- Email: [email protected]
Validation of the Exhaled Breath Temperature Measure
Accepted Manuscript Validation of the exhaled breath temperature measure: reference values in healthy subjects Giovanna E. Carpagnano, Maria P. Foschino-Barbaro, Corrado Crocetta, Donato Lacedonia, Valerio Saliani, Luigi Davide Zoppo, Peter J. Barnes PII:
S0012-3692(16)62394-2
DOI:
10.1016/j.chest.2016.11.013
Reference:
CHEST 819
To appear in:
CHEST
Received Date: 2 August 2016 Revised Date:
29 September 2016
Accepted Date: 2 November 2016
Please cite this article as: Carpagnano GE, Foschino-Barbaro MP, Crocetta C, Lacedonia D, Saliani V, Zoppo LD, Barnes PJ, Validation of the exhaled breath temperature measure: reference values in healthy subjects, CHEST (2016), doi: 10.1016/j.chest.2016.11.013. 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 Validation of the exhaled breath temperature measure: reference values in healthy subjects
Giovanna E Carpagnano, Maria P Foschino-Barbaro, 1Corrado Crocetta, Donato Lacedonia, Valerio
RI PT
Saliani, Luigi Davide Zoppo, 2Peter J Barnes
Department of Medical and Surgical Sciences, Institute of Respiratory Diseases, University of Foggia – Italy; 1Section of Statistics, Faculty of Economics, University of Foggia ; 2Airway Disease
SC
Section, National Heart and Lung Institute, Imperial College London, Royal Brompton Hospital,
Address correspondence Dr Giovanna Elisiana Carpagnano Institute of Respiratory Diseases,
M AN U
London, UK.
University of Foggia, Italy
TE D
Department of Medical and Surgical Sciences,
EP
Email: [email protected]
AC C
Running title: EBT in healthy
Key words: exhaled breath temperature, validation, non-invasive methods, airways inflammation, reference value, correction factor
1
ACCEPTED MANUSCRIPT Abstract Background: Exhaled breath temperature (EBT) is a new non-invasive method for the study of inflammatory respiratory diseases with a potential to reach clinical practice. However, few studies,
and the range of normal values is not well established.
RI PT
and mainly derived from small, paediatric populations, are available on the validation of the method
The aim of this study was to measure EBT values in an Italian population of 298 (45.2±15.5 yrs, 143 male, FEV1 97.2±5.8%, FVC 98.4±3.9%) selected from 867 adult volunteers, in order to define
SC
reference values in healthy subjects and to analyse the influence of individual and external variables on this parameter.
M AN U
Methods: We measured EBT with a X-halo PRO device to different ambient temperature that ranged from 0 to 38°C
Findings: We report reference value of EBT in healthy Caucasian never smoker subjects. EBT values are strongly influenced by the external temperature and to a lesser extent by gender.
TE D
Interpretation: These data provide reference values in a large population of healthy never smokers subjects for measuring EBT as a basis for future studies. Our results are a contribution to the promotion of the EBT from bench to bed side.
AC C
EP
Fundings: No funding was provided for this review.
2
ACCEPTED MANUSCRIPT Introduction: The exhaled breath temperature (EBT) is a new non-invasive method for the analysis of airway inflammation. Since calor (heat) is listed as one of the cardinal signs of inflammation, the measure of the EBT has been developed as a marker of airway inflammation and therefore used in the study
[4], and lung cancer [5] and cystic fibrosis [6].
RI PT
of inflammatory respiratory diseases. Several studies have been performed in asthma [1-3], COPD
A new third generation thermometer, the X-halo (Delmedica), is now commercially available for
SC
the measure of the EBT. The smaller version of X-halo is now being developed for personal daily use as an “inflammometer” [7]. The use of EBT devices is particularly attractive in asthmatic
M AN U
patients that largely showed significantly higher EBT values compared to healthy subjects and that therefore are encouraged to use these exhaled thermometers in clinical practice for the maintenance of the asthma control and in the therapeutic management.
There are several potential advantages of EBT measurement with the X-halo device. The
TE D
measurement is very easy for patients and requires only few minutes, it is completely non-invasive and is therefore also suitable for children and patients with severe diseases. The device is well accepted by patients and ethics committee, it is inexpensive and finally it does not affect the
EP
underlying airway disease. Despite the potential of EBT, it still has not yet reached the clinical setting, partly because of a lack of a standardization and validation of the method.
AC C
Popov et al. evaluated the influence of room temperature, atmospheric pressure and relative humidity on EBT measurements with X-halo in a group of healthy subjects and asthmatic patients [7]. Furthermore they demonstrated the reproducibility of EBT measurements [7]. Vermeulen S et al. confirmed the reproducibility of EBT in a group of 124 healthy and asthmatic children [8]. Some important data are available on standardization and validation of EBT measurements that are needed to advance the clinical applicability of EBT measurements as a potential non-invasive marker of airway inflammation [9].
3
ACCEPTED MANUSCRIPT In order to validate EBT measurements it is important to follow the validation used for other noninvasive methods, such as induced sputum and exhaled nitric oxide that have now earned a place in the routine management of asthma and COPD and are now included in guidelines [2]. Data from previous studies on thousands of subjects have shown that EBT is a safe and entirely
RI PT
non-invasive method. No contraindications exist to its measurement and no adverse events have ever been reported. As for standardization of EBT measurements, the Delmedica manual for users has some recommendations for its use and for the cleaning the device (http://www.x-halo.com).
