Validity of Fractional Exhaled Nitric Oxide in Diagnosis of Corticosteroid-Responsive Cough.

Validity of Fractional Exhaled Nitric Oxide

in Diagnosis of Corticosteroid-Responsive

Cough

Fang Yi,MD;Ruchong Chen,MD,PhD;Wei Luo,MSc;Danyuan Xu,MD;Lina Han,MD;Baojuan Liu,MD;Siqi Jiang,MD; Qiaoli Chen,BSc;and Kefang Lai,MD,PhD

BACKGROUND:Whether fractional exhaled nitric oxide(FeNO)measurement alone or

combined with sputum eosinophil and atopy is useful in predicting corticosteroid-responsive

cough(CRC)and non-CRC(NCRC)is not clear.

METHODS:A total of244patients with chronic cough and59healthy subjects as control were

enrolled.The causes of chronic cough were con?rmed according to a well-established

diagnostic algorithm.FeNO measurement and induced sputum for differential cell were

performed in all subjects.

RESULTS:CRC occurred in139(57.0%)patients and NCRC occurred in105.The FeNO level

in CRC signi?cantly correlated with sputum eosinophils(r s?0.583,P<.01).The median

(quarter)of FeNO level in CRC was signi?cantly higher than NCRC(32.0ppb[19.0-65.0

ppb]vs15.0ppb[11.0-22.0ppb],P<.01).FeNO of31.5ppb had a sensitivity and speci-

?city of54.0%and91.4%,respectively,in predicting CRC from chronic cough,with a positive

predictive value of89.3%and a negative predictive value of60.0%.If the patients had a

combination of low level of FeNO(<22.5ppb),normal sputum eosinophil(<2.5%),and

absence of atopy,the sensitivity and speci?city would be30.3%and93.5%for predicting

NCRC.

CONCLUSIONS:In our cohort,a high level($31.5ppb)of FeNO indicates more likelihood

of CRC,but the sensitivity is insuf?cient to rule out a diagnosis of CRC.A combination of

low-level FeNO,normal sputum eosinophil,and absence of atopy suggests a lower likelihood

of CRC.CHEST2016;149(4):1042-1051

KEY WORDS:atopy;chronic cough;corticosteroids-responsive cough;eosinophil;fractional

exhaled nitric oxide

ABBREVIATIONS:AC=atopic cough;CRC=corticosteroid-responsive cough;CVA=cough variant asthma;EB=eosinophilic bronchitis; FeNO=fractional exhaled nitric oxide;GERC=gastroesophageal re?ux-related cough;ICS=inhaled corticosteroids;NCRC= noncorticosteroid-responsive cough;NPV=negative predictive value; PLR=positive likelihood ratio;PPV=positive predictive value; UACS=upper airway cough syndrome

AFFILIATIONS:State Key Laboratory of Respiratory Disease, Guangzhou Institute of Respiratory Diseases,The First Af?liated Hospital of Guangzhou Medical University,Guangzhou,P.R.China. Drs Yi and Chen contributed equally to this manuscript.FUNDING/SUPPORT:This study was supported by National Natural Science Foundation of China[Grants81070019and81000033]. CORRESPONDENCE TO:Kefang Lai,MD,PhD,State Key Laboratory of Respiratory Disease,Guangzhou Institute of Respiratory Disease,the First Af?liated Hospital of Guangzhou Medical University,151 Yanjiang Rd,Guangzhou,510120,China;e-mail:klai@https://www.360docs.net/doc/d6124508.html, Copyrightó2016American College of Chest Physicians.Published by Elsevier Inc.All rights reserved.

DOI:https://www.360docs.net/doc/d6124508.html,/10.1016/j.chest.2016.01.006

[Original Research Signs and Symptoms of Chest Diseases]

The common causes of chronic cough include cough variant asthma(CVA),upper airway cough syndrome (UACS),eosinophilic bronchitis(EB),gastroesophageal re?ux-related cough(GERC),and atopic cough(AC).1-3 CVA,EB,and AC respond well to corticosteroids; therefore,they can be categorized as corticosteroid-responsive cough(CRC).CRC has accounted for a range of44%to65.5%of chronic cough in different studies.4-10 The diagnosis algorithm usually involves induced sputum test,spirometry,bronchial provocation test, 24-h esophageal pH monitoring,skin prick test,or other tests to identify the causes of chronic cough.1,2,11 However,in primary practice,most of these measurements are not available,and induced sputum test is time-consuming and labor dependent. Fractional exhaled nitric oxide(FeNO)is a new kind of biomarker that re?ects the eosinophilic in?ammation of the lower airway.As a noninvasive method,the FeNO test has been widely used in the evaluation of eosinophilic airway in?ammation and guiding the treatment of asthma.12-16American Thoracic Society guidelines have suggested that a FeNO level>50ppb indicates a higher likelihood of eosinophilic

