Assessment of newly developed real-time human respiration carbon dioxide measurement device for management of asthma outside of hospital

Abstract

BACKGROUND:

Assessment of asthma outside of the hospital using a patient independent device is highly in demand due to the limitation of existing devices, which are manual and unreliable if patients are not cooperative.

OBJECTIVE:

The study aims to verify the use of newly developed human respiration, carbon dioxide (CO ${}_{2}$ ) measurement device for the management of asthma outside of the hospital.

METHOD:

The data were collected from 60 subjects aged between 18–35 years via convenience sampling method reported in UTM Health Center using the device. Furthermore, the data were normalized and analyzed using descriptive statistics, t-test, and area ( $A_{z}$ ) under receiver operating characteristic curve (ROC).

RESULT:

Findings revealed that the normalized mean values of end-tidal carbon dioxide (EtCO ${}_{2}$ ), Hjorth Activity (HA), and respiratory rate (RR) were lower in asthmatic compared with healthy subjects with minimum deviation from the mean. In addition, each parameter was found to significantly differ statistically for asthma and non-asthma with $p<$ 0.05. Furthermore, the $A_{z}$ shows the strong association for the screening of asthma and non-asthma with an average of 0.71 (95% CI: 0.57–0.83), 0.77 (95% CI: 0.64–0.90), and 0.83 (95% CI: 0.73–0.94) for RR, EtCO ${}_{2}$ , and HA, respectively.

CONCLUSIONS:

This study demonstrates that the newly developed handheld human respiration CO ${}_{2}$ measurement device may possibly be used as an effort-independent asthma management method outside of the hospital.

Keywords

Asthma capnography early screening CO2 measurement

1. Introduction

Asthma is one of the major non-communicable and preventable diseases that is characterized by the inflammation or swelling of smaller airways (bronchioles) of the lung in response to different stimuli [1]. According to the Global Burden of Disease, asthma accounts for 242 million people globally, which is estimated to be up by 183 million from 1990 [2]. Furthermore, studies conducted by Anandan et al. revealed that asthma has been raised significantly since the 1960s and affected around 8.6% of the world’s young adults (aged 18–45) [3]. Of those, 4.5% of young asthmatic patients were diagnosed and treated, whereas 4.1% were undiagnosed due to a lack and limitations of existing methods for asthma diagnosis [4, 5]. According to the Global Initiative for Asthma (GINA 2016), existing devices (Spirometer and peak flow meter) are patient dependent and required to follow a set of instructions during manoeuvres [6]. Besides, young children, elderly, injured anaesthetized, sore, and ill patients cannot perform the test [7, 8]. Moreover, asthma is considered one of the predominant chronic diseases that has become the main reason to visit the emergency department and admission to the hospital settings each year [9]. In fact, it affects the millions of asthmatic patients’ personal lives as well as their professional lives [10]. In line, Hilger and Krull recommended that young adults suffering from asthma should have a self-managing asthmatic alert device in order to reduce their risk of exacerbation and hospital admission [11]. Hence, achieving acceptable asthma management and diagnosis in children and young adult on the right time remains evasive, despite the existing methods.

To overcome the limitation of existing methods, human respiration CO ${}_{2}$ measurement device such as capnography has been proposed as a noninvasive and effort-independent asthma screening tool [8, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25]. In support, the study conducted by Guthrie et al. [12] and Langhan et al. [13] revealed that measurement of EtCO ${}_{2}$ using capnograph device is viable for assessing acute asthma in children. Nagurka et al. [14] did a study on 299 asthmatic patients and measured EtCO ${}_{2}$ . The findings revealed that extreme (both low and high) EtCO ${}_{2}$ values may be used for the assessment asthmatic condition outside of the hospital that concurs with the finding of Lamba et al. [15]. Besides, Kesten et al. [16] performed a study of 47 acute asthmatic and 42 non-asthmatic patients and reported that RR increases during the natural attack of asthma that overlaps with the finding of Kassabian et al. [17] and Azab et al. [18]. In addition, a study performed by Kean and Malarvili [19] revealed that measurement of Hjorth Activity (HA) can be used as an index to differentiate asthma and non-asthma. It is calculated by squaring the standard deviations (see Fig. 1) of the amplitude of the waveform. The reason to calculate this parameter is that the analysis of it includes the slope of curves that are like CO ${}_{2}$ signal’s shape as presented in Fig. 1. Additionally, a study conducted by Howe et al., on the CO ${}_{2}$ signal elucidate that the computation of the slope of the capnogram can provide the significant information about the asthmatic and non-asthmatic CO ${}_{2}$ signal. Equation (1) was used to compute the HA. However, further study is needed to check the possibility of concurrent measurement of EtCO ${}_{2}$ , RR and HA for early diagnosis of asthmatic episodes beyond clinical assessment.