SC
For validation of EBT measurements, several factors need to be addressed: 1) the reproducibility of the measurement, 2) comparison with other biomarkers, 3) differences between measurements in
M AN U
diseased and healthy subjects, 4) sensitivity to risk factors or treatment and 5) reference normal values. Several studies reported good reproducibility of EBT measurements [7,8]. 2) Some comparison studies have also reported between EBT, exhaled nitric oxide [8] and induced sputum [8,10]. 3) Differences in EBT measurements in healthy subjects and patients with asthma [1-3],
TE D
COPD [4], cystic fibrosis (CF) [6], and lung cancer [5] have been reported. However, these studies have included only small numbers of subjects a. Reference values in the general population are therefore not well defined. 4) Also there are few studies looking at how the measurement may be
EP
changed physiologically with exercise and ventilation or with inhaled therapy [11,12]. This study was designed to measure EBT in a large population of healthy never-smoking adults in
AC C
order to provide reference values for the healthy population. Furthermore we aimed to analyse the influence of external variables on EBT measurements.
4
ACCEPTED MANUSCRIPT Material and Methods
Population We enrolled 867 consecutive Caucasian Italian volunteers (50.8±16.3 yrs, 60% male, BMI
RI PT
27.11±5.2 kg/m2, 59% smokers, FEV1 89.3±8.4%, FVC 90.7±4.8%, 90% living in industrialized areas while 10% in rural) during public regional meetings (“Breath Days” – Puglia - Italy) in shopping centres. The study was approved by the institutional ethics committee, University of
SC
Foggia (IRB approval number 17/CE/2014). All subjects were informed of the purpose of breath temperature measurement and after signing informed consent form, at the time of recruitment,
M AN U
anthropometric, physiological and clinical data were collected, including date of birth, height, weight, gender, area of residence, work, hobbies, diagnosed diseases, drugs, smoking history, blood pressure, blood oxygen saturation, lung function (with Pony FX- Cosmed), axillary body temperature and EBT measurement. We excluded from the population enrolled, all subjects with
TE D
any respiratory, cardiac or others diseases and all smokers and former smokers to obtain a new population of 298 healthy never smoked subjects (Table 1).
EP
Exhaled breath temperature measurement
EBT was measured with an X-halo device (Delmedica Investments, Singapore) according to
AC C
previously validated methods [13,14]. Briefly, patients were requested to inhale freely through the nose and to exhale into the device at a rate and depth typical of their normal tidal-breathing rhythm.
Ambient temperature measurement In consideration that EBT measurement were done in three different places (outside, hospital, shopping center) and that only when in hospital the temperature is usually controlled (between 19 and 24°C) we decided to measure and register ambient temperature before each EBT measurement. We used an external thermometer (PCE-HT 110, Italy) with a range of measure among 0-50 °C, 5
ACCEPTED MANUSCRIPT with precision variability of ±0.8°C, resolution: 0.1°C . We measured EBT to different ambient temperature that ranged from 0 to 38°C.
Statistical analysis
RI PT
To find the variables that have influenced EBT measurements we used a classification tree called ECHAID, that is an automatic classification algorithm that imposes a tree-like structure of the data. At each split, the algorithm looks for the predictor variable that if split, most "explains" the category
SC
response variable. In order to decide whether to create a particular split based on this variable, the CHAID algorithm tests a hypothesis regarding dependence between the split variable and the
M AN U
categorical response (using the chi-squared test for independence). Using a pre-specified significance level, if the test shows that the split variable and the response are independent, the algorithm stops the tree growth. Otherwise the split is created, and the next best split is searched. Using the result obtained by the classification tree we estimated a regression model to study the
AC C
EP
TE D
relationship between external temperature and EBT. The level of significance was set at p <0.05.
6
ACCEPTED MANUSCRIPT Results
Population 867 subjects volunteers were enrolled consecutively and successively divided on the basis of
RI PT
presence or absence of respiratory or other diseases and smoking habit to obtain 298 healthy never smoked subjects. Anthropometric, functional and clinical data of healthy subjects enrolled were
SC
summarized in Table 1.
EBT in healthy non smoker subjects
M AN U
E-CHAID tree estimated using the 298 healthy subjects, found that reference value of the EBT in healthy Caucasian non-smoking subjects was 30.459±2.955°C (Fig. 1). External temperature is the main factor influencing the EBT (F=58.991; df1=2, df2=295; p<0.00001). There are three groups of external temperatures that differently influence EBT.
