobin values were not dif-
ferent between the groups.
WBC counts obtained from tail blood samples were higher
than those from either heart or eye and were within the ranges
previously obtained in our laboratory from 24, unanes-
thetized BALB/c mice (Table 1). The differences in the total
WBC counts between the tail samples and either the eye or
heart samples were significant. The lower total WBC counts
in the eye samples reflected significant changes in the num-
bers of neutrophils, monocytes, eosinophils, and basophils
compared to the tail samples. The lymphocyte counts were
not significantly different between the tail and eye samples.
The even lower WBC counts obtained from heart samples
represented significant decreases in lymphocyte numbers as
well as neutrophils and the other WBC types when compared
to the tail samples. Although all of the types of the WBCs
were generally lowest in the heart samples, the significant
difference in total WBC counts between the eye and heart
samples were primarily due to a difference in the lymphocyte
counts.
As with the WBC counts, the tail blood samples demon-
strated higher amounts of CD11b than did eye samples and
the heart samples demonstrated the least amount of CD11b
(Fig. 2). There was a significant difference in mean fluores-
Fig. 3. Effect of Sample Order on WBC Counts. BALB/c mice were
anesthetized with ketamine and xylazine. Blood was drawn from the tail
either before (First, n = 6) or after (Second, n = 5) a sample was taken
from the eye and cells were counted with an automated counter. Total
WBC counts were significantly higher in the blood taken before addi-
tional sampling. The change in total WBC count reflected changes in
neutrophil, monocyte, and eosinophil counts. Values are expressed as
the mean SEM. * = p < 0.05.
Fig. 2. Effects of Sample Site on CD11b. BALB/c mice were anes-
thetized with ketamine and xylazine. Blood was drawn from either the
tail (n = 6), eye (n = 6), or heart (n = 15) and cells were labeled with
pycoerythrin-labeled anti-CD11b for flow cytometry. Fluorescence was
significantly increased on neutrophils obtained from tail and eye blood
samples as compared to the heart blood samples. * = p < 0.05 as com-
pared to the Heart group.

526
J. A. Nemzek et al.
Inflamm. res.
Table 3. Hematological data from tail blood samples drawn either
before (first) or after (second) eye samples.
The results of this study demonstrated consistently that
cell counts are highest in tail blood, lower in eye blood and
lowest in the heart. These results are similar to the results of
other studies of blood obtained from rodents. Sakaki [5]
reported that RBC counts and WBC counts were significant-
ly higher in blood obtained from the tail of mice when com-
pared to that taken from the eye. Although the percentage of
neutrophils and lymphocytes found in the eye samples was
often higher than those in the tail samples, the absolute num-
bers of neutrophils and lymphocytes was not reported in that
study. Goldie [13] examined murine blood obtained from
several sites from the tail tip to the heart and found that WBC
counts progressively decreased from the periphery to the
heart. Likewise, Quimby [4, 14] showed dramatic decreases
in WBC counts in samples taken from the heart versus the tail
of rats. In one of those studies, no differences in WBC dif-
ferential counts in rats were seen when samples from the
heart and tail were compared [4]. In addition, that study did
not demonstrate an effect due to the order of sampling. The
results of differential counts and order of sampling in that
study contradicted the findings of our study but may be due
to variation in species or anesthetic protocols.
Parameter
First
Second
Platelets (¥103/ul)
Hematocrit (%)
RBC (¥106/ul)
MCV (fL)
882 95
563 69*
53.15 2.27
10.94 0.54
48.85 1.94
15.5 1.15
44.5 2.29*
9.65 0.05
46.15 0.43
12.84 0.59
Hemoglobin (g/dL)
* = p < 0.05.
While the reason for the differences seen in the WBC
counts may be rooted in the immunological demands of the
host at the various sites [14], the physiological explanation
for the differences has not been completely defined. It has
been suggested that the vascular resistance and blood stasis
created by the larger surface area in the capillary beds may be
responsible for the higher WBC counts from peripheral sites
[4]. This theory is supported by the fact that anesthesia with
ether equalized the counts in the tail and heart due to its
vasodilatory effects [4, 6]. However, the “damming up” that
is believed to occur in the capillary beds of the tail [4] was
not evident upon histological evaluation of tail sections tak-
en from mice in our study (data not shown). In addition, anti-
coagulation with heparin will also equalize the blood counts
without inducing vasomotor affects [13]. Alternatively, it has
been suggested that there is a fluid exchange between the
vessels and tissues of the tail in rodents [13].
Since CD11b also varied with sample site, we considered
the possibility that it was responsible for the changes in WBC
counts. The amount of CD11b was lowest in the heart blood
samples as were the total WBC counts, suggesting that the
neutrophils with the higher CD11b levels could have adhered
and essentially been filtered out by the lung capillary bed.
However, this would not completely explain the effects of sam-
ple site on the WBC count since all types of WBCs and RBCs
were lower in heart blood than in tail blood, suggesting a flow
mechanics mechanism. It is important to note that the differ-
ences in CD11b were detectable in the face of ketamine anes-
thesia, known to attenuate the upregulation of CD11b [11].
