Accepted Manuscript

Improved Early Detection of Sepsis in the Emergency Department with a Novel

Monocyte Distribution Width Biomarker

Elliott D. Crouser, M.D., Joseph E. Parrillo, M.D., Christopher Seymour, M.D., M.D,
M.P.H. Derek C. Angus, Keri Bicking, Pharm. D., Liliana Tejidor, Ph.D., Robert
Magari, Ph.D., Diana Careaga, JoAnna Williams, M.D., Douglas R. Closser, M.D.,
Michael Samoszuk, M.D., Luke Herren, Emily Robart, Fernando Chaves, M.D.

PII: S0012-3692(17)31072-3
DOI: 10.1016/j.chest.2017.05.039
Reference: CHEST 1153

To appear in: CHEST

Received Date: 1 March 2017

Revised Date: 1 May 2017
Accepted Date: 31 May 2017

Please cite this article as: Crouser ED, Parrillo JE, Seymour C, Angus . DC, Bicking K, Tejidor L, Magari
R, Careaga D, Williams J, Closser DR, Samoszuk M, Herren L, Robart E, Chaves F, Improved Early
Detection of Sepsis in the Emergency Department with a Novel Monocyte Distribution Width Biomarker,
CHEST (2017), doi: 10.1016/j.chest.2017.05.039.

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

Improved Early Detection of Sepsis in the Emergency Department with a Novel

Monocyte Distribution Width Biomarker

Elliott D. Crouser, M.D.1, Joseph E. Parrillo, M.D.2, Christopher Seymour, M.D. 3, Derek
C. Angus, M.D, M.P.H..3, Keri Bicking, Pharm. D.2, Liliana Tejidor, Ph.D.4, Robert
Magari, Ph.D. 4, Diana Careaga4, JoAnna Williams, M.D.5, Douglas R. Closser, M.D. 1,
Michael Samoszuk, M.D. 4, Luke Herren1, Emily Robart1, Fernando Chaves, M.D. 4

PT
1. Department of Medicine, The Ohio State University Wexner Medical Center,
Columbus, OH
2. Heart and Vascular Hospital, Hackensack University Medical Center, Hackensack,

RI
NJ
3. Department of Critical Care Medicine, University of Pittsburgh School of Medicine
Pittsburgh, PA

SC
4. Beckman Coulter, Inc.
5. Department of Pathology, The Ohio State University Wexner Medical Center,
Columbus, OH

U
Address for Reprints (reprints will be ordered):

Elliott D. Crouser, M.D.

AN
Professor of Medicine
The Ohio State University Wexner Medical Center
Room 201F, Davis Heart & Lung Research Institute
473 West 12th Avenue
M

Columbus, OH 43210
E-mail: elliott.crouser@osumc.edu
D

Financial Support for the Project:

Beckman Coulter, Inc.
TE

Word Count, Abstract/Manuscript Body: 250/3126

Key Words: Sepsis, biomarker, blood, cell volume, monocyte, emergency department
EP

Summary of Conflicts of Interest:

All study participants were supported financially through Beckman Coulter, Inc. for their
efforts on the study. Otherwise, there are no other conflicts of interest to declare relating
C

to this study.
AC
 ACCEPTED MANUSCRIPT

Abstract:

Background: Sepsis most often presents to the emergency department (ED), and

delayed detection is harmful. The white blood count (WBC) is often used to detect

sepsis, but changes in WBC size also correspond with sepsis. We sought to determine

PT
if volume increases of circulating immune cells adds value to the WBC for early sepsis

detection in the ED.

RI
Methods: A blinded, prospective cohort study was conducted in two different ED

SC
populations within a large academic hospital.

Results: Neutrophil and monocyte volume parameters were measured in conjunction

U
with routine CBC testing on a UniCel® DxH 800 analyzer at the time of ED admission

and were evaluated for the detection of sepsis. 1320 ED subjects were consecutively
AN
enrolled and categorized as controls (n=879), systemic inflammatory response

syndrome (SIRS) (n=203), infection (n=140), or sepsis (n=98). Compared to other

M

parameters, monocyte distribution width (MDW) best discriminated sepsis from all other
D

conditions [AUC 0.79 (95% CI: 0.73-0.84)]; sensitivity 0.77, specificity 0.73; MDW
TE

threshold of 20.50.], sepsis from SIRS [AUC 0.74 (95% CI: 0.67-0.84)] and severe

sepsis from non-infected ED patients [AUC 0.88 (95% CI: 0.75-0.99); NPV 99%]. The
EP

added value of MDW to WBC was statistically significant (AUC 0.89 for MDW + WBC vs

0.81 for WBC alone; p < 0.01); and a decision curve analysis also showed improved
C

performance compared to WBC alone.

AC

Conclusions: The incorporation of MDW with WBC is shown in this prospective cohort

study to improve detection of sepsis compared to WBC alone at the time of admission in

the ED.

Clinical Trial Registration: ClinicalTrials.gov (NCT02232750)

 ACCEPTED MANUSCRIPT

Introduction:

Sepsis contributes significantly to hospital resource utilization and patient

mortality (1-3). Approximately 70% of sepsis cases are admitted through the ED (4), and

delays in the identification and treatment of these patients contributes significantly to

PT
adverse outcomes (5-7).

Whereas the definition of sepsis was recently revised (Sepsis-3) based upon

RI
systemic manifestations that portend adverse outcomes (8), a majority of hospital deaths

SC
due to sepsis occur in those presenting with milder sepsis, as reflected by Sepsis-2

criteria (3,5). A standard approach to identifying sepsis includes a clinical suspicion of

U
infection coupled with systemic inflammatory response syndrome (SIRS) criteria, and

this approach is shown to perform well relative to Sepsis-3 criteria for predicting hospital
AN
mortality or ICU transfer (9). Thus, the criteria used for the older Sepsis-2 definition

performs well in the ED population for predicting adverse sepsis outcomes, and
M

presumably identifies those who would most benefit from early treatment.
D

In response to the new Sepsis-3 definition the “Surviving Sepsis Campaign”

TE

recommends that the identification and treatment of sepsis should be unchanged, with

the first essential step being to identify “suspected infection” (10). The early detection
EP

and treatment of sepsis remains a major clinical challenge. The detection of infection is

typically delayed by >4 hrs after admission to the hospital (11) and by >8 hrs in ~30% of
C

cases presenting to the ED (12). Unlike other sepsis biomarkers, such as procalcitonin
AC

(PCT) or C-reactive protein (CRP), the WBC is among the first laboratory tests available

to clinicians in the ED. As such, PCT and CRP are not typically used for early detection.

