CLINICAL RESEARCH STUDY

Use of Neutrophil Count in Early Diagnosis and Risk

Stratification of AMI
Julia Meissner, MD,a Affan Irfan, MD,a Raphael Twerenbold, MD,a Sandra Mueller,a Miriam Reiter, MD,a
Philip Haaf, MD,a Tobias Reichlin, MD,b Nora Schaub, MD,a Katrin Winkler, MD,c Otmar Pfister, MD,b
Corinna Heinisch, MD,a Christian Mueller, MDa
a
Department of Internal Medicine and bDepartment of Cardiology, University Hospital Basel, Basel, Switzerland; cServicio de
Urgencias y Pneumologia, CIBERES ISC III, Hospital del Mar – Institut Municipal d’Investigació Mèdica, Barcelona, Spain.

ABSTRACT

BACKGROUND: Neutrophils are rapidly released into the circulation upon acute stress such as trauma or
acute myocardial infarction (AMI). We hypothesized that neutrophil count might provide incremental
value in the early diagnosis and risk stratification of AMI.
METHODS: We conducted a prospective observational multicenter study to examine the diagnostic accuracy
of the combination of neutrophil count and cardiac troponin T from 1125 consecutive patients who
presented to the Emergency Department with symptoms suggestive of acute myocardial infarction. The
final diagnosis was adjudicated by 2 independent cardiologists.
RESULTS: Neutrophil count was higher in patients with acute myocardial infarction compared with other
diagnoses (median 6.7 vs. 5.0 ⫻ 109/L, respectively, P ⬍.001). The accuracy of the neutrophil count for
diagnosing acute myocardial infarction, quantified by the area under the receiver operating characteristic
curve (AUC) was 0.69, which was significantly lower than that of cardiac troponin T (AUC 0.89, P ⬍.001).
The combination of the neutrophil count and cardiac troponin T did not improve the early diagnosis of
acute myocardial infarction versus cardiac troponin T alone (P ⫽ .79). The prognostic accuracy of
neutrophil count for death and AMI was significantly lower than that of cardiac troponin T. However,
patients in the highest tertile of neutrophil count had a significantly increased risk of death and AMI at 90
and 360 days compared with patients in the lowest tertile (hazard ratios 2.47 [95% confidence interval,
1.63–3.72] and 2.28 [95% confidence interval, 1.55–3.36], respectively).
CONCLUSION: The neutrophil count does not improve the early diagnosis of AMI in patients presenting
with chest pain but identifies patients at increased risk of death.
© 2011 Elsevier Inc. All rights reserved. • The American Journal of Medicine (2011) 124, 534-542

KEYWORDS: Diagnosis; Myocardial infarction; Neutrophils; Risk stratification

Despite being one of the leading causes of death worldwide,
the early diagnosis of acute myocardial infarction still re-
mains a challenge for the clinician. Currently, cardiac tro-
ponins, in conjunction with the clinical assessment and
Funding: None.
Conflict of Interest: None of the authors has any potential financial or
nonfinancial conflicts of interest.
Authorship: All authors meet criteria for authorship, had full access to
the data, participated in the preparation of the manuscript and accept
responsibility for the scientific content of the manuscript and have read and
approved the final manuscript.
Requests for reprints should be addressed to Christian Mueller, MD,
Department of Internal Medicine, University Hospital Basel, Petersgraben
4, Basel CH-4031, Switzerland.
E-mail address: chmueller@uhbs.ch
electrocardiogram (ECG),1 are considered in diagnosing
acute myocardial infarction.2 However, the early diagnosis
of myocardial necrosis is still limited by the delayed in-
crease of the circulating levels of cardiac enzymes,3 which
means a diagnostic gap within these early hours.
There are several different pathophysiological responses
to stress in the human body, including the release of the
acute phase reactants and hormones. Recently we4 and oth-
ers5 were able to confirm this hypothesis by quantifying the
endogenous stress response with copeptin, the c-terminal
part of the vasopressin prohormone, which correlates with
the individual stress level.6
Under normal conditions, neutrophils are produced in the
bone marrow by granulopoiesis, and only a small fraction of

