Neutrophil left shift and white blood cell count as markers of bacterial
infection

Takayuki Honda, Takeshi Uehara, Go Matsumoto, Shinpei Arai, Mitsu-

toshi Sugano

PII: S0009-8981(16)30102-4
DOI: doi: 10.1016/j.cca.2016.03.017
Reference: CCA 14315

To appear in: Clinica Chimica Acta

Received date: 31 August 2015

Revised date: 18 March 2016
Accepted date: 22 March 2016

Please cite this article as: Honda Takayuki, Uehara Takeshi, Matsumoto Go, Arai Shin-
pei, Sugano Mitsutoshi, Neutrophil left shift and white blood cell count as markers of
bacterial infection, Clinica Chimica Acta (2016), doi: 10.1016/j.cca.2016.03.017

This is a PDF file of an unedited manuscript that has been accepted for publication.
As a service to our customers we are providing this early version of the manuscript.
The manuscript will undergo copyediting, typesetting, and review of the resulting proof
before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that
apply to the journal pertain.
 ACCEPTED MANUSCRIPT

Neutrophil left shift and white blood cell count as markers of bacterial infection

Takayuki Hondaa,b, Takeshi Uehara a,b, Go Matsumoto b,c, Shinpei Araic, Mitsutoshi Sugano c

T
IP
a
Department of Laboratory Medicine, Shinshu University School of Medicine, Asahi 3-1-1, Matsumoto

R
390-8621, Japan

SC
b
Department of Laboratory Medicine, Shinshu University Hospital, Asahi 3-1-1, Matsumoto 390-8621,
Japan

NU
c
Division of Infection Control, Shinshu University Hospital, Asahi 3-1-1, Matsumoto 390-8621, Japan
MA

Corresponding author:
D

Takayuki Honda, MD PhD

TE

Department of Laboratory Medicine, Shinshu University School of Medicine,

Asahi 3-1-1, Matsumoto 390-8621, Japan
P

Tel: +80-263-37-2802
CE

Fax: +80-263-34-5316
E-mail: thondat@shinshu-u.ac.jp
AC
 ACCEPTED MANUSCRIPT

Abstract
Neutrophil left shift and white blood cell (WBC) count are routine laboratory tests used to assess

T
neutrophil state, which depends on supply from the bone marrow and consumption in the tissues. If WBC

IP
count is constant, the presence of left shift indicates that neutrophil consumption is equal to production. A
decrease in WBC count indicates that neutrophil consumption surpasses supply. During a bacterial infection,

R
large numbers of neutrophils are consumed. Thus, from onset of infection to recovery, dynamic changes

SC
occur in WBC count and left shift data, reflecting the mild to serious condition of the bacterial infection.
Although various stimuli in healthy and pathological conditions also cause left shift, a change as sudden and

NU
significant is only seen in bacterial infection. Left shift does not occur in the extremely early or late phases of
infection; therefore, assessing data from a single time point is unsuitable for diagnosing a bacterial infection.
MA
We argue that time-series data of left shift and WBC count reflect real-time neutrophil consumption during
the course of a bacterial infection, allowing more accurate evaluation of patient condition.
D
TE

Key Words:
Left shift; bacterial infection; neutrophil; segmented neutrophil; band neutrophil; metamyelocyte; myelocyte
P
CE
AC
 ACCEPTED MANUSCRIPT

Contents:
1. Introduction

T
2. Neutrophils

IP
2-1. Neutrophil function
2-2. Neutrophil kinetics

R
3. Left shift

SC
3-1. Definition of band neutrophils
3-2. Definition of left shift
3-3. Amechanism causing left shift
4. Four phases of bacterial infection
NU
MA
5. Left shift and WBC count in patients
6. Bacterial infections without left shift
D

7. Left shift in other conditions

TE

8. Neutrophil count and left shift as routine laboratory tests

9. Arguments against left shift as a marker of bacterial infection
P

10. How to use left shift and WBC counts

CE
AC
 ACCEPTED MANUSCRIPT

1. Introduction
When a bacterial infection occurs, a large number of neutrophils are required to kill the

T
microorganisms at the bacterial infection site. As such, the neutrophils in the blood rapidly migrate to the

IP
infected site and, to compensate for neutrophil consumption, the bone marrow increases neutrophil
production [1-4]. Therefore, the change in the ratio of non-segmented to segmented neutrophils in the blood

R
is a much more accurate index for evaluating the severity of bacterial infection than the granulocyte

SC
concentration [5]. Left shift, an increase in myelocytes, metamyelocytes, and band neutrophils, was used as a
laboratory test representing acute bacterial infections in the 1960s and 1970s [6-9].In the 1980s and 1990s,

NU
automated differential leukocyte counts became available and many reports showed that manual band
counts added no useful information regarding bacterial infection [3, 10-14].
MA
Many studies have concluded that left shift is not useful for diagnosing bacterial infections
because it is not observed upon admission for all patients with bacterial infection.
D