Unfortunately, WBC elevation is non-specific for sepsis in the ED patient population (13).

Thus, there is a pressing need for a biomarker that is routinely available during the initial

ED clinical evaluation to enhance the detection of sepsis.

 ACCEPTED MANUSCRIPT

Circulating neutrophils and monocytes are among the first to respond to

pathogenic organisms (14). Recent studies suggest that immune cell volume increases

may be useful for the detection of sepsis (14-21). These recent studies, while

encouraging, are limited by their small sample sizes and did not include ED patients.

PT
Thus, we sought to determine the utility of measuring volumetric changes of circulating

immune cells as predictors of sepsis in a large prospective study of adult ED patients.

RI
We hypothesized that volume increases in circulating neutrophils and/or monocytes,

SC
alone or in combination with other established CBC parameters, such as WBC, would

improve sepsis detection in the ED.

U
Methods:
AN
Patient enrollment:

The study, registered with ClinicalTrials.gov (NCT02232750) and approved by

M

the Western Institutional Review Board, Inc. (Protocol #20141542), was a blinded,

prospective cohort study enrolling patients presenting to two different EDs at the
D

OSUWMC. The study enrolled adults (age ⩾18 to 90 years of age) whose initial
TE

evaluation included a complete blood count (CBC) within 24 hours of admission to the

ED, and no subject was enrolled more than once. Based on available literature (4,22)
EP

we estimated that approximately 1350 ED study subjects with concurrent CBC testing

would be required to achieve the lower limits of 65% for the target sensitivity estimate of
C

75% using a two sided 95% confidence interval. These calculations were based on the
AC

approach recommended by Burderer (23). Given the large volume of admissions to the

two EDs, we enrolled patients using a weekday cap of 25 consecutive patients per day.

White cell volume determination:

 ACCEPTED MANUSCRIPT

All blood samples were analyzed on a UniCel® DxH 800 analyzer (Beckman

Coulter, Inc.) with version 2.0 software within 8 hours of collection. This instrument

measures specific cell volume parameters, and the distribution of cell volumes within a

group of cells (Figure 1) 1. The clinical research team was blinded to the these results at

PT
the time of clinical data entry and during assignment of the patients to a clinical category.

RI
Clinical Classification of ED Patients:

SC
Study subjects were categorized as follows based upon the “Sepsis-2”

consensus criteria (24,25): non-SIRS (i.e., zero or one *SIRS criterion) and no infection;

U
SIRS (systemic inflammatory response syndrome: ≥ 2 SIRS criteria); Sepsis (infection

plus SIRS); Severe Sepsis (sepsis with one or more organ failure); Septic Shock (sepsis
AN
with refractory hypotension); Infection, but no sepsis (i.e., zero or one SIRS criterion).

The presence of infection was determined using all available clinical data available 7
M

days after ED admission, including results from cultures, molecular tests (e.g.,
D

polymerase chain reaction, antigens), relevant imaging, and tissue pathology. (*SIRS
TE

criteria are as follows: WBC > 12,000 or <4,000 or >10% bands; pulse >90; respiratory

rate > 20; Temperature <96.8°F or >100.4°F.)

EP

Sepsis criteria had to be fulfilled within 6 hours of the CBC, and categorization

(see above) was verified by expert review of the electronic medical record >7 days after
C

ED admission by investigators EDC and DRC. Discordances were adjudicated, and a

AC

final determination was made by EDC prior to data analysis.

Preexisting Conditions:

1Cell volume parameters: [mean neutrophil volume (MNV), neutrophil distribution width (NDW), mean
monocyte volume (MMV), monocyte distribution width (MDW)] are research use parameters on the
DxH80, their clinical utility has not been established. These parameters are not yet approved for use as
sepsis biomarkers.
 ACCEPTED MANUSCRIPT

We anticipated that certain clinical conditions could confound the results of the

study. For example, vitamin deficiencies (e.g., B12, folate), primary bone marrow

disorders and certain drugs are common causes of generalized macrocytosis (26).

Furthermore, immune suppression could suppress immune cell activation and

PT
associated changes in morphology. As such, we performed independent analyses of the

data wherein we considered pre-existing conditions known to cause macrocytosis or to

RI
be associated with impaired immune cell activation (see Table S1) to determine if these

SC
pre-existing conditions independently influenced the study results.

U
Statistical Approach

General descriptive statistics and box-plots were calculated for CPD parameters.
AN
Diagnostic ability was evaluated in terms of the area under the curve (AUC), sensitivity,

specificity, positive predictive value (PPV), and negative predictive value (NPV) along
M

with their 95% confidence intervals (CI). The score approach was used to calculate CI
D

for sensitivity, specificity, PPV and NPV. The initial cut-off values were obtained using
TE

the Youden Index approach, which were optimized iteratively to maximize the sensitivity

and specificity.
EP

Three approaches were used to demonstrate the added value of MDW in

comparison to WBC: 1) Differences in areas under the curve (AUC): AUC was
C

calculated using a one-predictor variable logistic model with WBC as the predictor and
AC

sepsis status as the response. In addition, AUC was calculated for the two-predictor

variables logistic model with both WBC and MDW as predictors and sepsis as response.