0002-9343/$ -see front matter © 2011 Elsevier Inc. All rights reserved.
doi:10.1016/j.amjmed.2010.10.023
Meissner et al Neutrophil Count and AMI 535

the total bone marrow neutrophil pool is released into cir- Outcome of Interests
culation. This large storage pool of mature neutrophils in the Primary Outcome. Evaluate the diagnostic accuracy of
bone marrow, termed the bone marrow reserve, may be neutrophil count among patients presenting with chest pain.
rapidly mobilized during acute stress.7 Neutrophils also are
considered to be the hallmark of inflammation, which is Secondary Outcome. Assess the prognostic value of neu-
closely associated in the formation trophil count in determining the
and rupture of atherosclerotic risk of mortality among patients
plaque. Numerous experimental with chest pain.
CLINICAL SIGNIFICANCE
ischemia-reperfusion models sug-
gest the pathological significance ● Despite significantly higher neutrophil Routine Clinical Data
of neutrophil-platelet interaction count in acute myocardial infarction pa- For all patients, a routine initial
in acute coronary syndrome.8,9 tients, its ability to diagnose acute clinical assessment was carried
Hence, many authors have exam-
myocardial infarction is lower than that out that, in general, included clin-
ined the prognostic value of in-
of cardiac troponin T, alone and in ical history, physical examination,
flammatory markers in predicting
10-12 combination. 12-lead ECG, continuous ECG
cardiovascular risk. monitoring, pulse oximetry, and
Based on our encouraging find- ● White blood cell count and its differen- chest radiography. Timing and
ing about the incremental diagnos- tials have poor prognostic ability, but treatment of patients were left to
tic value of the endogenous stress may be used to identify patients at in- the discretion of the attending
quantified by copeptin, we sought creased risk for short- and long-term physician and were performed ac-
to investigate whether quantify-
death or nonfatal acute myocardial in- cording to the standard practice of
ing stress response by neutrophil
farction in all chest pain patients, when the hospital.
count also may be helpful in the
early diagnosis of acute myocardial cardiac troponin T is unavailable.
infarction. As secondary objec-
Laboratory Analysis
The blood samples were collected
tives, we also evaluated the diag-
in ethylenediamenetetraacetic acid-
nostic and prognostic performance of other white blood cell
containing tubes from each patient at the time of presentation
differential count.
to the ED. Samples were assayed at the local laboratories.
Plasma samples were aliquot and stored in a freezer maintained
METHODS at ⫺70°C. These samples will be used for future testing.
Study Design and Setting
From April 2006 to March 2010, a total of 1247 consecutive Hematological Measurements. Total white blood cell
chest pain patients were enrolled in the Advantageous Pre- count and differential cell count (including neutrophils and
dictors of Acute Coronary Syndrome Evaluation study, a lymphocytes) were measured at presentation using a quanti-
prospective observational multicenter study in 6 tertiary tative hematology analyzer, at the local laboratories. The co-
hospitals, coordinated by the University Hospital of Basel, efficient of variation for white blood cell count and differential
Switzerland. All participating centers were required to adhere cell count were ⬍2.5% at all the participating centers. Neutro-
to a standard protocol for data and sample collection, storage, phil/lymphocyte ratio was obtained by dividing total count of
and laboratory analyses to ensure quality control. neutrophils by lymphocytes count.

Population Cardiac Enzymes. Troponin T or troponin I, myoglobin,

All patients presenting to the Emergency Department (ED)
with symptoms suggestive of acute myocardial infarction
such as chest pain and angina pectoris were screened for
study eligibility. Criteria for inclusion were patients with
new onset or peak of chest pain within the last 12 hours.
Patients with terminal kidney failure requiring dialysis and
patients with acute trauma were excluded. Participants with
preexisting diseases or medication use known to influence
white blood cell count or its differential cell counts were
also excluded from the current analyses.
The study was approved by the institutional ethics com-
mittee of the participating centers, and was carried out
according to the principles of the Declaration of Helsinki.
Written informed consent was obtained from all included
patients.
creatine kinase, and creatine kinase-MB were measured at
presentation, and after 3 hours and 6-9 hours as long as
clinically indicated. When only cardiac troponin I was mea-
sured for any patient from the local laboratory, the stored
samples were used by an external laboratory to measure
cardiac troponin T (TnT4; Roche, Basel, Switzerland),
which was then used for this analysis.
Adjudicated Final Diagnosis
To determine the final diagnosis for each patient, 2 inde-
pendent cardiologists reviewed all available medical records
(clinical history, findings on physical examination, and re-
sults of laboratory tests, radiologic testing, ECG, echocar-
diography, cardiac exercise test, coronary angiography)
from the time of the patient’s arrival in the ED to the end of
536 The American Journal of Medicine, Vol 124, No 6, June 2011