TE D
The first group considers the cases with external temperature below 23°C. In this case the average EBT is 28.268±2.872°C. The second group considers cases measured with an external temperature of 23-28°C. In this case the EBT is 30.949±2.511°C. The third group shows that if the test is
EP
performed with an external temperature greater then 28°C the EBT is 32.558±1.805°C. The third level of the tree shows that gender influences the EBT for the first two groups. The female
AC C
EBT is respectively 1.486°C and 0.915°C lower than the male EBT. Other factors analysed as the age, the BMI, the blood pressure, internal temperature, lung function and area of residence influenced but without reaching statistical significance in the EBT. According to the result of the segmentation tree we estimated the following regression model:
EBT= α+β external temperature + ε
with α= 26.40 and β= 0.19, with E[ε]=0 and VAR [ε]=1whit R2= 0.2, F=175.4. 7
ACCEPTED MANUSCRIPT The model shows that the EBT is influenced by the external temperature. By average when the external temperature grows of 1°C the EBT grows of 0.19°C. For this reason it is important to measure the external temperature and if necessary to apply a correction factor to the results obtained. Using this model, the reference value of EBT in healthy non smoker subjects is 30.66°C
RI PT
at 22°C of external temperature that is common in hospital rooms. At 24°C the value is 31.4°C; at 26°C 31.40°C and at 28°C is 31.8°C.
AC C
EP
TE D
M AN U
SC
No adverse events were reported during or after the EBT measurement.
8
ACCEPTED MANUSCRIPT Discussion: In this study we report reference values for EBT in 298 healthy non-smoking subjects. We found that the EBT is strongly influenced by the external temperature and less by gender. Almost 200 articles are currently available in the literature on the use of EBT in asthma and other
RI PT
respiratory diseases. However, few of these studies have looked at the of EBT measurements in healthy subjects, which is needed if this non-invasive technique is to be useful in the clinic or measuring airway inflammation. There are no studies providing reliable reference values of EBT in
SC
healthy subjects. For this reason, the main aim of our study was to measure the EBT in a large normal population to define the reference EBT values in healthy never smoking subjects. We
M AN U
enrolled 867 apparently healthy volunteers who agreed to participate in this study. From this population we excluded all subjects with respiratory, cardiac or other diseases and all smokers or ex smokers giving a selected population of 298 healthy subjects that we used for the analysis. From this large healthy population, that is a great strength of this work, we report a reference value
could signify inflammation.
TE D
for the EBT that researchers and clinicians might use to identify subjects with increased EBT that
A second aim in this validation study was to explore the influence of ambient temperature and
EP
anthropometric, physiological and clinical variables (age, gender, height, weight, blood pressure, axillary temperature, lung function) on the EBT value. Popov et al. first addressed some of these
AC C
issues but in small numbers of patients [7]. In contrast to Popov et al, we found on a large population of 298 healthy subjects, that external (ambient) temperature strongly influences EBT values. In our daily practice, we noticed that the EBT, when measured in ambient temperatures up to 28°C was significantly higher and therefore we decided to perform our measurements at different ambient temperatures from 0°C to 38°C. The study was performed not only in hospital wards but also outside during public exhibitions and events in shopping centres. As expected, the ambient temperature strongly influences the EBT especially below 20 °C and over 27°C. To have tested EBT to the high external temperature found 9
ACCEPTED MANUSCRIPT in Southern Italy could be the reason for our discordant data compared to those of Popov et al. study. Usually the EBT measurement is performed in hospital where the temperature is usually controlled to be between 19 and 24°C. Also the Delmedica instruction for using X-halo suggest measuring EBT between 19 and 24°C assuming no influence of the external temperature on EBT
RI PT
values. However, there are wards where the temperature is not automatically adjusted where wider variations in ambient air temperature are frequent. Furthermore, if EBT is measured at home there may be greater temperature variations. While the measurement of EBT in controlled ambient
SC
temperature is the best option, we report a correction factor for ambient temperature that can normalize, in any situation, the EBT measurement. This correction factor calculated with a complex
M AN U
statistical method is really very simple to be applied and in its simplest form operators might increase the EBT value measured of 0.19° C for each degree of external temperature above and below 22°C. We believe that this correction factor that we calculated is very important for the promotion of the EBT from a research setting to clinical practice.
TE D
Another factor that influenced in our analysis the EBT was gender. Male appeared to have higher EBT compared to female, especially in the group of 23°C of external temperature and in that from 23 to 28°C. Higher external temperature makes the gender less important in influencing the EBT
EP
value. We suggest that higher blood pressure and cardiac diseases usual in males compared to females may influence EBT values. Another factor that might play an important role in sexual
AC C
differences is the sexual hormones with their selective inflammatory or antinflammatory actions. Particularly the female sexual hormones (estradiol and progesterone) have been demonstrated to have an antiflammatory effect that modulate the cellular and the immune response that might consequently influence the basal EBT in this sex resulting lower compared to those of males [15]. Crespo Lessmann et al, previously also reported differences of the EBT with gender [10]. In agreement with the data from Popov et al. studies, we did not find any influence of other variables, such as height, weight, area of residence, work, blood pressure, blood oxygen saturation, lung function, axillary temperature on the EBT. However, Bijnens et al previously reported a strong 10
ACCEPTED MANUSCRIPT relationship between the EBT and factors as age, BMI and residential proximity that in our study didn’t significantly influence the segmentation tree of the EBT. In conclusion, for the first time we reported reference values of EBT in a large population of healthy never smoked subjects. Furthermore, we found that the EBC is strongly influenced by
RI PT
external temperature and therefore we built a correction factor for EBT values in consideration of external temperature. In consideration of the importance that EBT could obtain in the study of respiratory diseases and of the role that it could gain in the clinical practice we support further
AC C
EP
TE D
M AN U
SC
studies in this field.