$\displaystyle\text{Hjorth Activity (HA)}=[\text{S(CO${}_{2}$(t))}]^{2}$ (1)

Where CO ${}_{2}$ ( $t$ ) is the measured CO ${}_{2}$ values, S(CO ${}_{2}$ ( $t$ )) is the standard deviation of the CO ${}_{2}$ waveform.

Figure 1.

Hjorth Activity computation for the complete breath cycle of the CO ${}_{2}$ signal; $t$ and CO ${}_{2}$ ( $t$ ) represent the time and output CO ${}_{2}$ values, respectively, whereas S(CO ${}_{2}$ ( $t$ )) elucidates the standard deviation of CO ${}_{2}$ waveform for the stipulated area presented as upward arrow.

Furthermore, Mieloszyk et al. [20] investigated four parameters such as exhalation duration, EtCO ${}_{2}$ , time spent at maximum CO ${}_{2}$ , and end-exhalation slope and concluded that these features can be used for differentiating chronic obstructive pulmonary disease (COPD)/Normal and COPD/congestive heart failure (CHF). Furthermore, Malik et al. [21] reported that CO ${}_{2}$ values vary with the asthmatic and non-asthmatic subject as per the severity level. In contrast to these studies, a study conducted by Howe et al. [8] revealed that EtCO ${}_{2}$ is not a reliable parameter for the differentiation of asthma and the control group. Thus, despite several positive studies, which supports that measurement of EtCO ${}_{2}$ , RR and HA are useful for the assessment of asthma in and outside of the hospital. To date, to the best of our knowledge, no study has been performed by combining all these parameters in correlation with the asthmatic condition due to lack of proof of evidence. In addition, one of the reasons is that there is no device to measure these parameters simultaneously. Moreover, a newly developed handheld real-time human respiration CO ${}_{2}$ measurement device may provide a better way to identify and manage the asthmatic episodes beyond clinical assessment [26]. The technical detail of the device can be found elsewhere [26]. Hence, the objective of this study is to conduct a study on young adult asthmatic and healthy subjects to verify the importance of these parameters for managing the asthmatic condition outside of the hospital.

2. Material and methods

2.1 Study design and settings

Sixty subjects, aged between 18–35 years, participated in this study. Thirty of these had an asthmatic disease that was identified based on records in the medical case remarks at the time of hospital admission and 30 were healthy volunteers. This study protocol was approved by the Medical Research and Ethics Committee (MREC), Ministry of Health Malaysia (Ref: (13) KKM/NIHSEC/P17-1027).

Informed consent and basic demographics were obtained on arrival at the healthcare centre before involvement in the study. The physical examination was performed with the asthmatic subject to assess the signs of asthma, including wheezing, air entry, and retractions. The enrolled subjects with asthma were evaluated by a medical physician. These evaluations assist in computing the Pulmonary Score (PS). The asthma severity level was recognized based on the PS score, which is composed of respiratory rate, sternocleidomastoid retractions, and wheezing. Each PS score is scaled from 0 to 3 [27]. The score from 0 to 3, 4 to 6, and 7 to 9 indicate a mild, moderate, and severe exacerbation of asthma, respectively. The assessment of asthma severity was performed prior to the measurement of vital signs, EtCO ${}_{2}$ and HA parameters. In addition, the patients were suffering from acute exacerbation of asthma were confirmed about their asthmatic condition by asking about their physical conditions. However, the subjects reported a more severe condition were excluded because this study focused on the screening of asthma and non-asthma rather than its severity level.

2.2 Data collection procedure

The data were collected using the sidestream technology-based human respiration CO ${}_{2}$ measurement device [26] as illustrated in Fig. 2 (A and B) from the UTM Health Center using a convenience sampling method [25], between February 2017 and June 2017. The device was made of four parts, namely, a CO ${}_{2}$ acquisition unit, a processing unit, real-time control (RTC) and a display unit. Firstly, CO ${}_{2}$ signal was acquired from the subjects through a sampling tube and diverted to the microcontroller unit for computation and transmission purpose. Furthermore, a high-resolution display was used to display the CO ${}_{2}$ signal and other parameters such as maximum and minimum CO ${}_{2}$ concentration, breathing rate, and activity) through serial communication. Moreover, the real-time control unit performs the data logging into SD card which is being controlled by the processing unit as reported in [26].

Besides, the developed device was small (12.50 cm $\times$ 13 cm $\times$ 8 cm) in size and light-weight (650 grams) as presented in Fig. 2, thereby, can be used outside of the hospital. Furthermore, the device was found highly reliable with an intra-class correlation coefficient of more than 0.8 for inter- and intra-rater. Besides, mean differences $\pm$ standard deviations were within the tolerable range (10–15%). Thus, the developed device was found highly reliable and valid measure for CO ${}_{2}$ , EtCO ${}_{2}$ , inspired carbon dioxide (ICO ${}_{2}$ ), RR, and HA.

During data collection, all the subjects were instructed to sit peacefully with relaxing on a chair to avoid the error in the EtCO ${}_{2}$ , RR and HA readings. Thereafter, they were instructed to breathe in and out via a simple nasal cannula with the relaxed manner at their own comfort. All the data were recorded for approximate 2 minutes and transferred to the computer for further analysis [28].