the 90-day follow-up period. When there was disagreement
about the diagnosis, cases were reviewed and adjudicated in
conjunction with a third cardiologist. The details of the
predefined criteria for diagnosis have been mentioned
elsewhere.13
Statistical Analysis
Analysis was made using Kruskal-Wallis, chi-squared,
Mann-Whitney U test, and Student’s t test as appropriate.
To identify the diagnostic and prognostic value of the 4
laboratory parameters (white blood cell count, neutrophil
count, neutrophil/lymphocyte ratio, and cardiac troponin T),
receiver operator characteristic (ROC) curves were per-
formed and the area under the curve was calculated. The
comparison of areas under the ROC curves (AUC) was
performed as recommended by DeLong et al.14 Multivariate
logistic regression was performed to determine other inde-
pendent variables included in the study associated with the
diagnosis of acute myocardial infarction.
Associations of the 4 laboratory parameters with mortal-
ity and acute myocardial infarction were assessed by rela-
tive risks derived from Cox regression analyses, that is,
hazard ratios (HRs). The multiple logistic and Cox regres-
sion analyses were adjusted for the confounders’ age, sex,
hypertension, coronary artery disease, peripheral vascular
disease, renal insufficiency, diabetes, systolic blood pres-
sure, previous myocardial infarction, previous stroke, and
abnormal ECG (ST elevation, ST depression, and T-wave
inversion). The time to death and acute myocardial infarc-
tion up to short term (90 days) and long term (360 days) in
patients with white blood cell count, neutrophil count, and
neutrophil/lymphocyte ratio levels categorized by tertiles

Table 1 Baseline Characteristics of 1125 Patients Enrolled in the APACE Study

Myocardial Infarction

All Patients Yes No

Characteristics n ⫽ 1125 n ⫽ 173 n ⫽ 952 P Value
Age (years), median [IQR] 64 [51-76] 73 [60-81] 62 [49-74] ⬍.001
Male sex, n (%) 755 (67) 124 (72) 631 (66) .165
BMI (kg/m2), median [IQR] 26 [24-30] 26 [24-29] 26 [24-30] .884
Cardiovascular risk factors, n (%)
Hypertension 722 (64) 134 (78) 588 (62) ⬍.001
Hypercholesterolemia 497 (44) 87 (50) 410 (43) .079
Renal insufficiency 107 (10) 31 (18) 76 (8) ⬍.001
Diabetes mellitus 228 (20) 45 (26) 183 (19) .041
Current smoking 383 (34) 53 (31) 330 (35) .301
Previous smoking 282 (25) 54 (31) 228 (24) .043
History-n (%)
Coronary artery disease 419 (37) 79 (46) 340 (36) .013
Previous myocardial infarction 275 (24) 56 (32) 219 (23) .008
Previous revascularization 268 (24) 41 (24) 227 (24) .967
Peripheral artery disease 76 (7) 22 (13) 54 (6) .001
Previous stroke 63 (6) 21 (12) 42 (4) ⬍.001
Medication – n (%)
ACE inhibitor 270 (24) 55 (32) 215 (23) .009
ARB 204 (18) 33 (19) 17 (18) .727
Beta-blocker 426 (38) 70 (41) 356 (37) .444
Aspirin 449 (40) 82 (47) 367 (39) .029
Vital status – median [IQR *]
Heart rate/min 76 [65-89] 81 [69-95] 75 [65-89] .001
Systolic blood pressure (mm Hg) 143 [127-160] 142 [125-163] 143 [128-159] .557
Diastolic blood pressure (mm Hg) 84 [75-93] 85 [75-94] 84 [75-93] .994
Glomerular filtration rate (GFR)
MDRD GFR (mL/min/1.73 m2), Mean, range 90 (9-176) 79 (9-160) 92 (9-176) ⬍.001
Electrocardiographic findings, n (%)
Left bundle branch block 43 (4) 17 (10) 26 (3) ⬍.001
ST-segment elevation 57 (5) 39 (23) 18 (2) ⬍.001
ST-segment depression 129 (12) 57 (33) 72 (8) ⬍.001
T-wave inversion 152 (14) 48 (28) 104 (11) ⬍.001
No significant electrocardiographic abnormalities 820 (73) 57 (33) 763 (81) ⬍.001
ACE inhibitor ⫽ angiotensin-converting enzyme inhibitor; APACE ⫽ Advantageous Predictors of Acute Coronary Syndrome Evaluation study;
ARB ⫽ angiotensin receptor blocker; BMI ⫽ body mass index; IQR ⫽ interquartile range; MDRD ⫽ Modification of Diet in Renal Disease.
Meissner et al Neutrophil Count and AMI 537

between the patients diagnosed with STEMI versus

NSTEMI (data not shown).