in?ammation and response to corticosteroids.17 However,the optional cutoff level of FeNO in chronic cough remains unknown.The diagnostic value of FeNO in patients with chronic cough has only been evaluated in a few studies with small sample sizes,and an induced sputum test was not conducted in some studies.18-22 Chatkin et al18found that a FeNO level of30ppb could be the cutoff point to differentiate asthmatic cough from nonasthmatic cough.In Sato’s study,20there was no difference in FeNO level between EB and other causes of chronic cough.The higher FeNO level(>38.8ppb) suggested more likelihood of asthmatic chronic cough, but it could not be used to con?rm EB.On the contrary, Mi-Jung et al found that a FeNO level<31.7ppb suggested little likelihood of EB.21Those studies mentioned previously showed that FeNO measurement was inconsistent in identifying EB.The FeNO level

of patients with other causes of chronic cough have only been investigated by Maniscalco et al22recently. Because both eosinophilic in?ammation of the airway and atopic cough are responsive to corticosteroids and atopy is a feature of atopic cough,we speculated that FeNO measurement,combined with sputum eosinophils and atopy test,would be useful in screening CRC.

Therefore,this study is aimed to explore whether FeNO measurement alone or in combination

with sputum eosinophil and atopy is useful in diagnosing CRC.

Materials and Methods

Subjects

Patients with chronic cough who visited The First Af?liated Hospital of Guangzhou Medical University between December2008and April 2014were enrolled.The inclusion criteria of patients were:(1)age 16to70years,(2)chronic cough lasting more than8weeks and without abnormalities on chest radiograph,(3)no treatment with oral corticoids in the past4weeks,(4)never-smokers or ex-smokers with cessation of smoking for at least6months before the study, and(5)no upper airway infection in the past8weeks.Nonatopic, nonsmoking healthy volunteers were enrolled as control subjects. The study was approved by the Ethic Committee of the First Af?liated Hospital of Guangzhou Medical University(IRB200902). Written informed consent was obtained from all participants.

Study Design

A validated,systematic,step-by-step diagnostic algorithm1,4was used in this prospective study.First,a detailed medical history taking and physical examination were performed.The next step consisted of chest radiography,spirometry,bronchial provocation test,induced sputum,and FeNO measurement,which were standard investigations in all patients.Additional tests(eg,24-h esophageal pH monitoring,thoracic or sinus CT scan)would also be conducted as needed.The?nal diagnosis was con?rmed based on response to treatment.Diagnostic criteria and management of different causes could be found in our previous report.4CVA,EB,and AC were categorized as CRC;UACS,GERC,and others were categorized as non-CRC(NCRC).The?owchart of the study was showed in Figure1.Diagnostic Methods

Sputum was induced and processed as described by the Chinese Cough Guidelines.1Brie?y,after inhalation of400m g salbutamol,sputum was induced with3%saline for15min via an ultrasonic nebulizer.If an insuf?cient amount of sputum was collected,inhaling with4%and then5%hypertonic saline in sequence for7min was nebulized repeatedly.The cell smear was stained with hematoxylin and eosin for a differential cell count.The differential count was obtained by counting400nonsquamous cells.

Spirometry and bronchial provocation test were performed by Jaeger Master Screen(Germany)according to the American Thoracic Society Recommendation.23The FEV1%predicted,FVC%predicted, and FEV1/FVC ratio were recorded.A provocative dose of methacholine causing a20%fall in FEV1was adopted as the marker for airway hyperresponsiveness.Airway hyperresponsiveness was de?ned as a provocative dose of methacholine causing a20%fall in FEV1<12.8m mol.

Concentration of FeNO was evaluated using NIOX MINO(Aerocrine Company,Sweden).Measurements were performed in accordance with standard procedure described by the American Thoracic Society and the European Respiratory Journal.24Brie?y,subjects were informed to inhale deeply via a mouthpiece,then exhale with a constant?ow(0.05L/s)for10s.