In conclusion, this study demonstrated that sampling site
can profoundly influence the total and differential white
blood cell, RBC, and platelet counts obtained from BALB/c
mice under ketamine and xylazine anesthesia. The order that
is used to draw samples from the tail can also influence these
parameters. Both sampling site and the order of tail blood
draw will also affect the amount of CD11b found on neu-
trophils. The findings emphasize the need to standardize
sampling site and order when measuring these parameters in
research studies.
Fig. 4. Effects of Sample Order on CD11b. BALB/c mice were anes-
thetized with ketamine and xylazine. Blood was drawn from the tail
either before (First, n = 6) or after (Second, n = 6) a sample was taken
from the eye and cells were labeled with phycoerythrin-labeled anti-
CD11b for flow cytometry. Fluorescence was significantly higher in the
samples taken prior to additional sampling. * = p< 0.05. All values are
expressed as the mean SEM.
The order of blood draw (either first or second) from the
tail and the eye was also evaluated. There were no significant
differences in any of the cell counts or the mean fluorescence
of the neutrophil adhesion molecules when the samples were
taken from the eye first (Group 2) or taken from the eye after
the tail sample (Group 1). In contrast, the tail samples taken
first (Group 1) demonstrated a significantly higher WBC
count when compared to tail samples drawn after eye sam-
ples (Group 2) (Fig. 3). These counts reflected significant
differences in neutrophil, monocyte, and eosinophil counts.
The platelet counts and hematocrit of samples taken from the
tail were higher when the sample was drawn before the eye
sample than when drawn after the eye sample (Table 3). In
addition, the mean fluorescence of CD11b was significantly
higher in the tail blood samples drawn first (Fig. 4).
Discussion
The published studies examining the effects of sample site on
cell counts were conducted decades ago on several different
strains of mice under various anesthetic protocols and pro-
duced variable results. Our study updates this information by
using an automated cell counter to obtain information from a
commonly used strain of mouse subjected to a frequently
used anesthetic protocol.

Vol. 50, 2001
WBC variation by sample site
527
[8] Fraticelli A, Serrano CV, Jr, Bochner BS, Capogrossi MC, Zweier
JL. Hydrogen peroxide and superoxide modulate leukocyte adhe-
sion molecule expression and leukocyte endothelial adhesion.
Biochim Biophys Acta 1996; 1310: 251–9.
[9] KuijpersTW, ToolAT, van der Schoot CE, et al. Membrane surface
antigen expression on neutrophils: a reappraisal of the use of sur-
face markers for neutrophil activation. Blood 1991; 78: 1105–11.
[10] Schmidt H, Ebeling D, Bauer H, et al. Ketamine attenuates endo-
toxin-induced leukocyte adherence in rat mesenteric venules. Crit
Care Med 1995; 23: 2008–14.
References
[1] Ebong SJ, Call DR, Bolgos G, Newcomb DE, Granger JL, O’Reil-
ly M, et al. Immunopathologic responses to non-lethal sepsis.
Shock 1999; 12: 118–26.
[2] Nemzek JA, Call DR, Ebong SJ, Newcomb DE, Bolgos GL,
Remick DG. Immunopathology of a two-hit murine model of acid
aspiration lung injury. Am J Physiol; Lung Cell Mol Physiol 2000;
278: L512–L520.
[3] Bannerman RM. Hematology. In: Foster HL, Small JD, Fox JG,
editors. The Mouse in Biomedical Research. vol. 3. New York,
N.Y.: Academic Press, 1983: 293–309.
[4] Quimby FH, Goff LG. Effect of source of blood sample on total
white cell count of the rat. Am J Physiol 1952; 170: 196–200.
[5] Sakaki K, Tanaka K, Hirasawa K. Hematological comparison of
the mouse blood taken from the eye and the tail. ExperAnim 1961;
10: 14–9.
[11] Weigand MA, Schmidt H, Zhao Q, Plaschke K, Martin E, Bar-
denheuer HJ. Ketamine modulates the stimulated adhesion mole-
cule expression on human neutrophils in vitro. Anesth Analg
2000; 90: 206–12.
[12] Cuncliffe-Beamer TL. Biomethodology and surgical techniques.
In: Foster HL, Small JD, Fox JG, editors. The Mouse in Biomed-
ical Research. vol. 3. New York, N.Y.: Academic Press, 1983:
402–37.
[6] Nichols J, Miller AT. A comparison of the total leucocyte count in
the heart blood and peripheral blood of the rat. Science 1948; 108:
378–9.
[13] Goldie H, Jones AM, Ryan H, Simpson M. Leucocyte counts in
the blood from the tail and heart of the mouse. Science 1954; 119:
353–4.
[7] McMillen MA, Huribal M, Sumpio B. Common pathway of
endothelial-leukocyte interaction in shock, ischemia, and reperfu-
sion. Am J Surg 1993; 166: 557–62.
[14] Quimby FH, Saxon PA, Goff LG. Total white blood cell counts of
peripheral and heart blood samples of the rat. Science 1948; 107:
447.