11
ACCEPTED MANUSCRIPT Contributions Conception and design; GEC, MPF, PJB
Analysis and interpretation: CC, DL, VS
RI PT
Drafting the manuscript for important intellectual content: GEC, PJB
AC C
EP
TE D
M AN U
SC
Conflicts of interest: The authors declare that they have no conflict of interest.
12
ACCEPTED MANUSCRIPT References
1. Paredi P, Kharitonov SA, Barnes PJ. Faster rise of exhaled breath temperature in asthma: a novelmarker of airway inflammation? Am J Respir Crit Care Med 2002; 165(2): 181-4.
RI PT
2. Kharitonov SA, Barnes PJ. Exhaled biomarkers. Chest 2006; 130(5):1541-6. 3. Melo RE, Popov TA, Solé D. Exhaled breath temperature, a new biomarker in asthma control: A pilot study. J Bras Pneumol 2010; 36(6):693-9. Lázár
Z,
Bikov
A,
Martinovszky
F, Gálffy
G, Losonczy
G, Horváth
I.
SC
4.
Exhaled breath temperature in patients with stable and exacerbated COPD. J Breath Res 2014;
M AN U
8(4):046002.
5. Carpagnano GE, Lacedonia D, Spanevello A, Martinelli D, Saliani V, Ruggieri C, FoschinoBarbaro MP. Exhaled breath temperature in NSCLC: could be a new non-invasive marker? Med Oncol 2014; 31(5):952.
6. Bade G, Gupta S, Kabra SK, Talwar A. Slower rise of exhaled breath temperature in cystic
Popov
TA, Dunev
S, Kralimarkova
TZ, Kraeva
S, DuBuske
LM.
Evaluation of
a simple, potentially individual device for exhaled breath temperature measurement.
Respir
EP
7.
TE D
fibrosis. Indian Pediatr 2015; 52(2):125-7.
Med 2007; 101(10):2044-50.
AC C
8. Vermeulen S, Barreto M, La Penna F, Prete A, Martella S, Biagiarelli F, Villa MP. Exhaled breath temperature in children: reproducibility and influencing factors. J Asthma 2014; 51(7):743-50.
9. Bijnens E, Pieters N, Dewitte H, Cox B, Janssen BG, Saenen N, Dons E, Zeegers MP, Int Panis L, Nawrot TS. Host and environmental predictors of exhaled breath temperature in the elderly. BMC Public Health 2013;13:1226. 10. Crespo Lessmann A, Giner J, Torrego A, Mateus E, Torrejón M, Belda A, Plaza V. Usefulness of the Exhaled Breath Temperature Plateau in Asthma Patients. Respiration 2015; 90(2):111-7.
13
ACCEPTED MANUSCRIPT 11. Couto M, Santos P, Silva D, Delgado L, Moreira A. Exhaled breath temperature in elite swimmers: the effects of a training session in adolescents with or without asthma. Pediatr Allergy Immunol 2015; 26(6):564-70. 12. Popov TA, Petrova D, Kralimarkova TZ, Ivanov Y, Popova T, Peneva M, Odzhakova T, Ilieva
Real
life
clinical
study
design
supporting
the
RI PT
Y, Yakovliev P, Lazarova T, Georgiev O, Hodzhev V, Hodzheva E,Staevska MT, Dimitrov VD. effectiveness
of
extra-fine
inhaled
beclomethasone/formoterol at the level of small airways of asthmatics. Pulm Pharmacol Ther 2013;
SC
26(6):624-9.
13. Piacentini GL, Peroni D, Crestati, Zardini F, Bodini A, Costella S, Boner AL. Exhaled air
M AN U
temperature in asthma: methods and relationship with markers of disease. Clin Exp Allergy 2007; 37:415–419.
14. Pifferi M, Ragazzo V, Previti A, Pioggia G, Ferro M, Macchia P, Piacentini GL, Boner AL. Exhaled air temperature in asthmatic children: a mathematical evaluation. Pediatr Allergy Immunol
TE D
2008; 20(2):164-71.
15. Perez-Alvarez MJ, Wandosell F. Stroke and neuroinflammation: role in sexual hormones. Curr
AC C
EP
Pharm Des 2016; 22(10): 1334-49.