Left shift does not appear in the extremely early phase of bacterial infections, but usually presents 12 to 24
TE

hours after onset. We have reported previously that a combination of white blood cell (WBC) count and left
shift analysis accurately reflects the real-time course of bacterial infection [15].
P

There are some exceptional bacterial infections (including infectious endocarditis[16], [17,
CE

18]meningitis, and abscess) in which left shift is not seen because the infection can be controlled by the
neutrophils in the circulating pool without increased production in the bone marrow.
AC

For many routine hematological and biochemical tests, analysis of time-series data indicates the
pathological condition of the organ or cell more accurately than data taken at a single time point. Moreover,
some routine tests must be used in combination in order to evaluate the clinical state of the patient. Therefore,
use of a single test at a single time point is insufficient for a diagnosis.
Another difficulty is that left shift has been observed in a number of instances where bacterial
infection was absent [3]. This can make diagnosis of bacterial infection difficult when left shift is assessed at
a single time point. However, time-series data can provide a more accurate diagnosis and can be used
subsequently to monitor the infection, because left shift follows a distinctive pattern over the course of a
bacterial infection [15].
 ACCEPTED MANUSCRIPT

In this review, we argue that the use of WBC count and left shift time-series data can be valuable
for diagnosis and monitoring of bacterial infections.

T
IP
2. Neutrophils
2-1. Neutrophil function

R
Neutrophils in the circulatory system are essential for an optimal host response to microorganisms.

SC
Neutrophils are rapidly recruited to bacterial infection sites where they evoke an immune response by
degrading and killing microbes [19]. Neutrophils bind and ingest microorganisms by phagocytosis, and the

NU
combination of neutrophil reactive oxygen species and granule components is highly effective in killing
most bacteria and fungi [20]. Therefore, a decrease in neutrophils or defects in their anti-microorganism
MA
activity dramatically increases the rate of morbidity and mortality in patients with a bacterial infection [19]
[21].Alarger number of neutrophils are consumed at the site of severe bacterial infections and they continue
D

to be supplied to the infected site from the bone marrow via the bloodstream. Therefore, the status of
TE

neutrophils in the blood may reflect the real-time condition of a patient with bacterial infection.
P

2-2. Neutrophil kinetics (Figure 1)

CE

In humans, neutrophils are produced from the multipotent stem cells in the bone marrow. Mature
neutrophils are released from the bone marrow into the blood and then they migrate into the tissues. Once in
AC

the tissues, neutrophils do not reenter the bloodstream [22] [23].

The production of a neutrophil lineage is achieved by the coordinated interaction of
hematopoietic stem cells, the bone marrow microenvironment, and various stimulatory and inhibitory
regulatory factors [24]. Neutrophils are delivered to the blood after proliferation, maturation, and storage [22].
It takes 3.4 days for myeloblasts to mature into myelocytes (after the promyelocyte stage) by mitosis, and 6.6
days for myelocytes to mature into segmented neutrophil (after the metamyelocyte and band neutrophil
stages). Therefore, more than 10 days are required for neutrophil maturation in the bone marrow [25].
In the blood, neutrophil equilibration occurs between the circulating pool and the marginal pool
(e.g., capillary endothelial cells in the lung, liver, spleen), and equilibration between these two pools is
 ACCEPTED MANUSCRIPT

sufficiently rapid and complete to allow them to be considered as one kinetic pool [2, 26-28]. Furthermore,
the neutrophils stay in these pools for only 6 to 8 hours and then migrate to the tissues [2, 25].

T
The WBC count changes in response to the neutrophil count. Because the neutrophil count

IP
changes dramatically in patients with a bacterial infection, we used the WBC count and the neutrophil count
interchangeably in this review. The real-time WBC count shows the status of neutrophil production and

R
consumption.

SC
3. Left shift
3-1. Definition of band neutrophils
NU
Band neutrophils, characterized by band or sausage-like nuclei, are immature neutrophils at the
MA
penultimate stage of maturation. Morphologically, it is sometimes difficult to differentiate band neutrophils
from mature (segmented) neutrophils. At least three definitions of band neutrophils exist in the literature,
D

established by the National Committee for Clinical Laboratory Standards[29], College ofAmerican
TE

Pathologists[30], and Committee on Clarification of the Nomenclature of Cells and Diseases of the blood
and Blood Forming Organs[31]; however, these definitions are inconsistent with one another, and band
P

reference ranges vary widely [3].

CE

Providing that the difference in band ratios according to each definition is within a few percentage
points, any one of the definitions can be used in the laboratory to differentiate band and segmented
AC

neutrophils. The absolute value of left shift is less important than the changes in left shift over time in order to
diagnose and monitor an infection.