Allergen skin sensitivity was measured by skin prick test with the common aeroallergens,controls consisted of normal saline and histamine,which was performed as described in our previous study.4

Atopy was de ?ned as the presence of positive skin reaction to any allergen or the increase of serum IgE.Statistical Analysis

Statistical analyses were conducted by using SPSS,version 18.0.The levels of FeNO and percentage of sputum eosinophil were expressed

as median and interquartile ranges.A Mann-Whitney test was applied to the comparison of FeNO from different causes.The optimal cutoff points of FeNO were obtained by receiver operating characteristic curve.The correlation between FeNO and sputum eosinophil was determined by Spearman rank correlation.P value <.05was considered statistically signi ?cant.

Results

Demographics

Among 244patients with chronic cough enrolled in our study,139(57.0%)received a diagnosis of CRC and 105received a diagnosis of NCRC.A total of 215quali ?ed sputum sample were obtained from those patients.Fifty-nine healthy subjects were

included in the control group.Clinical characteristics including of FEV 1%,FVC%,FEV 1/FVC,level of FeNO,and percentage of sputum eosinophil are shown in Table 1.

Values of FeNO

The FeNO level in CRC was signi ?cantly higher than in NCRC (32.0ppb [19.0-65.0ppb]vs 15.0ppb

[11.0-22.0ppb],P <.01;Fig 2A).For single cause,the FeNO level in CVA was higher than those in EB,AC,UACS,GERC,others,and the healthy control group (all P <.01).The FeNO level was signi ?cantly higher in EB compared with AC,UACS,GERC,others,and the healthy control group (all P <.01).However,there was no signi ?cant difference in FeNO value among groups of AC,UACS,GERC,others,and the healthy control (all P >.05)(Fig 2B).No signi ?cant differences of FeNO level between patients with or without allergic rhinitis were found in different groups (P >.05,respectively).

Sputum Eosinophil and Its Correlation With FeNO

The percentages of sputum eosinophil in both of CVA and EB group were signi ?cantly higher

than

Figure 1–Flowchart of the study.AC ?atopic cough;BPT ?bronchial provocation test;CRC ?corticosteroid-responsive cough;CVA ?cough variant asthma;EB ?eosinophilic bronchitis;FeNO ?fractional exhaled nitric oxide;GERC ?gastroesophageal re ?ux-related chronic cough;NCRC ?noncorticosteroid-responsive cough;UACS ?upper airway cough syndrome.

those in AC,UACS,GERC,others,and the healthy control subjects (all P <.01).The percentage of sputum eosinophil in CRC was signi ?cantly higher than that in NCRC (7.5%[3.0-27.9%]vs 0.5%[0.0-1.3%],P <.001).The eosinophil counts were similar in sputum among AC,UACS,GERC,others,and healthy subjects (all P >.05)(Fig 3).

In CVA,EB,CRC,and all patients with chronic cough,FeNO value and eosinophil percentage in induced sputum were signi ?cantly correlated (r s ?0.463,P <.01;r s ?0.494,P <.01;r s ?0.583,P <.01;r s ?0.597,P <.01)(Fig 4

A-D).

T A B L E 1

]D e m o g r a p h i c s o f 244P a t i e n t s W i t h C h r o n i c C o u g h a n d 59H e a l t h y S u b j e c t s

A g e ,h e i g h t ,w e i g h t ,F E V 1%p r e d ,F V C %p r e d ,a n d F E V 1/F V C r a t i o a r e e x p r e s s e d a s m e a n ?S D .F e N O a n d s p u t u m e o s i n o p h i l a r e e x p r e s s e d a s m e d i a n (i n t e r q u a r t i l e r a n g e ).A C ?a t o p i c c o u g h ;C V A ?c o u g h v a r i a n t a s t h m a ;E

B ?e o s i n o p h i l i c b r o n c h i t i s ;E o s %?p e r c e n t a g e o f s p u t u m e o s i n o p h i l s ;F e N O ?f r a c t i o n a l e x h a l e d n i t r i c o x i d e ;G E R

C ?g a s t r o e s o p h a g e a l r e ?u x -r e l a t e d c h r o n i c c o u g h ;p r e d ?p r e d i c t e d ;U A C S ?u p p e r a i r w a y c o u g h s y n d r o m e

.

F e N O (p p b )

30020010010075

5025CRC NCRC

P < .01

0F e N O (p p b )

300*

#

200100100755025CVA

UACS

EB

GERC AC

OTHERs

A

B

Figure 2–Exhaled NO values by diagnostic category.A,FeNO

value in CRC (n ?139)was signi ?cantly higher than that in NCRC (n ?105)(P <.01).B,FeNO values by diagnostic category.FeNO values in CVA and EB were signi ?cantly higher than those in AC,UACS,GERC,others,and the healthy control subjects (*P <.01and #P <.01respectively).FeNO values in CVA were signi ?cantly higher than those in EB (O P <.01).See the Figure 1legend for expansion of abbreviations.