The comparison between the AUC from the two models (WBC vs. WBC+MDW) along

with their confidence intervals were calculated as described by De Long et al (27); 2)

Cochran–Mantel–Haenszel (CMH) approach: Both WBC and MDW were dichotomized

to 0 and 1 based on their values falling into the ‘normal’ or ‘abnormal’ category. WBC
 ACCEPTED MANUSCRIPT

was considered to be normal if the recovered value was between 3.6 and 12.0. MDW

was considered normal when its recovered value was below 20.5. CMH statistics and

the common odds ratio for MDW adjusted for WBC effects were calculated as previously

described (28). MDW provided added value if the estimated odds ratio was statistically

PT
different from one (i.e., the 95% confidence interval did not include 1); 3) Decision curve

analysis: The number of true positives (TP) and false positive (FP) were calculated from

RI
the two logistic models (WBC alone and WBC+MDW). The net benefit for a given

SC
probability (p, 0 ≤ p ≤ 1) was calculated for each model as,

TP FP p

U
Net benefit = −  
N N 1−p
AN
Where, N was the total number of subjects in the trial, and p the threshold probability.
M

Decision curves (net benefit vs. threshold probability) were plotted for each model and

compared to each other (29,30). SAS 9.3 (SAS Institute, Cary, NC) statistical program
D

was used for data analyses.

TE

Results:
EP

Patient Demographics:
C

From September 2014 through December 2014, 1320 patients were enrolled, the
AC

demographic features of whom are shown in Table S2. ~30% of the sepsis cases were

admitted through a community hospital ED in eastern Columbus, OH; and ~70% were

admitted to the ED of a larger tertiary hospital located near the center of the city. The

distribution of controls, SIRS, infection, sepsis, severe sepsis and septic shock are

provided in Table 1. The prevalence of sepsis in this ED population was 7.4%, with
 ACCEPTED MANUSCRIPT

severe sepsis or septic shock accounting for 18.8% of the septic patients (1.4% of the

total ED population). In keeping with other sepsis studies (31), more than half of the

septic patients (61%) had positive cultures (blood, urine, other body fluids), the rest

being diagnosed based upon clinical criteria (e.g., surgical tissue analysis), serological

PT
or polymerase chain reaction (PCR) testing. Among those with positive cultures,

diagnostic serologies or PCR results, the causes of sepsis were categorized as follows:

RI
79% bacterial, 14% viral, and 7% fungal.

SC
Detection of Sepsis based upon Cell Volume criteria:

U
Four independent variables, as well as routine CBC parameters, were analyzed

for their ability to identify patients with sepsis (all sepsis phenotypes were combined for
AN
this analysis: sepsis, severe sepsis, septic shock) in the ED population: mean neutrophil

volume (MNV); neutrophil distribution width (NDW, a measure of size variation within the
M

cell population); mean monocyte volume (MMV); and monocyte distribution width (MDW)
D

(Figure 1 and Table S3). Neutrophil parameters MNV and NDW were significantly higher
TE

in the sepsis group compared to all others, as was the case for monocyte parameters

MMV and MDW (Figure 2). Although WBC and the WBC neutrophil percentage (NE)
EP

were higher in the sepsis group, these parameters were also elevated in the SIRS group

(Figure S1). After establishing cut-off values best discriminating sepsis from the other
C

conditions for each parameter (see Table S4), we compared the discrimination of each
AC

for sepsis. Overall, MDW best discriminated sepsis from all other conditions based on

AUC (Table S5). The negative predictive value of a normal MDW was 97%.

The performance of MDW was slightly better when subjects with preexisting

conditions associated with macrocytosis or compromised immune cell function,

comprising ~17% of the total study population, were excluded from the analysis (Table

S6).
 ACCEPTED MANUSCRIPT

Correlation of MDW with infection severity:

As shown in Figure S2, MDW increases incrementally with progression from

milder forms of infection to sepsis, and is highest in those with more organ dysfunction in

sepsis.

PT
MDW Added Value Analysis for Sepsis Detection:

RI
The estimated AUC for WBC was 0.81 (95% CI: 0.75-0.86) and for MDW + WBC

SC
was 0.89 (95% CI: 0.84-0.92). The difference between the AUCs of the two models was

0.08 (95% CI: 0.02-0.13). CMH Odds Ratio was 6.9 ((95% CI: 4.2-11.4). The benefit of

U
MDW is reflected when difference between AUCs is significantly >zero and CMH odds

ratio is significantly >1. Inclusion of MDW with WBC provided an 8% improvement on the
AN
AUC (Figure 3A), with a P-value of 0.0035, and by CMH odds ratio (95% confidence

intervals are >1). Since both criteria were satisfied, the MDW contribution to improved
M

sepsis detection is statistically significant. Although the same assumptions cannot be

D

made for comparisons of AUC of sepsis and SIRS, MDW is shown to have higher AUC
TE

alone, and in combination with WBC, compared to WBC alone (Figure 3B). As shown in

Figure 4, the MDW improves sepsis detection relative to WBC alone across a wide
EP

range of decision probability thresholds.

Discussion:
C

Volumetric increases are an early manifestation of immune cell responses to

AC

severe infections, and as such have shown potential as sepsis biomarkers in humans

(14-21). We show that volumetric changes of circulating immune cells, particularly

increased monocyte distribution width (MDW), correspond with previously established

criteria for sepsis (24) in a large ED patient population. MDW distinguished sepsis from

SIRS and the magnitude of MDW elevation correlated with infection severity and organ
 ACCEPTED MANUSCRIPT

dysfunction, ranging from normal MDW in those with limited infections, increasing

incrementally with sepsis severity. Moreover, the MDW provides added value relative to

WBC alone for early detection of sepsis in the ED. To the extent that earlier detection of

sepsis leads to earlier treatment, these findings have implications for reducing sepsis

PT
mortality (5-7).

Consistent with the results of recent human studies (14-21) and the established

RI
function of neutrophils as “first responders” during the innate immune response to

SC
invading organisms, both mean neutrophil volume (MNV) and neutrophil distribution

width (NDW) were reasonably sensitive and specific for the detection of sepsis (Table

U
S5). Based upon previous studies demonstrating that changes in neutrophil volume and

cell activation markers (CD64) can detect sepsis in hospitalized pediatric patients
AN
(14,17,18,32), we originally expected that neutrophil volume changes would be superior

for the detection of sepsis in the ED population. However, MDW outperformed MNV,
M

NDW and all other volumetric metrics [MMV, MCV, mean platelet volume (data not
D

shown)] for the detection of sepsis, particularly in the diagnostically challenging

TE

subgroup of ED patients presenting with SIRS (Figure 3B). We posit that monocyte

parameters may perform better because circulating monocytes are first responders to
EP

infections (33,34) and the response is proportional to the intensity of the exposure to

either bacterial, fungal, or viral pathogens (35), resulting in an acute increase in

C

monocyte size (36,37). In contrast, the neutrophil response to certain infections [e.g.,
AC

viral (38,39)] is relatively delayed.