14
ACCEPTED MANUSCRIPT TABLE LEGEND: Table 1: Anthropometric and functional data of healthy non smokers subjects
FIGURE LEGEND:
AC C
EP
TE D
M AN U
SC
RI PT
Figure 1: Classification tree for 298 healthy subjects
15
ACCEPTED MANUSCRIPT Table 1: Anthropometric and functional data of healthy non smokers subjects Healthy population N Age (years)
298 48 Q1=39 Q3=70 143 (48%)
Sex (M) BMI (Kg/m2) Median Interquartile range
27,7 Q1=24.6 Q3=31.2
Blood Pressure (mmHg) 120/79 Q1=115 Q3=121.25 (pressure max) Q1=73.75 Q3=81.25 (pressure min)
FEV1, % Pred 97.5 Q1= 95.75 Q3=100.25
FVC, % Pred
98 Q1=96 Q3=100.75
Median Interquartile range
FEV1/FVC ratio 97.7
Median Interquartile range
99 Q1=98 Q3=99.5
EP
28 (9%) 270 (91%)
AC C
Area of residence: rural industrialized
TE D
Median
SaO2 %
M AN U
Median Interquartile range
SC
Median Interquartile range
RI PT
Median Interquartile range
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
Figure 1: Classification tree for 298 healthy subjects
S0012-3692(16)62394-2
DOI:
10.1016/j.chest.2016.11.013
Reference:
CHEST 819
To appear in:
CHEST
Received Date: 2 August 2016 Revised Date:
29 September 2016
Accepted Date: 2 November 2016
Please cite this article as: Carpagnano GE, Foschino-Barbaro MP, Crocetta C, Lacedonia D, Saliani V, Zoppo LD, Barnes PJ, Validation of the exhaled breath temperature measure: reference values in healthy subjects, CHEST (2016), doi: 10.1016/j.chest.2016.11.013. 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 Validation of the exhaled breath temperature measure: reference values in healthy subjects
Giovanna E Carpagnano, Maria P Foschino-Barbaro, 1Corrado Crocetta, Donato Lacedonia, Valerio
RI PT
Saliani, Luigi Davide Zoppo, 2Peter J Barnes
Department of Medical and Surgical Sciences, Institute of Respiratory Diseases, University of Foggia – Italy; 1Section of Statistics, Faculty of Economics, University of Foggia ; 2Airway Disease
SC
Section, National Heart and Lung Institute, Imperial College London, Royal Brompton Hospital,
Address correspondence Dr Giovanna Elisiana Carpagnano Institute of Respiratory Diseases,
M AN U
London, UK.
University of Foggia, Italy
TE D
Department of Medical and Surgical Sciences,
EP
Email: [email protected]
AC C
Running title: EBT in healthy
Key words: exhaled breath temperature, validation, non-invasive methods, airways inflammation, reference value, correction factor
1
ACCEPTED MANUSCRIPT Abstract Background: Exhaled breath temperature (EBT) is a new non-invasive method for the study of inflammatory respiratory diseases with a potential to reach clinical practice. However, few studies,
and the range of normal values is not well established.
RI PT
and mainly derived from small, paediatric populations, are available on the validation of the method
The aim of this study was to measure EBT values in an Italian population of 298 (45.2±15.5 yrs, 143 male, FEV1 97.2±5.8%, FVC 98.4±3.9%) selected from 867 adult volunteers, in order to define
SC
reference values in healthy subjects and to analyse the influence of individual and external variables on this parameter.
M AN U
Methods: We measured EBT with a X-halo PRO device to different ambient temperature that ranged from 0 to 38°C
Findings: We report reference value of EBT in healthy Caucasian never smoker subjects. EBT values are strongly influenced by the external temperature and to a lesser extent by gender.
TE D
Interpretation: These data provide reference values in a large population of healthy never smokers subjects for measuring EBT as a basis for future studies. Our results are a contribution to the promotion of the EBT from bench to bed side.
AC C
EP
Fundings: No funding was provided for this review.
2
ACCEPTED MANUSCRIPT Introduction: The exhaled breath temperature (EBT) is a new non-invasive method for the analysis of airway inflammation. Since calor (heat) is listed as one of the cardinal signs of inflammation, the measure of the EBT has been developed as a marker of airway inflammation and therefore used in the study
[4], and lung cancer [5] and cystic fibrosis [6].
RI PT
of inflammatory respiratory diseases. Several studies have been performed in asthma [1-3], COPD
A new third generation thermometer, the X-halo (Delmedica), is now commercially available for
SC
the measure of the EBT. The smaller version of X-halo is now being developed for personal daily use as an “inflammometer” [7]. The use of EBT devices is particularly attractive in asthmatic
M AN U
patients that largely showed significantly higher EBT values compared to healthy subjects and that therefore are encouraged to use these exhaled thermometers in clinical practice for the maintenance of the asthma control and in the therapeutic management.
There are several potential advantages of EBT measurement with the X-halo device. The
TE D
measurement is very easy for patients and requires only few minutes, it is completely non-invasive and is therefore also suitable for children and patients with severe diseases. The device is well accepted by patients and ethics committee, it is inexpensive and finally it does not affect the
EP
underlying airway disease. Despite the potential of EBT, it still has not yet reached the clinical setting, partly because of a lack of a standardization and validation of the method.
AC C
Popov et al. evaluated the influence of room temperature, atmospheric pressure and relative humidity on EBT measurements with X-halo in a group of healthy subjects and asthmatic patients [7]. Furthermore they demonstrated the reproducibility of EBT measurements [7]. Vermeulen S et al. confirmed the reproducibility of EBT in a group of 124 healthy and asthmatic children [8]. Some important data are available on standardization and validation of EBT measurements that are needed to advance the clinical applicability of EBT measurements as a potential non-invasive marker of airway inflammation [9].
3
ACCEPTED MANUSCRIPT In order to validate EBT measurements it is important to follow the validation used for other noninvasive methods, such as induced sputum and exhaled nitric oxide that have now earned a place in the routine management of asthma and COPD and are now included in guidelines [2]. Data from previous studies on thousands of subjects have shown that EBT is a safe and entirely
RI PT
non-invasive method. No contraindications exist to its measurement and no adverse events have ever been reported. As for standardization of EBT measurements, the Delmedica manual for users has some recommendations for its use and for the cleaning the device (http://www.x-halo.com).