Diagnostic Accuracy
The diagnostic accuracy was significantly higher with car-
diac troponin T than with white blood cell count, neutrophil
count, and neutrophil/lymphocyte ratio (AUC [95% CI],
0.89 [0.86-0.92]; 0.67 [0.63-0.72]; 0.69 [0.65-0.73]; and
0.65 [0.61-0.70], respectively; P ⬍.001) (Figure 2A). The
AUC for white blood cell count was not statistically differ-

Figure 1 White blood cell count, neutrophil count, neutro-

phil/lymphocyte ratio, and cardiac troponin T at presentation to
the Emergency Department according to the adjudicated final
diagnosis, acute myocardial infarction. Boxes represent inter-
quartile ranges, while whiskers display ranges (without outliers
further than 1.5 interquartile ranges from the end of the box).

were analyzed by Kaplan-Meier survival curves and com-

pared by log-rank tests. For all tests, a P value of ⬍.05
(2-sided) was considered significant. All statistical anal-
yses were performed with the use of SPSS for Windows
version 15.0 (SPSS Inc., Chicago, Ill) and MedCalc soft-
ware version 9.6.4.0 (MedCalc Software bvba, Mari-
akerke, Belgium).

RESULTS
Characteristics of Patients
Of the 1170 consecutive patients enrolled in this study, 45
patients were excluded because of a steroid, immunosup-
pressive, or cytostatic therapy, or a history of malignant
blood disease. The baseline characteristics of the remaining
1125 patients are shown in Table 1.
Acute myocardial infarction was diagnosed in 173 (15.4%)
of the total patients. Of these patients, 42 (24%) suffered from
ST-segment elevation myocardial infarction (STEMI) and the
remaining 131 from non-ST-segment elevation myocardial in-
farction (NSTEMI). The other adjudicated final diagnoses
were unstable angina in 163 (14.5%), cardiac symptoms from
causes other than coronary artery disease in 146 (13%), non-
cardiac causes in 543 (48%), and symptoms of unknown origin Figure 2 (A) Receiver operating characteristic curves show-
in 100 patients (9%). ing the diagnostic accuracy of cardiac troponin T, white blood
As shown in Figure 1, patients with acute myocardial cell count, neutrophil count, and neutrophil/lymphocyte ratio,
and cardiac troponin T alone at presentation; and (B) in com-
infarction as a final diagnosis had significantly higher mea-
bination with white blood cell count, neutrophil count, and
surements in all 4 laboratory parameters than patients with neutrophil/lymphocyte ratio to diagnose acute myocardial in-
other adjudicated diagnoses (P ⬍.001). There was no sta- farction in all patients.
tistical difference in any of the 4 laboratory parameters
538
Figure 3 (A) Diagnostic performance of white blood cell
count, neutrophil count, neutrophil/lymphocyte ratio, and car-
diac troponin T for acute myocardial infarction in patients
presenting within 3 hours after the onset of symptoms. (B)
Comparison of the area under the receiver operating character-
istic (ROC) curve according to the time since the onset of
symptoms for the 4 laboratory assays performed on blood
samples obtained at presentation.
ent from that of neutrophil count (P ⫽ .14) and neutrophil/
lymphocyte ratio (P ⫽ .39), but the diagnostic accuracy of
neutrophil count was significantly higher than that of neu-
trophil/lymphocyte ratio (P ⫽ .04). The combination of
white blood cell count, neutrophil count, and neutrophil/
lymphocyte ratio with cardiac troponin T did not improve
the diagnostic accuracy of cardiac troponin T (Figure 2B).
Time Since Onset of Symptoms
As shown in Figure 3A, in patients presenting early (ⱕ3
hours) after onset of symptoms, the diagnostic accuracy of
The American Journal of Medicine, Vol 124, No 6, June 2011
cardiac troponin T remained significantly superior to white
blood cell count, neutrophil count, and neutrophil/lympho-
cyte ratio (P ⬍.001). The diagnostic performance of cardiac
troponin T increased with the onset of time for chest pain
but not for the white blood cell count, neutrophil count, and
neutrophil/lymphocyte ratio (Figure 3B).
Baseline Neutrophil Count as a Predictor for
Acute Myocardial Infarction
The optimal cutoff value of neutrophil count for distinguish-
ing between patients diagnosed with and without acute
myocardial infarction was found at 4.85 ⫻ 109/L at a sensi-
tivity of 81%, a specificity of 48%, with a negative predic-
tive value of 93%, a positive predictive value of 22%, and
a positive likelihood ratio of 1.55 and negative likelihood
ratio of 0.40. This positive likelihood ratio corresponds to a
posttest probability of 22%. After adjusting for baseline
characteristics, patients presenting with neutrophil count
higher as compared with lower than 4.85 ⫻ 109/L had a
more than 2-fold increase in the odds ratio of acute myo-
cardial infarction (adjusted odds ratio 2.44; 95% CI, 1.51-
3.94; P ⬍.001). Among all the cardiovascular risk factors
and other relevant history, a significantly higher neutrophil
count was associated with hypertension, renal insufficiency,
diabetes, current smokers, and previous stroke (P ⱕ.001 for
all variables).
Table 2 shows the validity indexes and the likelihood
ratios for the best discriminatory value derived from the
ROC and 2 values obtained by dividing the laboratory
parameters into tertiles, to diagnose acute myocardial in-
farction at presentation.
Prognostic Value for Death or Acute
Myocardial Infarction
A total of 26 (2%) patients died and 190 (19%) acute
myocardial infarctions occurred within the 90-day follow-
up, while 360-day follow-up had 49 (5%) deaths and 207
(23%) acute myocardial infarctions.
All 4 laboratory parameters were significantly higher in
nonsurvivors and patients who experienced an acute myo-
cardial infarction compared with the rest, in both short- and
long-term follow-up (P ⬍.0001 for all, data not shown). In
Cox regression analysis, white blood cell count, neutrophil
count, and cardiac troponin T were independent predictors
of death and acute myocardial infarction at 90 and 360 days;
neutrophil/lymphocyte ratio was not (Table 3). However,
they had a significantly lower accuracy to predict death and
acute myocardial infarction at 90 days and 360 days com-
pared with cardiac troponin T (P ⬍.001). When stratified to
tertiles, Kaplan-Meier analysis showed that there was sig-
nificant increase in clinical end point with the increase from
lowest to highest tertile (log-rank for all 3 variables;
P ⬍.001) (Figure 4, A-C).
Meissner et al Neutrophil Count and AMI 539