Receiver Operating Characteristic Curve for Predicting Certain Cause According to FeNO

For distinguishing CVA from chronic cough,the optimal cutoff point for FeNO was33.5ppb with sensitivity of69.6%and speci?city of85.1%(Fig5A). The sensitivity and speci?city of FeNO for detecting EB from those without asthma were69.8%and76.2%at a cutoff point of22.5ppb(Fig5B).The sensitivity and speci?city for differentiating chronic cough present with eosinophilic in?ammation from those absent from eosinophilic in?ammation,using31.5ppb as the cutoff point,were60.7%and91.8%(Fig5C). With a cutoff point of31.5ppb,the sensitivity and speci?city was54.0%and91.4%for differentiating CRC from NCRC(Fig5D),and the positive predictive value(PPV)and negative predictive value(NPV), positive likelihood ratio(PLR),and negative likelihood ratio were89.3%and60.0%,6.3,and

0.5,respectively.The sensitivity and speci?city of FeNO for detecting sputum eosinophilia($2.5%)

were74.6%and77.2%at a cutoff point of22.5ppb (Fig5E).If the speci?city of FeNO for detecting sputum eosinophilia($2.5%)was adjusted to95%, the cutoff value of FeNO would increase to33.5ppb. An elevated sputum eosinophil($2.5%)alone had

a speci?city and PPV of90.4%and92.9%for predicting CRC.

All sensitivities,speci?cities,PPVs,NPVs,PLRs and negative likelihood ratios of differentiating different group according to the FeNO are shown in Table2.Predictive Value for CRC by Combining FeNO, Sputum Eosinophil,and Atopy

Ninety-?ve patients with chronic cough had available atopy data.The clinical characteristics,FeNO level, and percentage of sputum eosinophil are shown in Table3.No signi?cant differences in FeNO level between patients with or without atopy were found

in different causes or all patients with chronic cough (P>.05,respectively).The speci?city and PPV would be64.5%and60.6%when atopy alone was used as the diagnostic criteria of CRC.If the patients had any one of the three characteristics(FeNO>31.5ppb, sputum eosinophil$2.5%,atopy),the sensitivity and speci?city would be93.5%and45.5%for diagnosis of CRC.If the patients had any one of the two characteristics(FeNO>31.5ppb with atopy,sputum eosinophil$2.5%with atopy)the sensitivity,speci?city, PPV,and NPV would be50.0%,93.9%,93.9%,and 50.0%for diagnosis of CRC,respectively.If the patients had a combination of a low FeNO level

(<31.5ppb),normal sputum eosinophil(<2.5%), and absence of atopy,the sensitivity,speci?city,PPV, and NPV would be48.4%,91.9%,76.2%,and77.0%, respectively,for predicting NCRC.If the FeNO level was adjusted to<22.5ppb,the sensitivity,speci?city, PPV,and NPV would be30.3%,93.5%,71.4%,and 71.6%,respectively.All the sensitivities,speci?cities, PPVs,and NPVs for predicting CRC according to

the FeNO,sputum eosinophil,and atopy are shown

in Table4.

Discussion

To the best of our knowledge,this study enrolled the largest sample of patients with chronic cough and was the ?rst study to evaluate the value of FeNO alone,combining sputum eosinophils and atopy in predicting CRC.

In this study,the prevalence of CRC in chronic cough was nearly60%,which was in agreement with our previous study.4Therefore,it would be of great importance if FeNO measurement could be used as

a diagnostic tool in predicting CRC in practice. Unfortunately,this present study showed a low sensitivity(54.0%)of FeNO alone in differentiating CRC from NCRC,meaning more than40%of patients with CRC would be misdiagnosed.However,the speci?city,PPV,and PLR were quite high when using 31.5pp

b as the cutoff point of the FeNO level,which indicates that corticosteroids could be consider as

an empirical treatment option for the chronic cough patients with an increased level of FeNO($31.5

ppb). E

o

s

%

*#

CVA UACS

EB GERC

AC OTHERs

100

75

50

25

20

15

10

5

Figure3–Percentages of sputum eosinophils by diagnostic category.

Percentages of sputum eosinophils in CVA or EB were signi?cantly

higher than those in AC,UACS,GERC,and others(*P<.01and

#P<.01,respectively).No signi?cant difference was shown in

percentage of sputum eosinophil among AC,UACS,GERC,and

others(all P>.05,respectively).See the Figure1legend for

expansion of abbreviations.