SC
For validation of EBT measurements, several factors need to be addressed: 1) the reproducibility of the measurement, 2) comparison with other biomarkers, 3) differences between measurements in
M AN U
diseased and healthy subjects, 4) sensitivity to risk factors or treatment and 5) reference normal values. Several studies reported good reproducibility of EBT measurements [7,8]. 2) Some comparison studies have also reported between EBT, exhaled nitric oxide [8] and induced sputum [8,10]. 3) Differences in EBT measurements in healthy subjects and patients with asthma [1-3],
TE D
COPD [4], cystic fibrosis (CF) [6], and lung cancer [5] have been reported. However, these studies have included only small numbers of subjects a. Reference values in the general population are therefore not well defined. 4) Also there are few studies looking at how the measurement may be
EP
changed physiologically with exercise and ventilation or with inhaled therapy [11,12]. This study was designed to measure EBT in a large population of healthy never-smoking adults in
AC C
order to provide reference values for the healthy population. Furthermore we aimed to analyse the influence of external variables on EBT measurements.
4
ACCEPTED MANUSCRIPT Material and Methods
Population We enrolled 867 consecutive Caucasian Italian volunteers (50.8±16.3 yrs, 60% male, BMI
RI PT
27.11±5.2 kg/m2, 59% smokers, FEV1 89.3±8.4%, FVC 90.7±4.8%, 90% living in industrialized areas while 10% in rural) during public regional meetings (“Breath Days” – Puglia - Italy) in shopping centres. The study was approved by the institutional ethics committee, University of
SC
Foggia (IRB approval number 17/CE/2014). All subjects were informed of the purpose of breath temperature measurement and after signing informed consent form, at the time of recruitment,
M AN U
anthropometric, physiological and clinical data were collected, including date of birth, height, weight, gender, area of residence, work, hobbies, diagnosed diseases, drugs, smoking history, blood pressure, blood oxygen saturation, lung function (with Pony FX- Cosmed), axillary body temperature and EBT measurement. We excluded from the population enrolled, all subjects with
TE D
any respiratory, cardiac or others diseases and all smokers and former smokers to obtain a new population of 298 healthy never smoked subjects (Table 1).
EP
Exhaled breath temperature measurement
EBT was measured with an X-halo device (Delmedica Investments, Singapore) according to
AC C
previously validated methods [13,14]. Briefly, patients were requested to inhale freely through the nose and to exhale into the device at a rate and depth typical of their normal tidal-breathing rhythm.
Ambient temperature measurement In consideration that EBT measurement were done in three different places (outside, hospital, shopping center) and that only when in hospital the temperature is usually controlled (between 19 and 24°C) we decided to measure and register ambient temperature before each EBT measurement. We used an external thermometer (PCE-HT 110, Italy) with a range of measure among 0-50 °C, 5
ACCEPTED MANUSCRIPT with precision variability of ±0.8°C, resolution: 0.1°C . We measured EBT to different ambient temperature that ranged from 0 to 38°C.
Statistical analysis
RI PT
To find the variables that have influenced EBT measurements we used a classification tree called ECHAID, that is an automatic classification algorithm that imposes a tree-like structure of the data. At each split, the algorithm looks for the predictor variable that if split, most "explains" the category
SC
response variable. In order to decide whether to create a particular split based on this variable, the CHAID algorithm tests a hypothesis regarding dependence between the split variable and the
M AN U
categorical response (using the chi-squared test for independence). Using a pre-specified significance level, if the test shows that the split variable and the response are independent, the algorithm stops the tree growth. Otherwise the split is created, and the next best split is searched. Using the result obtained by the classification tree we estimated a regression model to study the
AC C
EP
TE D
relationship between external temperature and EBT. The level of significance was set at p <0.05.
6
ACCEPTED MANUSCRIPT Results
Population 867 subjects volunteers were enrolled consecutively and successively divided on the basis of
RI PT
presence or absence of respiratory or other diseases and smoking habit to obtain 298 healthy never smoked subjects. Anthropometric, functional and clinical data of healthy subjects enrolled were
SC
summarized in Table 1.
EBT in healthy non smoker subjects
M AN U
E-CHAID tree estimated using the 298 healthy subjects, found that reference value of the EBT in healthy Caucasian non-smoking subjects was 30.459±2.955°C (Fig. 1). External temperature is the main factor influencing the EBT (F=58.991; df1=2, df2=295; p<0.00001). There are three groups of external temperatures that differently influence EBT.
TE D
The first group considers the cases with external temperature below 23°C. In this case the average EBT is 28.268±2.872°C. The second group considers cases measured with an external temperature of 23-28°C. In this case the EBT is 30.949±2.511°C. The third group shows that if the test is
EP
performed with an external temperature greater then 28°C the EBT is 32.558±1.805°C. The third level of the tree shows that gender influences the EBT for the first two groups. The female
AC C
EBT is respectively 1.486°C and 0.915°C lower than the male EBT. Other factors analysed as the age, the BMI, the blood pressure, internal temperature, lung function and area of residence influenced but without reaching statistical significance in the EBT. According to the result of the segmentation tree we estimated the following regression model:
EBT= α+β external temperature + ε
with α= 26.40 and β= 0.19, with E[ε]=0 and VAR [ε]=1whit R2= 0.2, F=175.4. 7
ACCEPTED MANUSCRIPT The model shows that the EBT is influenced by the external temperature. By average when the external temperature grows of 1°C the EBT grows of 0.19°C. For this reason it is important to measure the external temperature and if necessary to apply a correction factor to the results obtained. Using this model, the reference value of EBT in healthy non smoker subjects is 30.66°C
RI PT
at 22°C of external temperature that is common in hospital rooms. At 24°C the value is 31.4°C; at 26°C 31.40°C and at 28°C is 31.8°C.