The ED population is most appropriate for sepsis screening given that a majority

of sepsis cases cared for in the hospital setting meet sepsis criteria at the time of

admission (4), and early identification of sepsis is critical in terms of clinical outcomes (5-

7). By recent estimates >500,000 cases of severe sepsis, or about two thirds of all

severe sepsis cases, present to the ED each year in the U.S. (1,4) This number would
 ACCEPTED MANUSCRIPT

be even higher if milder cases of sepsis had been considered, and patients presenting

with less severe sepsis make up the majority of sepsis deaths (1,3). In keeping with

previous studies (4,22), we found that a minority (7.4%) of ED admissions present with

sepsis, and that SIRS of non-infectious etiology was ~2 times more prevalent than

PT
sepsis.

Although this study was not designed to compare monocyte and neutrophil

RI
volumetric parameters to all other available sepsis biomarkers, MDW is shown to

SC
provide added value to the WBC, the biomarker most commonly used to screen for

sepsis in the ED. Three different approaches demonstrated added value of MDW in

U
comparison to WBC alone. (30) Area under the ROC curve are often criticized because

they provide an overall measure of diagnostic ability and are limited in terms of the
AN
detection of positive and negative patients (30), and the test statistics and variances

used for comparison of different AUC may be biased (40). Nonetheless, an increase in
M

AUC indicates improved diagnostic ability. CMH odds ratio is also indicative of an overall
D

added value of the diagnostic test and is valid even if data from the other tests used in
TE

the model are sparse (28). Decision curves are considered to ultimately quantify the

clinical relevance of the new test in comparison to the existing ones over a wide range of
EP

clinical situations wherein the risk of sepsis is variable, and all of these approaches

support indicate that MDW adds value to WBC for the detection of sepsis.
C

It is important to emphasize that the MDW and WBC are not comparable to other
AC

available sepsis biomarkers (e.g., PCT, CRP) because they are available to the

healthcare providers at an earlier stage of the clinical evaluation, at which time the

diagnosis of sepsis has not yet been considered. PCT and CRP, which are not

incorporated within the routine CBC, are exclusively employed to screen for sepsis in

patients who are deemed to have a high clinical likelihood of sepsis by clinicians in the

ED based upon the available evidence (e.g., WBC, SIRS criteria). Thus, the MDW and
 ACCEPTED MANUSCRIPT

WBC are more convenient for early sepsis detection and the diagnostic performance is

excellent (AUC 0.89), comparing favorably to that of PCT (41). Furthermore, a normal

MDW + WBC index is associated with a 98% negative predictive value which may prove

clinically useful in excluding a sepsis diagnosis. While further validation is needed, this

PT
relatively large prospective clinical trial that includes two distinct ED patient populations,

one a community hospital ED, the other a tertiary hospital ED, provides compelling

RI
evidence that MDW in combination with WBC could improve early sepsis detection in the

SC
ED population. A critical distinction of MDW from PCT, CRP, lactate and other existing

sepsis biomarkers is that the integration of MDW with WBC could occur during the initial

U
phase of ED evaluation, even before the healthcare provider has considered the

diagnosis of sepsis.
AN
There are several limitations of the study. First, there is no gold standard for the

diagnosis of sepsis and misclassification of sepsis and non-sepsis SIRS inevitably limits
M

the accuracy of the biomarker (42). Furthermore, it is unclear how the MDW will perform
D

for the newly established “Sepsis-3” criteria (8). In that regard, it is notable that a
TE

subgroup of patients with more severe sepsis presentations (i.e., equivalent to Sepsis-

3), had the highest MDW values (Figure S2) and Sepsis-2 criteria (SIRS) may be
EP

superior to Sepsis-3 criteria for identifying high risk patients in non-ICU populations (9).

A larger study is needed to confirm that MDW results can be standardized across
C

multiple sites using multiple instruments and to determine if MDW + WBC performs
AC

better when monitored over time. Ultimately, it will be important to determine how the

availability of the MDW influences the management of sepsis in the ED, including early

and appropriate antibiotic treatments, and sepsis-related clinical outcomes such as

hospital and ICU lengths of stay, morbidity (e.g., organ failures) and mortality. The study

was conducted over a 4 month period spanning the fall and early winter seasons during

which a seasonal effect on sepsis was evident (more cases in late fall/early winter),
 ACCEPTED MANUSCRIPT

emphasizing the diversity of sepsis cases and the need to consider how the test

performs in each sepsis subset based upon the infectious etiology (e.g., viral, bacterial,

fungal). Finally, host factors, as shown in Table S1, could influence the immune cell

volumetric analysis. Whereas the inclusion of these pre-existing conditions (present in

PT
17% of the ED patients) in our data analysis had only a small effect on the overall study

results (Table S6), the effects of these factors may need to be considered on a case-by-

RI
case basis in the ED setting.

SC
Conclusions:

U
This study shows that the monocyte distribution width (MDW), a measure of a change in

the size distribution of circulating monocytes, provides significant added value to WBC
AN
for the detection of sepsis in the ED population. From a practical perspective, when

validated by a larger prospective study, incorporation of MDW and WBC parameters

M

during initial CBC analysis could be readily used in the ED to provide a timely and
D

convenient sepsis diagnostic tool and lead to early initiation of antimicrobial therapy.
TE

Acknowledgements:
EP

Author contributions: E.D.C. assisted with study design and interpretation of the data, he

had full access to the study data and assumes responsibility for the integrity of the data
C

and the accuracy of the analysis; he also drafted manuscript. J.E.P., C.S., D.C.A., K.B.,
AC

D.C., J.W., L.T., F.C. and M.S. assisted with study design and interpretation of the data,

and edited the initial draft of the manuscript. L.H., E.R., contributed to data collection

and management. D.R.C., contributed to categorization of sepsis patients. R.M.

conducted all of the statistical analyses for the manuscript.