Table 2 Diagnostic Accuracy: Validity Indexes for Values from Receiver Operating Characteristic Curve and Tertiles to Diagnose
Acute Myocardial Infarction

Sensitivity Specificity PPV NPV LR⫹ (95% CI) LR⫺ (95% CI)
White blood cell count ⫻ 10 /L9

6.91* 86 37 20 94 1.36 (1.26-1.47) 0.38 (0.26-0.55)

7.37† 82 45 21 93 1.49 (1.37-1.64) 0.4 (0.29-0.55)
8.89* 54 71 25 89 1.84 (1.56-2.18) 0.65 (0.55-0.77)
Neutrophil count ⫻ 109/L
4.40* 86 38 20 94 1.38 (1.28-1.50) 0.38 (0.26-0.55)
4.85† 81 48 22 93 1.55 (1.41-1.70) 0.40 (0.29-0.55)
6.30* 56 72 26 90 1.98 (1.67-2.33) 0.61 (0.52-0.73)
Neutrophil/lymphocyte ratio
2.32* 80 36 19 91 1.25 (1.15-1.37) 0.55 (0.40-0.75)
2.60† 75 46 20 91 1.37 (1.23-1.52) 0.56 (0.43-0.73)
4.00* 55 71 26 90 1.91 (1.62-2.26) 0.63 (0.53-0.74)
CI ⫽ confidence interval; LR⫹ ⫽ positive likelihood ratio; LR⫺ ⫽ negative likelihood ratio; PPV ⫽ positive predictive value; NPV ⫽ negative predictive
value.
*The best cut off value as determined from the receiver operating characteristic curve.
†The lower and upper values to divide the variable into tertiles.