AC C
EP
TE D
M AN U
SC
No adverse events were reported during or after the EBT measurement.
8
ACCEPTED MANUSCRIPT Discussion: In this study we report reference values for EBT in 298 healthy non-smoking subjects. We found that the EBT is strongly influenced by the external temperature and less by gender. Almost 200 articles are currently available in the literature on the use of EBT in asthma and other
RI PT
respiratory diseases. However, few of these studies have looked at the of EBT measurements in healthy subjects, which is needed if this non-invasive technique is to be useful in the clinic or measuring airway inflammation. There are no studies providing reliable reference values of EBT in
SC
healthy subjects. For this reason, the main aim of our study was to measure the EBT in a large normal population to define the reference EBT values in healthy never smoking subjects. We
M AN U
enrolled 867 apparently healthy volunteers who agreed to participate in this study. From this population we excluded all subjects with respiratory, cardiac or other diseases and all smokers or ex smokers giving a selected population of 298 healthy subjects that we used for the analysis. From this large healthy population, that is a great strength of this work, we report a reference value
could signify inflammation.
TE D
for the EBT that researchers and clinicians might use to identify subjects with increased EBT that
A second aim in this validation study was to explore the influence of ambient temperature and
EP
anthropometric, physiological and clinical variables (age, gender, height, weight, blood pressure, axillary temperature, lung function) on the EBT value. Popov et al. first addressed some of these
AC C
issues but in small numbers of patients [7]. In contrast to Popov et al, we found on a large population of 298 healthy subjects, that external (ambient) temperature strongly influences EBT values. In our daily practice, we noticed that the EBT, when measured in ambient temperatures up to 28°C was significantly higher and therefore we decided to perform our measurements at different ambient temperatures from 0°C to 38°C. The study was performed not only in hospital wards but also outside during public exhibitions and events in shopping centres. As expected, the ambient temperature strongly influences the EBT especially below 20 °C and over 27°C. To have tested EBT to the high external temperature found 9
ACCEPTED MANUSCRIPT in Southern Italy could be the reason for our discordant data compared to those of Popov et al. study. Usually the EBT measurement is performed in hospital where the temperature is usually controlled to be between 19 and 24°C. Also the Delmedica instruction for using X-halo suggest measuring EBT between 19 and 24°C assuming no influence of the external temperature on EBT
RI PT
values. However, there are wards where the temperature is not automatically adjusted where wider variations in ambient air temperature are frequent. Furthermore, if EBT is measured at home there may be greater temperature variations. While the measurement of EBT in controlled ambient
SC
temperature is the best option, we report a correction factor for ambient temperature that can normalize, in any situation, the EBT measurement. This correction factor calculated with a complex
M AN U
statistical method is really very simple to be applied and in its simplest form operators might increase the EBT value measured of 0.19° C for each degree of external temperature above and below 22°C. We believe that this correction factor that we calculated is very important for the promotion of the EBT from a research setting to clinical practice.
TE D
Another factor that influenced in our analysis the EBT was gender. Male appeared to have higher EBT compared to female, especially in the group of 23°C of external temperature and in that from 23 to 28°C. Higher external temperature makes the gender less important in influencing the EBT
EP
value. We suggest that higher blood pressure and cardiac diseases usual in males compared to females may influence EBT values. Another factor that might play an important role in sexual
AC C
differences is the sexual hormones with their selective inflammatory or antinflammatory actions. Particularly the female sexual hormones (estradiol and progesterone) have been demonstrated to have an antiflammatory effect that modulate the cellular and the immune response that might consequently influence the basal EBT in this sex resulting lower compared to those of males [15]. Crespo Lessmann et al, previously also reported differences of the EBT with gender [10]. In agreement with the data from Popov et al. studies, we did not find any influence of other variables, such as height, weight, area of residence, work, blood pressure, blood oxygen saturation, lung function, axillary temperature on the EBT. However, Bijnens et al previously reported a strong 10
ACCEPTED MANUSCRIPT relationship between the EBT and factors as age, BMI and residential proximity that in our study didn’t significantly influence the segmentation tree of the EBT. In conclusion, for the first time we reported reference values of EBT in a large population of healthy never smoked subjects. Furthermore, we found that the EBC is strongly influenced by
RI PT
external temperature and therefore we built a correction factor for EBT values in consideration of external temperature. In consideration of the importance that EBT could obtain in the study of respiratory diseases and of the role that it could gain in the clinical practice we support further
AC C
EP
TE D
M AN U
SC
studies in this field.
11
ACCEPTED MANUSCRIPT Contributions Conception and design; GEC, MPF, PJB
Analysis and interpretation: CC, DL, VS
RI PT
Drafting the manuscript for important intellectual content: GEC, PJB
AC C
EP
TE D
M AN U
SC
Conflicts of interest: The authors declare that they have no conflict of interest.