.
 ACCEPTED MANUSCRIPT

References Cited:

1. Liu V, Escobar GJ, Greene JD, et al. Hospital deaths in patients with sepsis from

2 independent cohorts. JAMA 2014; 312:90-2.

2. http://www.hcup-us.ahrq.gov/reports/statbriefs/sb160.pdf

PT
3. Jones SL, Ashton CM, Kiehne LB, et al. Outcomes and resources of sepsis-

associated stays by presence on admission, severity, and hospital time. Med

RI
Care 2016; 54:303-310.

SC
4. Wang HE, Shapiro NI, Angus DC, Yealy DM. National estimates of severe sepsis

in United States emergency departments. Crit Care Med 2007; 35:1928-36.

U
5. Ferrer R, Martin-Loeches I, Phillips G, et al. Empiric antibiotic treatment reduces

mortality in severe sepsis and septic shock from the first hour: results from a
AN
guideline-based performance improvement program. Crit Care Med 2014;

42:1749-55.
M

6. Kumar A, Roberts D, Wood KE, et al. Duration of hypotension before initiation of

D

effective antimicrobial therapy is the critical determinant of survival in human

TE

septic shock. Crit Care Med 2006;34:1589-1596.

7. Gaieski DF, Mikkelsen ME, Band RA, et al. Impact of time to antibiotics on
EP

survival in patients with severe sepsis or septic shock in whom early goal-

directed therapy was initiated in the emergency department. Crit Care Med
C

2010;38:1045-53.
AC

8. Singer M, Deutschman CS, Seymour CW, et al. The Third International

Consensus Definitions for Sepsis and Septic Shock (Sepsis-3). JAMA

2016;315(19):801-810.

9. Churpek MM, Snyder A, Han X, et al. qSOFA, SIRS, and early warning scores

for detecting clinical deterioration in infected patients outside the ICU. Am J

Respir Crit Care Med 2017 (in press) PMID:27649072.

 ACCEPTED MANUSCRIPT

10. Surviving Sepsis Campaign Responds to Sepsis-3. March 1, 2016.

http://survivingsepsis.org/SiteCollectionDocuments/SSC-Statements-Sepsis-

Definitions-3-2016.pdf

11. Seymour CW, Liu VX, Iwashyna TJ, et al. Assessment of clinical criteria for

PT
sepsis: For the Third International Consensus Definitions for Sepsis and Septic

Shock (Sepsis-3). JAMA 2016; 315:762-774.

RI
12. Filbin MR, Arias SA, Camargo CA, Barche A, Pallin DJ. Sepsis visits and

SC
antibiotic utilization in U.S. emergency departments. Crit Care Med 2014; 42:528-

535.

U
13. Horeczko T, Green JP, Panacek EA. Epidemiology of the systemic inflammatory

response syndrome (SIRS) in the emergency department. West J Emerg Med

AN
2014; 15:329-336.

14. Raimondi F, Ferrara T, Capasso L, Sellitto M, Landolfo F, Romano A, Grimaldi E,

M

Scopacasa F. Automated determination of neutrophil volume as screening test

D

for late-onset sepsis in very low birth infants. Pediatr Infect Dis J 2010;29:288.
TE

15. Dilmoula A, Kassengera Z, Turkan H, Dalcomune D, Sukhachev D, Vincent JL,

Pradier O. Volume, conductivity and scatter properties of leukocytes (VCS

EP

technology) in detecting sepsis in critically ill adult patients. Blood

2011;118:4729 (Abstract).
C

16. Celik IH, Demirel G, Askoy HT, Erdeve O, Tuncer E, Biyikli Z, Dilmen U.
AC

Automated determination of neutrophil VCS parameters in diagnosis and

treatment efficacy of neonatal sepsis. Pediatr Res 2012;71:121-125.

17. Bhargava M, Saluja S, Sindhuri U, Saraf A, Sharma P. Elevated mean neutrophil

volume + CRP is a highly sensitive and specific predictor of neonatal sepsis. Int

J Lab Hematol 2013;[Epub ahead of print].

18. Lee AJ, Kim SG. Mean cell volumes of neutrophils and monocytes are promising
 ACCEPTED MANUSCRIPT

markers of sepsis in elderly patients. Blood Res 2013;48:193-7.

19. Chaves F, Tierno B, Xu D. Quantitative determination of neutrophil VCS

parameters by the Coulter automated hematology analyzer: new and reliable

indicators for acute bacterial infection. Am J Clin Pathol 2005;124:440-444.

PT
20. Chaves F, Tierno B, Xu D. Neutrophil volume distribution width: a new

automated hematologic parameter for acute infection. Arch Pathol Lab Med

RI
2006;130:378–380.

SC
21. Mardi D, Fwity B, Lobmann R, Ambrosch A. Mean cell volume of neutrophils and

monocytes compared with C-reactive protein, interleukin-6 and white blood cell

U
count for prediction of sepsis and nonsystemic bacterial infections. Int J Lab

Hematol 2010; 32:410–418.

AN
22. Rezende E, Silva JM, Isola AM, Campos EV, Amendola CP, Almeida SL.

Epidemiology of severe sepsis in the emergency department and difficulties with

M

initial assistance. Clinics (Sao Paulo) 2008; 63:457-464.

D

23. Burderer NM. Statistical methodology. I. Incorporating the prevalence of disease

TE

into the sample size calculation for sensitivity and specificity. Acad Emerg Med

1996; 3(9):895-90.
EP

24. Levy MM, Fink MP, Marshall JC, et al. 2001 SCCM/ESICM/ACCP/ATS/SIS

international sepsis definitions conference. Crit Care Med 2003; 31:1250-56.

C

25. Members of the American College of Chest Physicians/Society of Crit Care Med
AC

Consensus Conference Committee: American College of Chest

Physicians/Society of Crit Care Med Consensus Conference: Definitions for

sepsis and organ failure and guidelines for the use of innovative therapies in

sepsis. Crit Care Med 1992; 20:864-874.