The value of FeNO for predicting the response to inhaled corticosteroids (ICS)varied in different

studies.6,7,25,26When using 20ppb as the cutoff point,the sensitivity and speci ?city were only 53%and 63%in Prieto ’s study,showing that patients with CRC could not be identi ?ed by FeNO measurement.6Hahn et al showed that,using 38ppb as the cutoff point,FeNO could predict the response to ICS therapy for patients with chronic cough.25However,in Hahn ’s study,there was a high proportion (48%)of people with asthma,which might have increased the response rate to the ICS.Another retrospective study also showed a high sensitivity (94.7%)for predicting the response to ICS when using 33.9ppb FeNO level as the cutoff point.7However,elevated FeNO,instead of an induced sputum

test,was used as the diagnostic criteria for EB.7Another retrospective study showed that the sensitivity and speci ?city of FeNO for predicting the response to ICS were 78.6%and 80.0%when using the 26.5ppb cutoff point in patients with chronic cough,26but

some patients with postinfectious cough were included and there was no induced sputum test in that study.We also combined FeNO,sputum eosinophil,and atopy to evaluate the predictive value for CRC.Our results showed that if the patients had a low level of FeNO (<22.5ppb)and normal sputum eosinophil and absence of atopy,the sensitivity was low but the

speci ?city was high (93.5%)for NCRC,which indicated that a low level of FeNO,normal sputum eosinophil,

A

300.00r s = 0.463P < .01

250.00200.00150.00100.00

50.000.00

0.00

20.00

40.0060.00Eos% of CVA

F e N O

80.00

100.00

C

300.00r s = 0.583P < .01

250.00200.00150.00100.0050.000.00

0.00

20.00

40.0060.00Eos% of CRC

F e N O

80.00

100.00

D

300.00r s = 0.597P < .01

250.00200.00150.00100.0050.000.00

0.00

20.0040.0060.00

Eos% of patients with chronic cough

F e N O 80.00100.00B

200.00

r s = 0.494P < .01

150.00

100.00

50.00

0.00

0.00

20.00

40.00

Eos% of EB

F e N O

60.00

Figure 4–Correlation of FeNO and percentage of sputum eosinophils (Eos%)in patients with CVA,EB,CRC,and all patients with chronic cough.A,Correlation between FeNO and Eos%was found in 63patients with CVA (r ?0.463,P <.01).B,Correlation between FeNO and

Eos%was found in 53patients with EB (r ?0.494,P <.01).C,Correlation between FeNO and Eos%was found in 132patients with CRC (r ?0.583,P <.01).D,Correlation between FeNO and Eos%was found in 215patients with chronic cough (r ?0.597,P <.01).See the Figure 1legend for expansion of abbreviations.

1.0A

0.80.6S e n s i t i v i t y

0.40.20.0

0.0

0.2

0.40.61-specificity

0.8

1.0

B

0.41-specificity

1.00.80.6S e n s i t i v i t y

0.40.20.0

0.0

0.2

0.60.8

1.0

0.0

0.0

0.2

0.40.61-specificity

0.8

1.0

D

1.00.80.6S e n s i t i v i t y

0.40.2C

1.00.80.6S e n s i t i v i t y

0.40.20.0

0.0

0.2

0.40.61-specificity

0.8

1.0

E

1.00.80.6S e n s i t i v i t y

0.40.20.0

0.0

0.2

0.40.61-specificity

0.8 1.0

Figure 5–Diagnostic value of FeNO in different categories.A,The area under the ROC curve was 0.791and the optimal cutoff level of FeNO was 33.5ppb,with sensitivity of 69.6%and speci ?city of 85.1%,for distinguishing CVA from EB,AC,UACS,GERC,and others.B,The area

under the ROC curve was 0.787and the optimal cutoff level of FeNO was 22.5ppb,with sensitivity of 69.8%and speci ?city of 76.2%,for distinguishing EB from AC,UACS,GERC,and others.C,The area under the ROC curve was 0.822and the optimal cutoff level of FeNO was 31.5ppb,with sensitivity of 60.7%and speci ?city of 91.8%,for distinguishing chronic cough related to eosinophilic airway in ?ammation from those not related to eosinophilic.D,The area under the ROC curve was 0.784and the optimal cutoff level of FeNO was 31.5ppb,with sensitivity of 54%and speci ?city of 91.4%,for distinguishing CRC from NCRC.E,The area under the ROC curve was 0.807and the optimal cutoff level of FeNO was 22.5ppb,with sensitivity of 74.6%and speci ?city of 77.2%,for distinguishing sputum eosinophilia.ROC ?receiver operating characteristic.See the Figure 1legend for expansion of other abbreviations.