DISCUSSION sentation increased with the use of the neutrophil count

In this study we explored the diagnostic and predictive
value of neutrophil count for acute myocardial infarction, as
well as the short- and long-term prognostic accuracy of
white blood cell count, neutrophil count, and neutrophil/
lymphocyte ratio in comparison with cardiac troponin T, in
1125 patients presenting to ED with chest pain. We report 3
major findings.
First, we found a significantly higher white blood cell
count, neutrophil count, and neutrophil/lymphocyte ratio in
patients who were diagnosed with acute myocardial infarc-
tion than in patients with other diagnoses. This finding
supports the study by Zazula et al,15 who found similar
results in a small sample of acute myocardial infarction
patients compared with patients with unstable angina and
noncardiac chest pain. They also report modest accuracy of
neutrophil/lymphocyte ratio for diagnosing acute coronary
syndrome (AUC 0.63), which corroborates our finding of an
AUC of 0.65 for diagnosing acute myocardial infarction.
However, the optimal neutrophil count cutoff for differen-
tiating acute myocardial infarction and nonacute myocardial
infarction patients in our study cohort (ⱖ4.85 ⫻ 109/L)
showed a high sensitivity, but a low specificity, which
makes it unsuitable to be used as an isolated, or even in
combination, diagnostic tool for myocardial infarction,
when cardiac troponin T is obtainable. Interestingly, we also
observed that the diagnostic value of neutrophil count and
white blood cell count was significantly higher than that of
neutrophil/lymphocyte ratio.
Second, the diagnostic accuracy of white blood cell
count, neutrophil count, and neutrophil/lymphocyte ratio
were significantly inferior to that of cardiac troponin T,
alone and in combination with cardiac troponin T. One
previous study with 2952 consecutive chest pain patients
showed that the sensitivity of creatine kinase-MB at pre-
from 50% to 70%.16
Third, white blood cell count and neutrophil count, but
not neutrophil/lymphocyte ratio, were significant indepen-
dent predictors of short- and long-term mortality and acute
myocardial infarction. The highest tertiles of these labora-
tory parameters were associated with significant increased
risk of short-term and long-term combined outcome (death
and acute myocardial infarction) relative to the lowest
tertile.
There have been several previous studies suggesting that
white blood cell count or its differential cell counts may be
used as an independent predictive marker for adverse out-
come.17-19 The AUC of neutrophil count found in our study
to predict long-term (360 days) death and acute myocardial
infarction is similar to that found by Takahashi et al20 to
predict long-term cardiac events (AUC ⫽ 0.69). Papa et al17
demonstrated that high neutrophil/lymphocyte ratio resulted
in higher cardiac death in patients with chronic coronary
artery disease, and similar results were observed by Horen
et al.21 O’Donoghue et al19 found a higher baseline neutro-
phil count in patients that presented with STEMI who died
in the 30-day follow-up as compared with patients who
survived to 30 days.19 The European Prevalence of Infec-
tion in Intensive Care study also showed that high neutro-
phil count was associated with higher risk of long-term
mortality in patients undergoing high-risk angioplasty.22
There are several strengths of our study. All the previous
studies on the diagnostic and prognostic role of white blood
cell count and its differential cell counts for acute myocar-
dial infarction were either observed in a small number of
chest pain patients15,16 or were conducted in a selected
group diagnosed with acute myocardial infarction/coronary
artery disease.20,23,24 We evaluated the diagnostic value of
neutrophil count in a large group of consecutive chest pain
 540
Table 3 Independent Association of Risk Predictors for Mortality and Acute Myocardial Infarction at 90 Days and 360 Days Among 1125 Patients Presenting with Chest Pain

Short Term (90 Days) Long Term (360 Days)

Cox Regression* Cox Regression*

Variable AUC Hazard Ratio (95% CI) P Value AUC Hazard Ratio (95% CI) P Value
Cardiac troponin T † 0.85 (0.82-0.89) 1.61 (1.38-1.87) ⬍.001 0.81 (0.77-0.85) 1.53 (1.31-1.78) ⬍.001
White blood cell count † 0.66 (0.62-0.71) 1.10 (1.05-1.14) ⬍.001 0.66 (0.62-0.70) 1.09 (1.05-1.13) ⬍.001
Tertile Median‡ 1.98 (1.28-3.07) .002 1.81 (1.20-2.71) .004
Highest‡ 2.81 (1.86-4.25) ⬍.001 2.52 (1.71-3.71) ⬍.001
Neutrophil count † 0.67 (0.63-0.72) 1.10 (1.05-1.15) ⬍.001 0.68 (0.64-0.72) 1.10 (1.05-1.14) ⬍.001
Tertile Median‡ 1.78 (1.15-2.76) .01 1.58 (1.05-2.38) .03
Highest‡ 2.47 (1.63-3.72) ⬍.001 2.28 (1.55-3.36) ⬍.001
Neutrophil/lymphocyte ratio † 0.64 (0.60-0.69) 1.02 (1.00-1.04) .07 0.65 (0.60-0.69) 1.02 (1.00-1.03) .06
Tertile Median‡ 1.06 (0.69-1.61) .8 0.98 (0.66-1.46) .94
Highest‡ 1.67 (1.14-2.45) .009 1.59 (1.11-2.28) .01
AUC ⫽ area under receiver operating characteristic curve; CI ⫽ confidence interval.
*Adjusted for age, sex, hypertension, coronary artery disease, peripheral vascular disease, renal insufficiency, diabetes, systolic blood pressure, previous myocardial infarction, previous stroke and abnormal
electrocardiogram (ST-elevation, ST-depression and T wave inversion).
†The data were entered as continuous variables.
‡Tertiles versus the lowest tertile.