26. Aslinia F, Mazza JJ, Yale SH. Megaloblastic anemia and other causes of

macrocytosis. Clin Med Res 2006; 4:236-241.

 ACCEPTED MANUSCRIPT

27. De Long, DeLong, and Clarke-Pearson (1988). Comparing the Areas under Two

or More Correlated Receiver Operating Characteristic Curves: A Nonparametric

Approach. Biometrics 1988; 44:837-845.

28. Agresti, Alan (2002). Categorical Data Analysis. Hooken, New Jersey: John

PT
Wiley & Sons, Inc. p. 413.

29. Vickers AJ, Elkin EB. Decision curve analysis: a novel method for evaluating

RI
prediction models. Med Decis Making 2006; 26:565-574.

SC
30. Moons KGM, de Groot JAH, Linnet K, Reitsma JB, Bossuyt PMM. Quantifying

the added value of a diagnostic test or marker. Clin Chem 2012; 58:1408-1417.

U
31. Martin GS, Mannino DM, Easton S, Moss M. The epidemiology of sepsis in the

United States from 1979 through 2000. New Engl J Med 2003; 348:1546-1554.
AN
32. Fitrolaki DM, Dimitriou H, Kalmanti M, Biassoulis G. CD64-neutrophil expression

and stress metabolic patterns in early sepsis and severe traumatic brain injury in
M

children. BMC Pediatr 2013; 13:31.

D

33. Henderson RB, Hobbs JA, Mathies M, Hogg N. Rapid recruitment of

TE

inflammatory monocytes is independent of neutrophil migration. Blood 2003;

102:328-335.
EP

34. Italiani P, Boraschi D. From monocytes to M1/M2 macrophages, phenotypical vs.

functional differentiation. Front Immunol 2014; 2014;5:514.

C

35. Xu D. Clinical applications of leukocyte morphological parameters. Int J Clin Res

AC

2015;1:1.

36. McCullough KC, Basta S, Knotig S, et al. Intermediate stages in monocyte-

macrophage differentiation modulate phenotype and susceptibility to virus

infection. Immunology 1999; 98:203-212.

37. Wang SY, Mak KL, Chen LY, Chou MP, Ho CK. Heterogeneity of human blood
 ACCEPTED MANUSCRIPT

monocyte: two subpopulations with different sizes, phenotypes and functions.

Immunology 1992; 77:298-303.

38. Iwasaki A, Pillai PS. Innate immunity to influenza virus infection. Nature Reviews

Immunology 2014; 14:315-328.

PT
39. Hou W, Gibbs JS, Lu X, et al. Viral infection triggers rapid differentiation of

human blood monocytes into dendritic cells. Blood 2012;119:3128-3131.

RI
40. Seshan VE, Gonen M, Begg CB (2013). Comparing ROC curves derived from

SC
regression models. Statistics in medicine 30, 32(9), 1483-1493.

41. Kibe S, Adams K, Barlow G. Diagnostic and prognostic biomarkers of sepsis in

U
critical care. J Antimicrob Chemother 2011; 66:ii30-ii40.

42. Lee TS, Hui Q. Testing the assumption of non-differential misclassification in

AN
case-control studies. Journal of Modern Applied Statistical Methods 2013;

12:211-30.
M
D
TE

Figure Legends:

Figure 1. Cell Population Distribution (CPD) Analysis. Representative histograms of

EP

white blood cell populations derived from the Beckman Coulter DxH 800 analyzer. The

histogram to the left is an example of a non-septic donor. The 2D histogram to the right
C

corresponds to an example of a patient with septic shock. The rotated one-dimensional

AC

histogram on the left represents the distribution of the monocyte population

volumes. The dotted blue line on top of the distribution represents the mean monocyte

volume (MMV). The dotted red line represents 1 standard deviation from the mean of the

distribution [a.k.a., monocyte distribution width (MDW)], which is shown to be increased

in the septic patient.

 ACCEPTED MANUSCRIPT

Figure 2. Neutrophil and Monocyte CPD performance for sepsis in the ED

population. Panel A, mean neutrophil volume (MNV) is noted to be lowest in controls,

PT
highest in the sepsis group, with intermediate values in SIRS and infection groups.

Panel B, the neutrophil distribution width (NDW) shows a similar pattern. [*P < 0.001 for

RI
sepsis vs control or infection or SIRS (individual comparisons) and for sepsis vs (control

+ infection + SIRS groups combined)]. Panel C, mean monocyte volume (MMV) is noted

SC
to be lowest in controls, highest in the sepsis group, with intermediate values in SIRS

and infection groups. Panel D, the monocyte distribution width (MDW) shows a similar

U
pattern, and the MDW is statistically higher in the sepsis group compared to each of the
AN
other groups. [*P < 0.001 for sepsis vs control or infection or SIRS (individual

comparisons) and for sepsis vs (control + infection + SIRS groups combined)].

M
D

Figure 3. Receiver operating characteristic (ROC) for sepsis and SIRS in ED

subjects using MDW alone and in combination with the WBC. Panel A, comparing
TE

sepsis to all other ED patients: the area under the curve (AUC) for MDW (cutoff value

20.5) is comparable to WBC (cutoff value of 12.0); however, the addition of MDW to
EP

WBC (using 20.5 and 12.2 as the cutoff values for MDW and WBC, respectively)

increases the AUC by 8% relative to WBC alone, which is statistically significant (see
C

text).. Panel B, comparing sepsis to SIRS, and using the same cutoff values, the AUC
AC

for MDW is higher than for WBC and is higher yet when MDW is added to WBC.

Figure 4. Decision probability curves for sepsis detection reflecting the added

value of MDW compared to WBC alone. The net benefit of WBC alone (blue line) and

in combination with MDW (red line) for the detection of sepsis was calculated over a
 ACCEPTED MANUSCRIPT

range of pre-test probabilities for sepsis (see methods for details). The Control (green

line) represents a model that provides no insight into the diagnosis of sepsis. MDW +

WBC shows greater benefit compared to WBC alone over a wide range of sepsis

probabilities (2-70%). The models converge above 70% sepsis probability, a situation

PT
that is not representative of the ED population.