3-2. Definition of left shift

Marsh et al. defined left shift as an increase in the ratio of non-segmented/segmented neutrophils
in the peripheral blood [5]. In other words, an increase in the amount of immature neutrophils (myelocytes,
metamyelocytes, and band neutrophils) in the peripheral blood. Therefore, the immature/total neutrophil
ratio, immature neutrophil/WBC ratio, band/total neutrophil ratio, band neutrophil/WBC ratio, absolute
immature neutrophil count, and band neutrophil count have been used as markers of left shift in previous
studies [5, 32-34]. In these studies, a band/total neutrophil ratio or WBC count above 6 to 20% was
 ACCEPTED MANUSCRIPT

commonly used to indicate left shift [15, 34-36] and the immature/total neutrophil ratio offered both practical
and theoretical advantages for evaluating the severity of neonatal sepsis [33]. On the other hand, in clinical

T
settings, the band neutrophil/WBC ratio was used as a marker of left shift for evaluating bacterial infection

IP
because it was more readily obtained in these settings [15].

R
3-3. Amechanism causing left shift (Figure２)

SC
In a patient with a bacterial infection, mature (segmented) neutrophils in the circulating and
marginal pools migrate to the infected site to kill the invading microorganisms. Adecrease of neutrophils in

NU
the blood stimulates the bone marrow to release immature neutrophils after mature ones that were stored in
the bone marrow pool are depleted.
MA
A decrease of mature and immature neutrophils stored in the bone marrow pool may also
stimulate the bone marrow to produce a neutrophil lineage. In addition, the bone marrow continues to
D

supply neutrophils to the infected site via the bloodstream until recovery from the bacterial infection. Alarger
TE

left shift indicates a larger consumption of neutrophils at the infected site and a larger production of
neutrophils in the bone marrow.
P
CE

4. Four phases of bacterial infection

The course of a bacterial infection can divided into 4 phases based on real-time left shift and
AC

WBC count data taken between onset and recovery. However, phase characterization can only be
established when a bacterial infection is a disease that consumes neutrophils in the blood [15].
The first phase occurs within 12 to 24 hours after onset of the bacterial infection. In this phase, the
WBC count drops under the normal reference range [5] and left shift has not yet appeared. Neutrophils from
the circulating and marginal pools migrate to the bacterial infection site; however, the bone marrow has not
yet started to increase neutrophil production and immature neutrophils have not been released into the blood
in response to the bacterial infection.
Neutrophil kinetics were rarely observed in the first phase. Because patients typically do not
consult a physician until the infection is well established, few reports describing left shift and WBC count in
this phase were available. In experimental studies using dogs, an increase in the neutrophil concentration and
 ACCEPTED MANUSCRIPT

the appearance of left shift occurred between 4 and 12 hours after bacterial inoculation [5], and it took more
than 4 hours for the bone marrow to respond to the decrease of neutrophils in the blood and to release more

T
neutrophils into the blood. In addition, our clinical study found that a neutrophil decrease without left shift

IP
occurred in the early stage of bacterial infections [15]. Another study found that the total, combined amount
of myelocytes, metamyelocytes, band neutrophil, and segmented neutrophils stored in the bone marrow

R
was 20 times more than those in the circulating pool, and the amount of segmented neutrophils in the bone

SC
marrow was 6.5 times greater than the amount in the circulating pool [2]. This shows that immature
neutrophils including myelocytes, metamyelocytes, and band neutrophils are released into the blood after

NU
depletion of the mature neutrophils that are stored in the bone marrow pool.
In the second phase, left shift increases from 15% to its highest ratio. In this phase, the WBC
MA
count will increase in the blood if neutrophil production surpasses consumption. This occurs when the bone
marrow accelerates neutrophil production. On the other hand, the WBC count in the blood, may also reveal
D

a balance between the neutrophil supply from the bone marrow and neutrophil consumption at the infected
TE

site. If the WBC count in the blood is above the reference range in this phase, the host is able to control the
bacterial infection. However, if the WBC count is under the reference range, the host cannot supply
P

sufficient neutrophils to cure the bacterial infection.

CE

In the third phase, the left shift values decrease from the highest ratio to 15% in accordance with
recovery from bacterial infection. The WBC count usually remains above the reference range. When the
AC

WBC count does not change in the blood, neutrophil production is equal to neutrophil consumption at the
infected site. Asmaller left shift indicates a decrease of neutrophil production in the bone marrow and shows
that a patient’s condition is improving.
In the fourth phase, the WBC count decreases to a level within the reference range and left shift is
once again under 15%. It is no longer necessary to maintain a high WBC count to supply neutrophils
effectively because they are no longer being consumed at the infection site.
Therefore, left shift values reflect the severity of the bacterial infection, and re-elevation of left shift
might help identify a deteriorating condition.

5. Left shift and WBC count in patients

 ACCEPTED MANUSCRIPT

When a bacterial infection consumes neutrophils in the blood, the real-time course of a bacterial
infection can be evaluated by a combination of WBC count (which can be used in the place of the absolute

T
neutrophil count) and the presence of left shift. However, both parameters should be interpreted based on a

IP
time-series because they change dramatically over the course of a bacterial infection [15]. In addition,
changes in these parameters might give more important information than those taken upon admission or at a

R
single time point.