The American Journal of Medicine, Vol 124, No 6, June 2011

tertiles (univariate analysis).
cell count, neutrophil count, and neutrophil/lymphocyte ratio
survival distribution for the different levels of the white blood
tiles, respectively. *Log-rank (Mantel-Cox) test of equality of
(B) neutrophil count, and (C) neutrophil/lymphocyte ratio ter-
survival curves for long term for (A) white blood cell count,
Figure 4 Kaplan-Meier plots for the cumulative event-free
Meissner et al Neutrophil Count and AMI
patients presenting to EDs of multiple centers. Therefore,
our study extends the previous results to the larger popula-
tion of consecutive, unselected patients presenting to the ED
with chest pain. We also compared the diagnostic and pre-
dictive ability of white blood cell count, neutrophil count,
and neutrophil/lymphocyte ratio with the gold-standard and
more conventional marker for myocardial necrosis, cardiac
troponin T, alone and in combination. We also calculated
the sensitivity and specificity, and the likelihood ratios,
which are the 2 most commonly accepted parameters de-
scribing the diagnostic power of clinical tests.25 When com-
bined with an accurate clinical diagnosis, likelihood ratios
from ancillary tests improve diagnostic accuracy in a syn-
ergistic manner.26
All the hematological investigations of interest were
routine blood tests (white blood cell count and its differ-
ential cell count) measured by standard laboratory meth-
ods at 6 different centers. There was no statistical differ-
ence in the neutrophil count among the patients recruited
between the different centers. Therefore, we believe that
the results can be safely generalized to all acute chest
pain patients in the corresponding ordinary clinic and
laboratory setting.
There are a few limitations of this study that merit
consideration. First, other markers known to be involved
in inflammation, such as C-reactive protein,11 fibrino-
gen,16 and interleukin-10,27 were not evaluated. A mul-
timarker approach including neutrophil count may be of
use in risk stratification for patients with chest pain.
Second, we did not measure serial neutrophil count levels
and so could not evaluate the potential dynamic diagnos-
tic value. However, the association of the risk predictors
with the onset of chest pain was evaluated, and did not
prove to be of benefit over cardiac troponin T. Third,
together with the mature segmented neutrophils, the im-
mature forms (band cells) also increase in number during
the acute stress and may provide some valuable insight in
the setting of acute myocardial infarction. However, as
most common laboratories report the absolute neutrophil
count as a sum of segmented and band cells, we were
unable to comment on its association.
In conclusion, high white blood cell count, neutrophil
count, and neutrophil/lymphocyte ratio were associated
with increased risk for acute myocardial infarction in pa-
tients presenting to the ED with chest pain. However, their
diagnostic accuracy was significantly lower than that of
cardiac troponin T, alone and in combination. Furthermore,
even their short- and long-term predictive ability for death
or acute myocardial infarction was fairly poor in compari-
son with cardiac troponin T. Hence, our findings suggest
that white-blood cell count, neutrophil count, or neutrophil/
lymphocyte ratio may be used in conjunction with clinical
findings for diagnosis suggestive of acute myocardial in-
farction and for risk prediction in chest pain patients, when
cardiac troponin T is unavailable. We also recommend that
before incorporating similar nonspecific inflammatory
markers into clinical practice for specific diagnosis, they
541
should be first compared with the widely available and
preferably gold-standard diagnostic tools and predictive risk
models.
ACKNOWLEDGMENTS
We thank the patients who participated in the study, the staff
of the Emergency Department, and the laboratory techni-
cians for their most valuable efforts.
References
1. Thygesen K, Alpert JS, White HD, et al. Universal definition of
myocardial infarction. Circulation. 2007;116:2634-2653.
2. Jaffe AS, Ravkilde J, Roberts R, et al. It’s time for a change to a
troponin standard. Circulation. 2000;102:1216-1220.
3. Melanson SE, Morrow DA, Jarolim P. Earlier detection of myocardial
injury in a preliminary evaluation using a new troponin I assay with
improved sensitivity. Am J Clin Pathol. 2007;128:282-286.
4. Reichlin T, Hochholzer W, Stelzig C, et al. Incremental value of
copeptin for rapid rule out of acute myocardial infarction. J Am Coll
Cardiol. 2009;54:60-68.
5. Neuhold S, Huelsmann M, Strunk G, et al. Comparison of copeptin,