RI
U SC
AN
M
D
TE
C EP
AC
 ACCEPTED MANUSCRIPT

PT
RI
U SC
AN
M
D
TE
EP
C
AC
 ACCEPTED MANUSCRIPT

PT
RI
U SC
AN
M
D
TE
EP
C
AC
 ACCEPTED MANUSCRIPT

PT
RI
U SC
AN
M
D
TE
EP
C
AC
 ACCEPTED MANUSCRIPT

PT
RI
U SC
AN
M
D
TE
EP
C
AC
 ACCEPTED MANUSCRIPT

PT
RI
U SC
AN
M
D

e-Figure 1. White blood count (WBC) and neutrophil percent (NE%) performance for
sepsis in the ED population. Left Panel, WBC is noted to be elevated in the SIRS and sepsis
groups relative to control and infection groups. Right Panel, NE% is elevated in SIRS and sepsis
TE

groups compared to control and infection groups. There was no statistically significant difference
of either parameter when comparing SIRS and sepsis.
C EP
AC

Online supplements are not copyedited prior to posting and the author(s) take full responsibility for the accuracy of all data.
 ACCEPTED MANUSCRIPT

PT
RI
U SC
AN
M
D
TE
C EP
AC

e-Figure 2. Monocyte distribution width (MDW) increases with severity of infection in the
ED population. MDW is shown to incrementally increase when comparing controls, SIRS and
infection groups. More dramatic increases in MDW are observed in patients with sepsis and are
highest in those with “advanced sepsis” (severe sepsis + septic shock, labeled as “Severe Sepsis”).
[*P < 0.01 advanced sepsis vs sepsis; P < 0.001 Advanced sepsis vs non-septic (control +
infection + SIRS)].

Online supplements are not copyedited prior to posting and the author(s) take full responsibility for the accuracy of all data.
 ACCEPTED MANUSCRIPT

 
e-Table 1: Preexisting conditions associated with macrocytosis or otherwise expected to promote
impaired immune responses in the ED population.

Condition Description
Neutropenia Absolute neutrophil count <1,500/µL

PT
Drugs used to treat Neupogen, Filgrastim, Granix, Sargramostime.
neutropenia
Immune HIV infection, organ or bone marrow transplantation, malignancy,
compromised state hematologic (lymphomas, multiple myeloma, acute and chronic

RI
leukemias, myelodysplastic syndromes, aplastic anemia).
Immune modulating Chronic treatments (≥2 weeks) with corticosteroids (prednisone ≥20
drugs mg/day), anti-TNFα molecules, methotrexate (≥15 mg/week),

SC
azathioprine (≥75 mg/day), cyclophosphamide (any dose) or
mycophenylate (≥1000 mg/day) or any combination of these
treatments (at any dose).

U
Hydroxyurea Treatment within one week
Alcoholism As defined by the CDC (≥15 drinks/week for men; ≥8 drinks/week
AN
for women) or a blood alcohol level exceeding the legal limit
(0.08%)
Current antibiotic Subjects treated with antibiotics within the past 7 days prior to ED
use presentation (i.e., recently infected).
M
D
TE
C EP
AC

Online supplements are not copyedited prior to posting and the author(s) take full responsibility for the accuracy of all data.
 ACCEPTED MANUSCRIPT

 
e-Table 2. Emergency Department Population Demographics

Summary Demographics by Group

Sepsis SIRS Non-Sepsis

Total Subjects 98 203 1222

PT
Subject Age - Mean (Min - Max) 50 (18 - 84) 48 (18 - 89) 46 ( 18 - 90)

RI
Male gender, no. (%) 43 (44%) 97 (48%) 610 (50%)
Race:
African/Black 16 56 421

SC
American Indian or Alaska Native 1 0 1
Asian 0 5 19
Middle Eastern 0 0 1

U
Native Hawaiian, other Pacific
0 2 4
Islander
White
Unknown
AN 77
4
131
9
729
47
Comorbid Illness, no. (%):
Alcoholism 2 (2) 17 (8) 43 (4)
M
Malignancy 16 (16) 14 (7) 33 (3)
Renal Disease 13 (13) 14 (7) 41 (4)
Liver Disease 3 (3) 2 (3) 25 (2)
D

Post-Surgical 5 (5) 0 (0) 3 (0)

Bronchitis / COPD 1 (1) 8 (4) 11 (1)
TE
C EP
AC

Online supplements are not copyedited prior to posting and the author(s) take full responsibility for the accuracy of all data.
 ACCEPTED MANUSCRIPT

 
e-Table 3. Summary of Cell Population Data (CPD) and CBC data from the ED population
(n=1320 total) based upon clinical category.

Standard
Analyte Status N Mean Deviation Min Max
MCV Control 879 88.02 6.56 58.96 115.00

PT
Infection 140 88.32 6.13 71.75 107.67
SIRS 203 87.93 7.75 62.36 118.78
Sepsis 98 87.36 7.05 59.00 106.33

RI
MMV Control 879 176.62 8.42 158.66 228.55
Infection 140 180.90 8.58 162.46 221.71
SIRS 203 177.44 8.70 157.09 206.29
Sepsis 98 186.94 11.20 163.01 220.16

SC
MDW Control 879 19.24 2.39 14.09 50.49
Infection 140 20.25 2.25 15.12 26.40
SIRS 203 19.99 3.18 14.74 37.71
Sepsis 98 22.63 3.66 15.48 37.33

U
MNV Control 879 151.38 8.73 130.47 188.25
Infection 140 153.59 8.58 132.45 178.39
SIRS
Sepsis
203
98
AN
153.61
159.09
9.46
11.01
132.05
140.53
189.43
189.16
NDW Control 879 18.25 1.84 14.81 35.90
Infection 140 18.75 1.90 15.09 30.21
M
SIRS 203 19.50 3.33 15.38 39.34
Sepsis 98 20.61 3.33 17.27 42.99
MO Control 879 8.25 2.97 0.31 27.91
D