SC
Figures 3, 4 and 5 were quoted from our previous report [15]. Figure 3 shows the percentage of band
neutrophil from the onset of a bacterial infection to recovery in six patients enrolled in our study [15]. The

NU
onset of various bacterial infections was determined by their symptoms, and patients were treated with
appropriate antibiotics, surgery, or both. The percentage of band neutrophil reflected the neutrophil
MA
consumption volume at the site of bacterial infection. Neutrophil consumption was equal to production in
the bone marrow if the WBC count in the blood did not change. WBC counts increased 10 to 20 hours after
D

the onset of infection in all patients, and decreased 48 to 120 hours later.
TE

Figure 4 shows the WBC count for the six patients from the onset of the bacterial infection to
recovery [15].AWBC count above the reference range indicated excessive neutrophils in the blood, and
P

that the host was able to supply enough neutrophils to the bacterial infection site. Conversely, a WBC count
CE

below the reference range indicated that the host could not supply the amount of neutrophils necessary for
the infection, and that the patient’s condition was poor. In some cases, WBC counts continued to increase
AC

after recovery from the bacterial infection. This may be because inflammatory reactions after extinction of
the bacteria might elevate WBC counts due to the presence of inflammatory cytokines.
Figure 5 shows the C-reactive protein (CRP) values from the onset of bacterial infection to
recovery [15]. Within 8 hours after onset of the bacterial infection, patients 2 and 4 had CRP values under 8.0
mg/dL and 3.0 mg/dL, respectively. CRP levels reached peak values 20 to 72 hours after the onset of
bacterial infection and then decreased. However, the CRP level remained above 5 mg/dL after recovery
from the bacterial infection (150 hours after onset) for five of six patients. CRP could not precisely reflect the
severity of the bacterial infection in real-time and only showed an inflammatory reaction regardless of the
presence or absence of a bacterial infection.
 ACCEPTED MANUSCRIPT

6. Bacterial infections without left shift

Some bacterial infections, including infective endocarditis, meningitis, and abscess, do not always

T
show left shift because neutrophils in the blood are not continuously consumed, and it is not necessary for

IP
the bone marrow to produce and release immature neutrophils.
For diagnosing infective endocarditis, a blood culture has been shown to be one of the most

R
accurate laboratory tests and the utility of other laboratory tests is limited. The WBC count, for example, is

SC
often in the normal range and leukocytosis does not occur frequently, limiting the usability of these tests to
diagnose [37]. An infected site of cardiac valve vegetation was a small lesion, and only a small amount of

NU
bacteria was released from the vegetation into the blood. Neutrophils in the circulating pool are sufficient to
kill them without additional neutrophils being supplied from the bone marrow. Therefore, an increase in the
MA
WBC count and the presence of left shift are not always found during the course of infective endocarditis.
For meningitis, no detailed descriptions about WBC counts or left shift have been provided in
D

textbooks, and routine blood work is often unrevealing. In meningitis cases, the WBC count is usually
TE

elevated with a shift toward immature neutrophil forms [38, 39]. However, in a previously reported case of
meningitis accompanied by leukocytosis without left shift [17], neither the WBC count nor left shift
P

occurrence were used to diagnose meningitis. In our case, the neutrophil count in the cerebrospinal fluid was
CE

not correlated with the neutrophil count, and left shift was not observed [18](Table 1). In healthy humans,
neutrophils in the bloodstream cannot easily cross the blood-cerebrospinal fluid barrier to migrate into the
AC

cerebrospinal fluid, and this barrier prevents neutrophil migration in patients with meningitis. Therefore, left
shift cannot be observed in all cases of meningitis because neutrophils in the blood are not constantly
consumed.
In cases with abscess (such as pyothorax and iliopsoas abscess), it is not always possible to
observe leukocytosis or left shift. Neutrophils in the blood may not migrate to the central portion of the
abscess because the thick fibrous wall and scanty vascular network prevent neutrophil supply to the infected
site. As with meningitis, consumption of neutrophils in the blood does not increase. Leukocytosis and left
shift do not always occur in patients with abscess.

7. Left shift in other conditions

 ACCEPTED MANUSCRIPT

Although rare, left shift has also been observed in conditions other than bacterial
infection. Several reports have detected left shift after the application of parenteral nutrition and

T
corticosteroids as well as after hemorrhage, anoxia, metabolic acidosis, and postoperative states

IP
[34, 40-42].
Andron et al. reported that a band count greater than 19% had a sensitivity of 49% and

R
a specificity of 87% for the diagnosis of bacterial infection [43]. Similarly, Seebach et al stated

SC
that a band count above 20% had a sensitivity of 54% and a specificity of 79%. In our study,
data from patients in the emergency care unit (n=72), including those with a bacterial infection

NU
(n=19), showed that a band count above 15% had a sensitivity of 47.4% and a specificity of
95.8% upon admission (data not published). We also observed left shift in one patient with a
MA
large hemorrhage. The hemorrhage may have decreased the amount of neutrophils in the blood
and stimulated the bone marrow to release immature neutrophils.
D

However, the conditions mentioned above are rare, and bacterial infection should be
TE

first suspected when left shift was observed.