B-type natriuretic peptide, and amino-terminal pro-B-type natri-
uretic peptide in patients with chronic heart failure: prediction of
death at different stages of the disease. J Am Coll Cardiol. 2008;
52:266-272.
6. Katan M, Morgenthaler N, Widmer I, et al. Copeptin, a stable peptide
derived from the vasopressin precursor, correlates with the individual
stress level. Neuro Endocrinol Lett. 2008;29:341-346.
7. Furze RC, Rankin SM. Neutrophil mobilization and clearance in the
bone marrow. Immunology. 2008;125:281-288.
8. Merten M, Thiagarajan P. P-selectin expression on platelets deter-
mines size and stability of platelet aggregates. Circulation. 2000;102:
1931-1936.
9. Arai M, Lefer DJ, So T, DiPaula A, Aversano T, Becker LC. An
anti-CD18 antibody limits infarct size and preserves left ventricular
function in dogs with ischemia and 48-hour reperfusion. J Am Coll
Cardiol. 1996;27:1278-1285.
10. Marciniak A, Gierblinski I, Stefanski R, et al. Predictive value of
plasma interleukin 1, interleukin 6, interleukin 8 and C-reactive protein
(CRP) in patients with myocardial infarction [Polish]. Pol Arch Med
Wewn. 2003;109:15-22.
11. Ziakas A, Gavrilidis S, Giannoglou G, et al. In-hospital and long-term
prognostic value of fibrinogen, CRP, and IL-6 levels in patients with
acute myocardial infarction treated with thrombolysis. Angiology.
2006;57:283-293.
12. Tousoulis D, Antoniades C, Bosinakou E, et al. Differences in inflam-
matory and thrombotic markers between unstable angina and acute
myocardial infarction. Int J Cardiol. 2007;115:203-207.
13. Reichlin T, Hochholzer W, Bassetti S, et al. Early diagnosis of myo-
cardial infarction with sensitive cardiac troponin assays. N Engl J Med.
2009;361:858-867.
14. DeLong ER, DeLong DM, Clarke-Pearson DL. Comparing the areas
under two or more correlated receiver operating characteristic curves:
a nonparametric approach. Biometrics. 1988;44:837-845.
15. Zazula AD, Precoma-Neto D, Gomes AM, et al. An assessment of
neutrophils/lymphocytes ratio in patients suspected of acute coronary
syndrome. Arq Bras Cardiol. 2008;90:31-36.
16. Basili S, Di Francoi M, Rosa A, et al. Absolute neutrophil counts and
fibrinogen levels as an aid in the early diagnosis of acute myocardial
infarction. Acta Cardiol. 2004;59:135-140.
17. Papa A, Emdin M, Passino C, Michelassi C, Battaglia D, Cocci F.
Predictive value of elevated neutrophil-lymphocyte ratio on cardiac
mortality in patients with stable coronary artery disease. Clin Chim
Acta. 2008;395(1-2):27-31.
542
18. Chia S, Nagurney JT, Brown DF, et al. Association of leukocyte and
neutrophil counts with infarct size, left ventricular function and out-
comes after percutaneous coronary intervention for ST-elevation myo-
cardial infarction. Am J Cardiol. 2009;103:333-337.
19. O’Donoghue M, Morrow DA, Cannon CP, et al. Association be-
tween baseline neutrophil count, clopidogrel therapy, and clinical
and angiographic outcomes in patients with ST-elevation myocar-
dial infarction receiving fibrinolytic therapy. Eur Heart J. 2008;29:
984-991.
20. Takahashi T, Hiasa Y, Ohara Y, et al. Relationship of admission
neutrophil count to microvascular injury, left ventricular dilation, and
long-term outcome in patients treated with primary angioplasty for
acute myocardial infarction. Circ J. 2008;72:867-872.
21. Horne BD, Anderson JL, John JM, et al. Which white blood cell
subtypes predict increased cardiovascular risk? J Am Coll Cardiol.
2005;45:1638-1643.
22. Gurm HS, Bhatt DL, Lincoff AM, et al. Impact of preprocedural white
blood cell count on long term mortality after percutaneous coronary
The American Journal of Medicine, Vol 124, No 6, June 2011
intervention: insights from the EPIC, EPILOG, and EPISTENT trials.
Heart. 2003;89:1200-1204.
23. Cheng ML, Chen CM, Gu PW, Ho HY, Chiu DT. Elevated levels
of myeloperoxidase, white blood cell count and 3-chlorotyrosine in
Taiwanese patients with acute myocardial infarction. Clin Biochem.
2008;41(7-8):554-560.
24. Huang G, Zhong XN, Zhong B, et al. Significance of white blood cell
count and its subtypes in patients with acute coronary syndrome. Eur
J Clin Invest. 2009;39:348-358.
25. Soltani A, Moayyeri A. What constitutes clinical evidence? A dynamic
approach to clinical diagnosis. Can Fam Physician. 2005;51:1578-1579,
1582-1583.
26. Grimes DA, Schulz KF. Refining clinical diagnosis with likelihood
ratios. Lancet. 2005;365(9469):1500-1505.
27. Goswami B, Rajappa M, Mallika V, Shukla DK, Kumar S. TNF-alpha/
IL-10 ratio and C-reactive protein as markers of the inflammatory
response in CAD-prone North Indian patients with acute myocardial
infarction. Clin Chim Acta. 2009;408(1-2):14-18.