Infection 140 8.60 3.18 2.97 25.77

SIRS 203 7.73 4.42 0.95 40.64
Sepsis 98 8.08 7.82 0.89 76.33
TE

NE Control 879 62.51 12.62 21.31 98.14

Infection 140 67.01 14.31 23.78 93.56
SIRS 203 71.38 15.27 2.62 93.71
Sepsis 98 80.15 13.55 2.88 94.91
EP

PLT Control 879 235.19 80.55 10.22 660.65

Infection 140 249.54 91.02 66.61 746.13
SIRS 203 253.49 98.87 3.51 575.33
Sepsis 98 259.45 117.77 15.40 635.09
C

WBC Control 879 8.62 3.47 1.23 60.36

Infection 140 10.00 3.82 3.15 26.16
AC

SIRS 203 12.63 7.08 1.01 80.81

Sepsis 98 15.61 6.53 0.39 32.80

MCV = mean corpuscular volume

MMV = mean monocyte volume
MDW = monocyte distribution width
MNV = mean neutrophil volume
NDW = neutrophil distribution width
MO = monocyte %
NE = neutrophil %
PLT = platelet count in thousands
WBC = white blood cell count in thousands

Online supplements are not copyedited prior to posting and the author(s) take full responsibility for the accuracy of all data.
 ACCEPTED MANUSCRIPT

 
e-Table 4. Cutoff values for each measure CPD and CBC parameter, and the distribution of ED
patients (n=1320 total) based upon test performance.

Sepsis vs. Non-Sepsis

True False True False

PT
Analyte Cut-off TOTAL
Positive Negative Negative Positive
MCV 88.00 49 49 612 610 1320
MMV 180.00 69 29 822 400 1320

RI
MDW 20.50 75 23 873 349 1320
MNV 152.00 70 28 659 563 1320

SC
NDW 18.80 73 25 819 403 1320
MO 6.30 37 61 887 335 1320
NE 75.00 71 27 938 284 1320
PLT 250.00 45 53 734 488 1320

U
WBC 12.00 73 25 983 239 1320

MCV = mean corpuscular volume

AN
MMV = mean monocyte volume
MDW = monocyte distribution width
M
MNV = mean neutrophil volume
NDW = neutrophil distribution width
MO = monocyte %
D

NE = neutrophil %
PLT = platelet count in thousands
WBC = white blood cell count in thousands
TE
C EP
AC

Online supplements are not copyedited prior to posting and the author(s) take full responsibility for the accuracy of all data.
 ACCEPTED MANUSCRIPT

 
e-Table 5. Clinical accuracy of CPD parameters for sepsis detection in the ED population.

Sepsis vs. Non-Sepsis

95% 95%
Confidence Confidence 95% Confidence
Parameter AUC Cut-Off Sensitivity Specificity
Interval Interval Interval

PT
Lower Upper Lower Upper Lower Upper
WBC 0.81 0.76 0.86 12.00 0.75 0.65 0.82 0.80 0.78 0.83
NE 0.83 0.79 0.88 75.00 0.73 0.63 0.80 0.77 0.74 0.79

RI
MNV 0.69 0.63 0.74 152.00 0.71 0.62 0.79 0.54 0.51 0.57
NDW 0.77 0.73 0.82 18.80 0.75 0.65 0.82 0.67 0.64 0.70
MMV 0.76 0.71 0.81 180.00 0.70 0.61 0.79 0.67 0.65 0.70

SC
MDW 0.78 0.73 0.84 20.50 0.77 0.67 0.84 0.72 0.69 0.74

U
MDW = monocyte distribution width
MMV = mean monocyte volume
MNV = mean neutrophil volume
NDW = neutrophil distribution width
AN
NE = neutrophil %
M
WBC = white blood cell count
D
TE
C EP
AC

Online supplements are not copyedited prior to posting and the author(s) take full responsibility for the accuracy of all data.
 ACCEPTED MANUSCRIPT

 
e-Table 6. Effect of preexisting conditions associated with altered immune status (see e-Table 1)
on the performance of CPD and CBC parameters for discrimination of sepsis from SIRS in the ED
population. Exclusion of ~17% of ED patients based upon conditions that are known to have risk
factors for altered immune status or macrocytosis (i.e., “without” exclusion criteria) modestly
improves the performance of CPD and CBC parameters in the ED population.

PT
Sepsis vs. SIRS
95% 95% 95%
Pre-
Confidence Cut- Confidence Confidence

RI
Parameter Existing AUC Sensitivity Specificity
Interval Off Interval Interval
Conditions
Lower Upper Lower Upper Lower Upper
Without 0.73 0.64 0.82 12.00 0.81 0.67 0.90 0.39 0.32 0.47

SC
WBC
With [All] 0.66 0.59 0.73 12.00 0.75 0.65 0.82 0.41 0.35 0.48
Without 0.76 0.69 0.84 75.00 0.79 0.64 0.88 0.52 0.44 0.59
NE
With [All] 0.70 0.64 0.77 75.00 0.73 0.63 0.80 0.522 0.45 0.59

U
Without 0.69 0.60 0.78 152.00 0.81 0.67 0.90 0.46 0.38 0.53
MNV
With [All] 0.64 0.58 0.71 152.00 0.71 0.62 0.79 0.44 0.38 0.51

NDW
Without
With [All]
0.75
0.68
0.67
0.61
0.83
0.74
AN
18.80
18.80
0.74
0.75
0.59
0.65
0.85
0.82
0.58
0.52
0.50
0.45
0.65
0.59
Without 0.77 0.69 0.85 180.00 0.69 0.54 0.81 0.70 0.62 0.76
MMV
M
With [All] 0.75 0.69 0.81 180.00 0.70 0.61 0.79 0.66 0.59 0.72
Without 0.78 0.70 0.87 20.50 0.79 0.64 0.88 0.68 0.60 0.75
MDW
With [All] 0.74 0.68 0.80 20.50 0.77 0.67 0.84 0.63 0.56 0.70
D
TE

MDW = monocyte distribution width

MMV = mean monocyte volume
MNV = mean neutrophil volume
NDW = neutrophil distribution width
EP

NE = neutrophil %
WBC = white blood cell count (in thousands)
C
AC

170512

Online supplements are not copyedited prior to posting and the author(s) take full responsibility for the accuracy of all data.