P

8. Neutrophil count and left shift as routine laboratory tests

CE

Neutrophil count and left shift are not specific tests to diagnose bacterial infections, but are used
routinely to evaluate patient status more generally. These metrics reflect a dynamic state of neutrophils in the
AC

blood, influenced by supply from the bone marrow and consumption in various tissues.
Reference ranges for neutrophil count and left shift are commonly used to identify abnormal
neutrophil states [4, 44-46]. However, these ranges have proved unhelpful for differentiating bacterial
infections from other causes of neutrophil activity. In the past, threshold values of left shift (5–20% bands)
have been used to diagnose bacterial infections [47-49]. However, assessing neutrophil left shift at admission,
or at any single time point, has shown low sensitivity (but relatively high specificity) for diagnosing bacterial
infections[47-49].
Evaluation of left shift at a single time point, especially at patient admission, is unsuitable for
diagnosing bacterial infection because there are many potential causes of consumption and production of
neutrophils. Furthermore, patients may not present with left shift if they are admitted in the early stages of
 ACCEPTED MANUSCRIPT

bacterial infection. However, by evaluating left shift data twice within some days of admission, it would be
possible to diagnose bacterial infection with approximately 80% sensitivity, 90% specificity, and 90%

T
efficacy (data not published). In order to judge improvement or exacerbation of a bacterial infection,

IP
time-series data of left shift and WBC count are necessary. Time-series analyses of other routine laboratory
tests, including many hematological and biochemical tests, have proved much more useful for evaluating

R
patient states than analysis at a single time point[18].

SC
9. Arguments against left shift as a marker of bacterial infection

NU
Many reports in the 1980s and 1990s stated that manual band counts added no useful diagnostic
information to automated differential leukocyte counts [3, 10-14]. These studies argued that automated
MA
hematology analyzers offered automated, accurate, and precise differential counts for the five major
subclasses of leukocytes. The authors suggested that laboratory efficiency for diagnosing infections could be
D

improved by relying on instruments that perform accurate total leukocyte and absolute neutrophil counts and
TE

not on manual differential counts. However, in these reports, the statistical analyses were not appropriate
because only left shift upon admission was evaluated. In addition, all patients with a bacterial infection,
P

including those with diseases that do not show left shift, as well as those who were admitted to the hospital
CE

before the appearance of left shift were include in the analysis.

Left shift did not always appear during the course of a bacterial infection, and it was not observed
AC

in the extremely early phase [5, 15]. Within 12 to 24 hours after the onset of the bacterial infection, the bone
marrow has not yet produced neutrophils nor supplied immature neutrophils into the blood. In the late phase,
leukocytosis and high CRP values remained present and some patients were diagnosed with bacterial
infections even when neutrophils were not being supplied to the infected site. Therefore, it was impossible to
diagnose a bacterial infection using a one-time left shift measurement (especially a measurement taken upon
admission). Thus, left shift has not been used as a marker for evaluating bacterial infection because of poor
interpretations of its low sensitivity.
However, a few reports did recognize some usefulness of manual band counts for diagnosing
bacterial infections [15, 36, 49]. Band counts are also required to calculate the immature to total neutrophil
ratio, a widely used index for diagnosing neonatal sepsis [33].
 ACCEPTED MANUSCRIPT

10. How to use left shift and WBC counts

T
Various factors concern left shift and WBC counts; their combination is useful for evaluating the

IP
course of a bacterial infection. In bacterial infections, values for both metrics changed dramatically over the
course of the infection (as described in section 5). In addition, using a time-series analysis (with at least two

R
points) might increase the sensitivity and specificity of both tests for diagnosing and evaluating a bacterial

SC
infection.
Left shift has a high specificity for diagnosing bacterial infections because it depends primarily on

NU
the bone marrow response to the neutrophil decrease in the blood. When left shift is detected upon
admission, a bacterial infection or a complication from a bacterial infection should be suspected. If the left
MA
shift values increase and are accompanied by CRP elevation within a few hours, the patient can be correctly
diagnosed as having a bacterial infection. Assessing the presence of left shift in addition to the WBC count is
D

useful for evaluating the severity of the bacterial infection and for assessing whether antibiotic treatment is
TE

appropriate.
P
CE
AC
 ACCEPTED MANUSCRIPT

References
[1] Cartwright GE, Athens JW, Haab OP, Raab SO, Boggs DR, Wintrobe MM. Blood Granulocyte

T
Kinetics in Conditions Associated with Granulocytosis. Ann N YAcad Sci 1964; 113:963-967.

IP
[2] CARTWRIGHT GE,ATHENS JW, WINTROBE MM. THE KINETICS OF
GRANULOPOIESIS IN NORMAL MAN. Blood 1964; 24:780-803.

R
[3] Cornbleet PJ. Clinical utility of the band count. Clin Lab Med 2002; 22:101-136.