and absence of atopy was less likely in CRC.If the patients had any one of the two characteristics (FeNO $31.5ppb with atopy,sputum eosinophil $2.5%with atopy),a high speci ?city and PPV (93.9%and 93.9%,respectively)suggested a high likelihood of CRC.We also found that sputum

eosinophilia had a speci ?city and PPV of 90.4%and 92.9%,respectively,for predicting CRC,which was similar to FeNO.However,if atopy alone was used as the diagnostic criteria of CRC,both the sensitivity (64.5%)and speci ?city (60.6%)were insuf ?cient for diagnosing CRC.

Our study showed a low sensitivity and speci ?city of FeNO measurement for distinguishing EB from nonasthmatic chronic cough.However,Oh et al reported that a low level of FeNO allowed an

exclusionary diagnosis of EB.21A much higher level of FeNO (median,51.4ppb)in their study may explain the difference compared with our study

(median,30.0ppb).For detecting CVA from chronic cough,we found a high speci ?city and NPV when

using 33.5ppb as the cutoff point.This suggested that FeNO measurement is useful in the exclusion of CVA.We also found that a FeNO level $31.5ppb in chronic cough patients had a high speci ?city,PPV,and a helpful PLR for determining the presence of eosinophilic airway in ?ammation (CVA and EB),although the sensitivity was low.In Maniscalco ’study,an ideal cutoff point that could discriminate CVA and nonasthmatic EB from GERC and UACS was 33.0ppb,with sensitivity and speci ?city of 92%and 88%22;the cutoff point was similar to ours,but a high sensitivity was observed in their research.The reasons we believe this follow.First,Maniscalco ’s study only recruited for the common causes of chronic cough,whereas our study also recruited for the uncommon causes.Second,the sample sizes of two studies were different.In classic asthma,FENO >50ppb can be used to identify eosinophilic in ?ammation and responsiveness to corticosteroids.17Therefore,the cutoff value to indicate the eosinophilic in ?ammation is different between classic asthma (>50ppb)and chronic cough.

There is no doubt that induced sputum analysis is an important tool for evaluating airway in ?ammation in patients with chronic cough or asthma.1,27,28However,not all the subjects can provide an suitable sample of sputum for measurements.In our study,the successful rate of sputum induction was 88.1%,which was similar to our previous reported and other studies.16,29,30We found that the FeNO value and sputum eosinophils were signi ?cantly correlated in patients with CVA or EB or CRC,but the correlations were not very high as reported by other authors.21,31Our study showed that one-third of the patients with eosinophilic in ?ammation of the airway had a normal FeNO value.It suggests that FeNO measurement may not be a replacement for sputum induction when monitoring airway in ?ammation in patients with chronic cough,but may act as

a

TABLE 2

]Diagnostic Value of FeNO on Different Categories

CRC ?corticosteroid-responsive couth;NLR ?negative likelihood ratio;NPV ?negative predictive value;PLR ?positive likelihood ratio;PPV ?positive predictive value.See Table 1legend for expansion of other abbreviations.a

Cough related to airway eosinophilic in ?ammation includes CVA and EB.b

Includes CVA,EB,and

AC.

TABLE 3

]Demographics of 95Patients With Available

Atopy Data

Age expressed as mean ?SD.FeNO and Eos%expressed as median (interquartile range).NCRC ?noncorticosteroid-responsive cough.See Table 1and 2legends for expansion of other abbreviations.a

Includes CVA,EB,and AC.b

Includes UACS,GERC,and others.

complementary method.If we adjusted the speci ?city of FeNO for detecting sputum eosinophilia ($2.5%)to 95%,the cutoff value of FeNO would be 33.5ppb,which means that sputum eosinophilia was very likely when the FeNO level was >33.5ppb.These patients may not need sputum assessment.

AC is one of common causes of chronic cough in Asian clinical practices.4Our study also showed a normal FeNO level in AC,which was consistent with the level of sputum eosinophils.Two studies from Japan also reported that the FeNO level was normal in AC.19,32In the Japanese guidelines,3sputum eosinophil counts can be normal or increased in diagnostic criteria of AC,which suggests that AC and EB may sometimes overlap.In China,however,the de ?nition of AC includes normal eosinophil counts in sputum,and EB is excluded.There were some limitations to this study.First,not all the patients had available atopy data.Although the

sample was not large,we did not ?nd any signi ?cant differences in FeNO between patients with or without atopy in CVA,EB,and CRC.We think that the

atopic status does not appear very useful in screening for CRC.Further studies with a larger sample should be focus on this issue.Second,evaluating the level of FeNO and other in ?ammation biomarkers in larger sample with AC is needed.