12
ACCEPTED MANUSCRIPT References
1. Paredi P, Kharitonov SA, Barnes PJ. Faster rise of exhaled breath temperature in asthma: a novelmarker of airway inflammation? Am J Respir Crit Care Med 2002; 165(2): 181-4.
RI PT
2. Kharitonov SA, Barnes PJ. Exhaled biomarkers. Chest 2006; 130(5):1541-6. 3. Melo RE, Popov TA, Solé D. Exhaled breath temperature, a new biomarker in asthma control: A pilot study. J Bras Pneumol 2010; 36(6):693-9. Lázár
Z,
Bikov
A,
Martinovszky
F, Gálffy
G, Losonczy
G, Horváth
I.
SC
4.
Exhaled breath temperature in patients with stable and exacerbated COPD. J Breath Res 2014;
M AN U
8(4):046002.
5. Carpagnano GE, Lacedonia D, Spanevello A, Martinelli D, Saliani V, Ruggieri C, FoschinoBarbaro MP. Exhaled breath temperature in NSCLC: could be a new non-invasive marker? Med Oncol 2014; 31(5):952.
6. Bade G, Gupta S, Kabra SK, Talwar A. Slower rise of exhaled breath temperature in cystic
Popov
TA, Dunev
S, Kralimarkova
TZ, Kraeva
S, DuBuske
LM.
Evaluation of
a simple, potentially individual device for exhaled breath temperature measurement.
Respir
EP
7.
TE D
fibrosis. Indian Pediatr 2015; 52(2):125-7.
Med 2007; 101(10):2044-50.
AC C
8. Vermeulen S, Barreto M, La Penna F, Prete A, Martella S, Biagiarelli F, Villa MP. Exhaled breath temperature in children: reproducibility and influencing factors. J Asthma 2014; 51(7):743-50.
9. Bijnens E, Pieters N, Dewitte H, Cox B, Janssen BG, Saenen N, Dons E, Zeegers MP, Int Panis L, Nawrot TS. Host and environmental predictors of exhaled breath temperature in the elderly. BMC Public Health 2013;13:1226. 10. Crespo Lessmann A, Giner J, Torrego A, Mateus E, Torrejón M, Belda A, Plaza V. Usefulness of the Exhaled Breath Temperature Plateau in Asthma Patients. Respiration 2015; 90(2):111-7.
13
ACCEPTED MANUSCRIPT 11. Couto M, Santos P, Silva D, Delgado L, Moreira A. Exhaled breath temperature in elite swimmers: the effects of a training session in adolescents with or without asthma. Pediatr Allergy Immunol 2015; 26(6):564-70. 12. Popov TA, Petrova D, Kralimarkova TZ, Ivanov Y, Popova T, Peneva M, Odzhakova T, Ilieva
Real
life
clinical
study
design
supporting
the
RI PT
Y, Yakovliev P, Lazarova T, Georgiev O, Hodzhev V, Hodzheva E,Staevska MT, Dimitrov VD. effectiveness
of
extra-fine
inhaled
beclomethasone/formoterol at the level of small airways of asthmatics. Pulm Pharmacol Ther 2013;
SC
26(6):624-9.
13. Piacentini GL, Peroni D, Crestati, Zardini F, Bodini A, Costella S, Boner AL. Exhaled air
M AN U
temperature in asthma: methods and relationship with markers of disease. Clin Exp Allergy 2007; 37:415–419.
14. Pifferi M, Ragazzo V, Previti A, Pioggia G, Ferro M, Macchia P, Piacentini GL, Boner AL. Exhaled air temperature in asthmatic children: a mathematical evaluation. Pediatr Allergy Immunol
TE D
2008; 20(2):164-71.
15. Perez-Alvarez MJ, Wandosell F. Stroke and neuroinflammation: role in sexual hormones. Curr
AC C
EP
Pharm Des 2016; 22(10): 1334-49.
14
ACCEPTED MANUSCRIPT TABLE LEGEND: Table 1: Anthropometric and functional data of healthy non smokers subjects
FIGURE LEGEND:
AC C
EP
TE D
M AN U
SC
RI PT
Figure 1: Classification tree for 298 healthy subjects
15
ACCEPTED MANUSCRIPT Table 1: Anthropometric and functional data of healthy non smokers subjects Healthy population N Age (years)
298 48 Q1=39 Q3=70 143 (48%)
Sex (M) BMI (Kg/m2) Median Interquartile range
27,7 Q1=24.6 Q3=31.2
Blood Pressure (mmHg) 120/79 Q1=115 Q3=121.25 (pressure max) Q1=73.75 Q3=81.25 (pressure min)
FEV1, % Pred 97.5 Q1= 95.75 Q3=100.25
FVC, % Pred
98 Q1=96 Q3=100.75
Median Interquartile range
FEV1/FVC ratio 97.7
Median Interquartile range
99 Q1=98 Q3=99.5
EP
28 (9%) 270 (91%)
AC C
Area of residence: rural industrialized
TE D
Median
SaO2 %
M AN U
Median Interquartile range
SC
Median Interquartile range
RI PT
Median Interquartile range
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
Figure 1: Classification tree for 298 healthy subjects