SC
[4] Longo D, Fauci A, Kasper D, Hauser S, Jameson J, Loscalzo J, Harrison's Principles of Internal
Medicine. New York: McGraw-Hill Health Professions Division, 2011.
[5]
NU
Marsh JC, Boggs DR, Cartwright GE, Wintrobe MM. Neutrophil kinetics in acute infection. J
Clin Invest 1967; 46:1943-1953.
MA
[6] Robertson OH, Fox JP. The relationship of infecting dosage, leucotytic response, bacteremia, and
extent of pulmonary involvement to the outcome of experimental labor pneumonia in the dog.. J Exp Med
D

1939; 69:229-246.
TE

[7] Akenzua GI, Hui YT, Milner R, ZipurskyA. Neutrophil and band counts in the diagnosis of
neonatal infections. Pediatrics 1974; 54:38-42.
P

[8] Hutter JJ, Hathaway WE, Wayne ER. Hematologic abnormalities in severe neonatal necrotizing
CE

enterocolitis. J Pediatr 1976; 88:1026-1031.

[9] Manroe BL, Rosenfeld CR, Weinberg AG, Browne R. The differential leukocyte count in the
AC

assessment and outcome of early-onset neonatal group B streptococcal disease. J Pediatr 1977; 91:632-637.
[10] Novak RW. The beleaguered band count. Clin Lab Med 1993; 13:895-903.
[11] Dutcher TF. Leukocyte differentials. Are they worth the effort? Clin Lab Med 1984; 4:71-87.
[12] Krause JR.Automated differentials in the hematology laboratory. Am J Clin Pathol 1990;
93:S11-16.
[13] Wenz B, Gennis P, Canova C, Burns ER. The clinical utility of the leukocyte differential in
emergency medicine. Am J Clin Pathol 1986; 86:298-303.
[14] Rimarenko S, Castella A, Salzberg MR, Strand CL. Evaluation of the automated leukocyte
differential count in emergency department patients. Am J Emerg Med 1987; 5:187-189.
 ACCEPTED MANUSCRIPT

[15] Ishimine N, Honda T, YoshizawaA, et al. Combination of white blood cell count and left shift
level real-timely reflects a course of bacterial infection. J Clin Lab Anal 2013; 27:407-411.

T
[16] Sexton D, J, Fowler B, G, Clinical manifestations and diagnosis of infective endocarditis. Wolters

IP
Kluwer Health, 2015:
http://www.uptodate.com/contents/clinical-manifestations-and-diagnosis-of-infective-endocarditis?source=s

R
earch_result&search=Infectious+endocarditis&selectedTitle=1~150.

SC
[17] Menaker J, Martin IB, Hirshon JM. Marked elevation of cerebrospinal fluid white blood cell
count: an unusual case of Streptococcus pneumoniae meningitis, differential diagnosis, and a brief review of

[18]
NU
current epidemiology and treatment recommendations. J Emerg Med 2005; 29:37-41.
Honda T, Sugano M. Atwenty years old man had a fever during hospitalization (in Japanese).
MA
Kensa to Gijutsu 2013; 41:294-300.
[19] Nauseef WM. How human neutrophils kill and degrade microbes: an integrated view. Immunol
D

Rev 2007; 219:88-102.

TE

[20] Kobayashi SD, Voyich JM, Burlak C, DeLeo FR. Neutrophils in the innate immune response.
Arch Immunol Ther Exp (Warsz) 2005; 53:505-517.
P

[21] Bodey GP, Buckley M, Sathe YS, Freireich EJ. Quantitative relationships between circulating
CE

leukocytes and infection in patients with acute leukemia. Ann Intern Med 1966; 64:328-340.
[22] Skubitz KM, Neutrophilic leukocytes. In: Greer JP, Arber DA, Glader B, et al., editors. Wintrobe's
AC

Clinical Pathology. Philadelphia: LIPPINCOTT WILLIAMS & WILKINS, 2013.

[23] Craddock CG, Adams WS, Perry S, Skoog WA, Lawrence JS. Studies of leukopoiesis: the
technique of leukopheresis and the response of myeloid tissue in normal and irradiated dogs. J Lab Clin
Med 1955; 45:881-905.
[24] Foucar K, Viswanatha DS, Wilson CA, Non-neoplastic Disorders of Bone Marrow. Washington,
DC: the American Resistry of Pathology, 2008.
[25] Deubelbeiss KA, Dancey JT, Harker LA, Finch CA. Neutrophil kinetics in the dog. J Clin Invest
1975; 55:833-839.
 ACCEPTED MANUSCRIPT

[26] Athens JW, Haab OP, Raab SO, et al. Leukokinetic studies. IV. The total blood, circulating and
marginal granulocyte pools and the granulocyte turnover rate in normal subjects. J Clin Invest 1961;

T
40:989-995.

IP
[27] Athens JW, Raab SO, Haab OP, et al. Leukokinetic studies. III. The distribution of granulocytes
in the blood of normal subjects. J Clin Invest 1961; 40:159-164.

R
[28] Mauer AM, Athens JW,Ashenbrucker H, Cartwright GE, Wintrobe MM. Leukokinetic studies.