In conclusion,our study ?nds that FeNO levels are higher in patients with CVA,EB,and CRC than other causes of chronic cough and correlate with sputum eosinophils.A combination of a low level of FeNO (<22.5ppb)and normal sputum eosinophil (<2.5%)and absence of atopy suggests less likelihood of CRC.FeNO level $31.5ppb had a high speci ?city,PPV,and a helpful PLR,which suggests more likelihood of CRC,but the sensitivity is insuf ?cient to rule out the diagnosis of

CRC.

TABLE 4

]Predictive Value for CRC on Different Categories by Combining FeNO,Eos%,and Atopy

See Table 1and 2legends for expansion of abbreviations.

Acknowledgments

Author contributions:F.Y.and R.C. contributed to data collection,data analysis, manuscript preparation and interpretation, revising of the submitted manuscript,and read and approved the?nal manuscript. W.L.was responsible for sputum induction and differential cell count,read and approved the?nal manuscript.D.X.,L.H.,B.L.,and S. J.contributed to data collection and read and approved the?nal manuscript.Q.C. was responsible for sputum induction and differential cell count,and read and approved the?nal manuscript.K.L. contributed to concept visualization of the study,study design,and critical review of the manuscript and read and approved the?nal manuscript.

Other contributions:We thank Steven G. Smith,PhD(Department of Medicine, McMaster University,Canada),and Patrick Mitchell,MD(Royal College of Physicians, Ireland),for comments and revision.We also thank Mei Jiang,PhD(State Key Laboratory of Respiratory Diseas,Guangzhou Institute of Respiratory Diseases,The First Af?liated Hospital of Guangzhou Medical University, China),for comments and data analysis. Financial/non?nancial disclosures:None declared.

References

1.Asthma Workgroup,Chinese Society,

Respiratory,Diseases,Chinese Medical,

Association.The Chinese national

guidelines on diagnosis and management

of cough(December2010).Chin Med J

(Engl).2011;124(20):3207-3219.

2.Irwin RS,Baumann MH,Bolser DC,et al.

Diagnosis and management of cough

executive summary:ACCP evidence-

based clinical practice guidelines.Chest.

2006;129(1Suppl):1-23.

3.Kohno S,Ishida T,Uchida Y,et al.The

Japanese Respiratory Society guidelines

for management of cough.Respirology.

2006;1(Suppl4):S135-S186.

https://www.360docs.net/doc/d6124508.html,i KF,Chen RC,Lin JT,et al.

A prospective,multicenter survey on

causes of chronic cough in China.Chest.

2013;143(3):613-620.

5.Fujimura M,Abo M,Ogawa H,et al.

Importance of atopic cough,cough variant

asthma and sinobronchial syndrome as

causes of chronic cough in the Hokuriku area of Japan.Respirology.2005;10(2):201-207.

6.Prieto L,Ferrer A,Ponce S,et al.Exhaled

nitric oxide measurement is not useful for predicting the response to inhaled

corticosteroids in subjects with chronic

cough.Chest.2009;136(3):816-822.

7.Hsu J-Y,Wang C-Y,Cheng Y-W,et al.

Optimal value of fractional exhaled nitric

oxide in inhaled corticosteroid treatment

for patients with chronic cough of

unknown cause.J Chin Med Assoc.

2013;76(1):15-19.

8.Smyrnios NA,Irwin RS,Curley FJ.

Chronic cough with a history of excessive

sputum production:the spectrum and

frequency of causes,key components of

the diagnostic evaluation,and outcome of

speci?c therapy.Chest.1995;108(4):

991-997.

9.Poe RH,Harder RV,Israel RH,et al.

Chronic persistent cough.Experience in

diagnosis and outcome using an anatomic

diagnostic protocol.Chest.1989;95(4):

723-728.

10.Irwin RS,Corrao WM,Pratter MR.

Chronic Persistent cough in the adult:the

spectrum and frequency of causes and

successful outcome of speci?c therapy.

Am Rev Respir Dis.1981;123(4Pt1):

413-417.

11.Morice A.The diagnosis and management

of chronic cough.Eur Respir J.2004;24(3):

481-492.

12.Malmberg L,Pelkonen A,Haahtela T,

et al.Exhaled nitric oxide rather than lung

function distinguishes preschool children

with probable asthma.Thorax.2003;58(6):

494-499.