SC
II. a method for labeling granulocytes in vitro with radioactive diisopropylefluorophosphate (DFP). J Clin
Invest 1960; 39:1481-1486.
[29]
NU
NCCLS, Referece leukocyte differential count (proportional) and evaluation of instrumental
methods; approved standard. In: NCCLS, editor. H20-A. Villanova: NCCLS, 1992.
MA
[30] Glassy E, Color Atlas of Hematology: An illustrated field guide based on proficiency testing.
North field: College of American Pathologists, 1998.
D

[31] organs Cftconocadobabf. First report. Am J Clin Pathol 1948; 6:454-457.

TE

[32] Kuppermann N, Walton EA. Immature neutrophils in the blood smears of young febrile children.
Arch Pediatr Adolesc Med 1999; 153:261-266.
P

[33] Christensen RD, Bradley PP, Rothstein G. The leukocyte left shift in clinical and experimental
CE

neonatal sepsis. J Pediatr 1981; 98:101-105.

[34] Seebach JD, Morant R, Rüegg R, Seifert B, Fehr J. The diagnostic value of the neutrophil left
AC

shift in predicting inflammatory and infectious disease. Am J Clin Pathol 1997; 107:582-591.
[35] Al-Gwaiz LA, Babay HH. The diagnostic value of absolute neutrophil count, band count and
morphologic changes of neutrophils in predicting bacterial infections. Med Princ Pract 2007; 16:344-347.
[36] Mathy KA, Koepke JA. The clinical usefulness of segmented vs. stab neutrophil criteria for
differential leukocyte counts.Am J Clin Pathol 1974; 61:947-958.
[37] Sexton D, Fowler V, Clinical manifestations and diagnosis of infective endocarditis. Alphen aan
den Rijn: Wolters Kluwer Health, 2015:
http://www.uptodate.com/contents/clinical-manifestations-and-diagnosis-of-infective-endocarditis?source=s
earch_result&search=%E6%84%89F%E86%89F%93%E86%80%A87%E85%BF%83%E85%86%8
5%E88%86%89C%E87%82%88E&selectedTitle=81~150.
 ACCEPTED MANUSCRIPT

[38] Dubos F, Korczowski B,Aygun DA, et al. Serum procalcitonin level and other biological
markers to distinguish between bacterial and aseptic meningitis in children: a European multicenter case

T
cohort study. Arch PediatrAdolesc Med 2008; 162:1157-1163.

IP
[39] Tunkel A, Clinical features and diagnosis of acute bacterial meningitis in adults. Alphen aan den
Rijn: Wolters Kluwer Health, 2015:

R
http://www.uptodate.com/contents/clinical-features-and-diagnosis-of-acute-bacterial-meningitis-in-adults?so

SC
urce=search_result&search=%E9%AB%84%E88%86%89C%E87%82%88E&selectedTitle=81~150.
[40] Hoyer JD. Leukocyte differential. Mayo Clin Proc 1993; 68:1027-1028.
[41]
6:411-419.
NU
Bull B, Korpman RA. Characterization of the WBC differential count. Blood Cells 1980;
MA
[42] Korpman RA, Bull B. Whither the WBC differential?--some alternatives. Blood Cells 1980;
6:421-429.
D

[43] Ardron MJ, Westengard JC, Dutcher TF. Band neutrophil counts are unnecessary for the
TE

diagnosis of infection in patients with normal total leukocyte counts. Am J Clin Pathol 1994; 102:646-649.
[44] Goldman L, Schafer AI, Cecil Textbook of Medicine. Philadelphia: WB Saunders, 2015.
P

[45] Dale D, Federman D, Scientific American Medicine. New York: Scientific American, 1997.
CE

[46] Cornbleet PJ, Novak RW. Classifying segmented and band neutrophils. CAP Today 1994;
8:37-41.
AC

[47] Wenz B, Ramirez MA, Burns ER. The H*1 hematology analyzer. Its performance
characteristics and value in the diagnosis of infectious disease. Arch Pathol Lab Med 1987; 111:521-524.
[48] Bates DW, Cook EF, Goldman L, Lee TH. Predicting bacteremia in hospitalized patients. A
prospectively validated model.Ann Intern Med 1990; 113:495-500.
[49] Wile MJ, Homer LD, Gaehler S, Phillips S, Millan J. Manual differential cell counts help predict
bacterial infection. Amultivariate analysis. Am J Clin Pathol 2001; 115:644-649.
 ACCEPTED MANUSCRIPT

Figure legends

T
Figure 1. Maturation and distribution of neutrophils

IP
It takes about 7~10 days for neutrophils to mature from stem cells in the bone marrow. Band and segmented
neutrophils are released from bone marrow and stay in the blood for several hours, and then migrate into the

R
organ tissues. The same amount of segmented neutrophils in the circulating pool is stored in the marginal

SC
pool and move easily between the circulating and marginal pools.
Abbreviations: Band cell, band neutrophil; Seg. cell, segmented neutrophil.

NU
Figure 2. Neutrophil migration and distribution to the bacterial infection site
MA
In bacterial infection cases, segmented neutrophils in the blood first migrate to the infected site, consuming
neutrophils. The resulting decrease of neutrophils in the blood then stimulates the bone marrow to release
D

mature neutrophils and, subsequently, immature neutrophils. When segmented and band neutrophils are not
TE

sufficient for controlling the bacterial infection, the bone marrow releases metamyelocytes and myelocytes
into the blood. Alarger left shift reflects a larger neutrophil consumption at the infected site.
P

Abbreviations: Band cell, band neutrophil; Seg. cell, segmented neutrophil.

CE

Figure 3. Ratios of band/white blood cell count (left shift) in six patients
AC

The data for six patients diagnosed with and successfully treated for a bacterial infection are shown. The
band/white blood cell count ratios were above 15% within 8 to 48 hours after the onset of bacterial infection,
and below 15% after recovery (150 hours after the onset). This figure was quoted from the reference of No.
16.

Figure 4. White blood cell count in six patients

Patients are the same as those in Figure 3. In the early phase of bacterial infection, WBC counts decreased to
fewer than 3,000 μL, and then, subsequently, increased. They continued to increase in three patients even
 ACCEPTED MANUSCRIPT

after recovery from the bacterial infection. Reference range (used by Nagano association of medical
technologists, Japan): 2,970–9,130/μL. This figure was quoted from the reference of No. 16.

T
IP
Figure 5. C-reactive protein
Patients are the same as Figure 3. C-reactive protein levels reached peak values 20 to 72 hours after the onset

R
of the bacterial infection, and remained above 5 mg/dL in all cases after recovery. Reference range (used by

SC
Nagano association of medical technologists, Japan): less than 0.1 mg/dL. This figure was quoted from the
reference of No. 16.

NU
MA
D
P TE
CE
AC
 ACCEPTED MANUSCRIPT

Table 1. Characteristics of a patient with meningitis

One patient presented with a high fever and was diagnosed with meningitis by a cerebrospinal fluid
examination on the 15th hospitalization day. Despite intravenous antibiotic therapy, high fever continued. The

T
IP
patient recovered after an intraspinal injection of antibiotics on the 23rd hospitalization day. Although he
suffered from meningitis between admission to the hospital and the antibiotic injection, left shift and WBC

R
elevation were not observed. The table from the reference of No. 36 was modified.

SC
1 day 7 12 15 18 22 25 32 38 Referece range

NU
Blood analysis

White bloo cell count 10.08 8.04 5.19 7.63 3.87 3.75 3.03 2.52 5.24 2.97-9.13x109/L
MA
Band cell 2 3 5 6 6 15 0-15 %

Segmented cell 73 83 46 67 49 55 69.5 28-68 %

D

Lymphocyte 9 10 27 13 33 23 17.9 17-57 %

TE

Monocyte 13 3 14 11 8 6 8.0 0-10 %

Eosinophil 3 1 4 1 2 0 4.4 0-10 %

P

Basophil 0 0 1 1 1 0 0.2 0-2 %

CE

Atyipical lymphocyte 0 0 0 0 0 1 0%

Metamyelocyte 1 0 0 0 0 0 0%
AC

Myelocyte 0 0 2 0 1 0 0%

Celebrospinal fluid

Total cell count 4270 9360 2500 18 5 1 <5/μL

Mononuclear cell 320 330 450 17 5 1 /μL

Segemnted cell 3950 9030 2050 1 0 0 /μL

Total protein 2863 282 93 55 49 36 10-40 mg/dL

Glucose 11 12 34 51 61 53 50-75 mg/dL

Band cell: Band

neutrophil, Segmented
 ACCEPTED MANUSCRIPT

cell: Segmented
neutrophil

T
IP
R
SC
NU
MA
D
P TE
CE
AC
 ACCEPTED MANUSCRIPT

T
IP
R
SC
NU
MA
D
TE

Fig. 1
P
CE
AC
 ACCEPTED MANUSCRIPT

T
IP
R
SC
NU
MA
D
TE

Fig. 2
P
CE
AC
 ACCEPTED MANUSCRIPT

T
IP
R
SC
NU
MA
D
TE

Fig. 3
P
CE
AC
 ACCEPTED MANUSCRIPT

T
IP
R
SC
NU
MA
D
TE

Fig. 4
P
CE
AC
 ACCEPTED MANUSCRIPT

T
IP
R
SC
NU
MA
D
TE

Fig. 5
P
CE
AC
 ACCEPTED MANUSCRIPT

Highlights

・Alarger left shift indicates a larger neutrophil consumption in the blood.

T
IP
・Alarger left shift represents a more severe bacterial infection.
・Left shift reflects the severity of a bacterial infection consuming neutrophils.

R
・High WBC count in the blood shows sufficient neutrophil supply to the infected site.

SC
・WBC (or neutrophil) count in the blood depends on neutrophil supply and consumption.

NU
MA
D
P TE
CE
AC