13.Dupont LJ,Demedts MG,Verleden GM.

Prospective evaluation of the validity of

exhaled nitric oxide for the diagnosis of

asthma.Chest.2003;123(3):751-756.

14.Pedrosa M,Cancelliere N,Barranco P,

et https://www.360docs.net/doc/d6124508.html,efulness of exhaled nitric oxide for

diagnosing asthma.J Asthma.2010;47(7):

817-821.

15.Jones SL,Kittelson J,Cowan JO,et al.The

predictive value of exhaled nitric oxide

measurements in assessing changes in

asthma control.Am J RespirCrit Care

Med.2001;164(5):738-743.

16.Jatakanon A,Lim S,Kharitonov SA,et al.

Correlation between exhaled nitric oxide,

sputum eosinophils,and methacholine

responsiveness in patients with mild

asthma.Thorax.1998;53(2):91-95.

17.Dweik RA,Boggs PB,Erzurum SC,et al.

An of?cial ATS clinical practice guideline:

interpretation of exhaled nitric oxide

levels(FENO)for clinical applications.

Am J RespirCrit Care Med.2011;184(5):

602-615.

18.Chatkin JM,Ansarin K,Silkoff PE,et al.

Exhaled nitric oxide as a noninvasive

assessment of chronic cough.Am J Respir

Crit Care Med.1999;159(6):1810-1813.

19.Fujimura M,Ohkura N,Abo M,et al.

Exhaled nitric oxide levels in patients with

atopic cough and cough variant asthma.

Respirology.2008;13(3):359-364.

20.Sato S,Saito J,Sato Y,et al.Clinical

usefulness of fractional exhaled nitric

oxide for diagnosing prolonged cough.

Respir Med.2008;102(10):1452-1459.

21.Oh MJ,Lee JY,Lee BJ,et al.Exhaled nitric

oxide measurement is useful for the

exclusion of nonasthmaticeosinophilic

bronchitis in patients with chronic cough.

Chest.2008;134(5):990-995.

22.Maniscalco M,Faraone S,So?a M,et al.

Extended analysis of exhaled and nasal

nitric oxide for the evaluation of chronic

cough.Respir Med.2015;109(8):970-974.

https://www.360docs.net/doc/d6124508.html,ler MR,Hankinson J,Brusasco V,et al.

Standardisation of spirometry.Eur Respir J.

2005;26(2):319-338.

24.ATS/ERS recommendations for

standardized procedures for the online

and of?ine measurement of exhaled lower

respiratory nitric oxide and nasal nitric

oxide,2005.Am J Respir Crit Care Med.

2005;171(8):912-930.

25.Hahn PY,Morgenthaler TI,Lim https://www.360docs.net/doc/d6124508.html,e

of exhaled nitric oxide in predicting

response to inhaled corticosteroids for

chronic cough.Mayo Clin Proc.

2007;82(11):1350-1355.

26.Watanabe K,Shinkai M,Shinoda M,et al.

Measurement of eNO with portable

analyser might improve the management

of persistent cough at primary care

practice in Japan.Clin Respir J.2014;13.

27.Fortuna AM,Feixas T,González M,et al.

Diagnostic utility of in?ammatory

biomarkers in asthma:exhaled nitric oxide

and induced sputum eosinophil count.

Respir Med.2007;101(11):2416-2421.

28.Berlyne GS,Parameswaran K,Kamada D,

et al.A comparison of exhaled nitric oxide

and induced sputum as markers of airway

in?ammation.J Allergy Clin Immunol.

2000;106(4):638-644.

29.Modena BD,Tedrow JR,Milosevic J,et al.

Gene expression in relation to exhaled

nitric oxide identi?es novel asthma

phenotypes with unique biomolecular

pathways.Am J RespirCrit Care Med.

2014;190(12):1363-1372.

30.Luo W,Chen RC,Liu CL,et al.Diagnostic

signi?cance of differential cell count in

induced sputum to chronic cough.Chin

J Lab Med(Chin).2007;30(3):280-283.

31.Kowal K,Bodzenta-Lukaszyk A,

Zukowski S.Exhaled nitric oxide in

evaluation of young adults with chronic

cough.J Asthma.2009;469(7):692-698.

32.Nakade Y,Fujimura M,Ohkura N,et al.

Reversibility of the pulmonary function

based on the partial?ow-volume

curve predicts the ef?cacy of

bronchodilator therapy for treating

chronic cough.Intern Med.2013;52(18):

2017-2023.