Figure 2.

A real-time human respiration carbon dioxide measurement device developed based on sidestream technology during data recording for non-asthmatic subjects (A) and (B) active state [26].

2.3 Data analysis

Firstly, data were normalized using Eq. (2), since normalized values of each parameter bring the parameter between 0 and 1 which makes simple in differentiating asthma and non-asthma condition. In addition, it is also useful to increase the efficiency of the classifier for differentiating the asthmatic condition provided the further implementation. Furthermore, data were analyzed using the SPSS software package (version 23.0) and necessary data were plotted using OriginPro (version 8.0). A descriptive statistical analysis was executed for subjects with and without asthma. Furthermore, the normality distribution of each parameter was established with respect to each measurement and the characteristic detail is incorporated into Table 1.

$\displaystyle\text{Normalized}=\frac{X_{i}-X_{\min}}{X_{\max}-X_{\min}}$ (2)

Where $X_{i}$ , $X_{\min}$ , and $X_{\max}$ represent the features (EtCO ${}_{2}$ , RR and HA), minimum value and maximum value, respectively in each feature; $i$ represents the number of samples.

Table 1

Characteristics detail of asthmatic and non-asthmatic patients

Characteristics	Asthma	Non-asthma
Gender (M/F)	18/12	14/16
Age, yrs	26.5 $\pm$ 8.24	24.86 $\pm$ 5.25
Weight, kg	62.24 $\pm$ 5.58	57.35 $\pm$ 10.23
Height, cm	165.42 $\pm$ 3.6	158.34 $\pm$ 3.24

Where M – male; F – female. ${}^{\text{a}}$ Data are depicted in terms of mean $\pm$ SD.

Table 2

Mean and SD values of the physiological parameter and PS score with increasing severity level for asthma on arrival at the hospital

Physiological parameter	Asthmatic subjects ( $n=$ 30, mean $\pm$ SD)
HR (beats/min)	91.32 $\pm$ 18.95
Oxygen saturation (%)	97.42 $\pm$ 2.19
Body temperature ( ${}^{\circ}$ C)	36.86 $\pm$ 0.34
PS score	Mild ( $n=$ 25)	Moderate ( $n=$ 5)
	2.59 $\pm$ 1.46	4.14 $\pm$ 0.37

In this, we calculated Shapiro-Wilk (S-W) $p$ -values and the skewness and kurtosis, $z$ -values for each parameter to check the normality of data, for both males and females. The S-W $p$ -value ( $p>$ 0.05) and $z$ -value ( $-$ 1.96 $\leqslant z\leqslant$ 1.96) was considered statistically significant [29]. Furthermore, the non-parametric t-test was performed to compare the mean of asthmatic and non-asthmatic subjects for each feature [30]. A $p$ -value ( $<$ 0.05) was considered statistically significant.

3. Result

This study reports the verification of the newly developed device which, measure the EtCO ${}_{2}$ , RR, and HA on the asthmatic subject for the management of asthma outside of the hospital. The data were collected from 30 asthmatic subjects that include 25 milds, and 5 moderate and 30 non-asthmatic subjects. The severity levels were identified based on the PS score computed during clinical investigations as reported in Table 2. The vital parameters recorded from the asthmatic subjects at the time of hospital admission and their pulmonary score as per the severity are tabulated in Table 2.

Furthermore, S-W $p$ -values ( $p<$ 0.05) and $z$ -values ( $-$ 1.96 $>z>$ 1.96) revealed that each parameter were non-normally distributed for both healthy and asthmatic subjects. Hence, a non-parametric t-test was performed to compare the mean of EtCO ${}_{2}$ , RR and HA in asthma and no-asthma condition. Table 3 shows the normalized mean and standard deviation (SD) of each parameter for non-asthma and asthma. Findings revealed that the normalized mean $\pm$ SD values were significantly higher for non-asthmatic compared with asthmatic subjects for all the parameters as illustrated in Table 3.

Table 3
Normalized mean and SD values for EtCO ${}_{2}$ , RR and HA for healthy and asthmatic subjects in terms of mmHg, bpm, and no unit

Parameters	Non-asthma (normalized, mean $\pm$ SD)	Asthma (normalized, mean $\pm$ SD)
EtCO ${}_{2}$	0.66 $\pm$ 0.22	0.42 $\pm$ 0.26
RR	0.42 $\pm$ 0.30	0.27 $\pm$ 0.23
HA	0.42 $\pm$ 0.25	0.33 $\pm$ 0.22

Table 4

Mean and SD values of the physiological parameter, and $A_{z}$ with 95% confidence interval (CI), $p$ -value, and SE for asthma and non-asthma subjects

Physiological parameter	Asthmatic subjects ( $n=$ 30)	Non-asthmatic subjects ( $n=$ 30)	$A_{z}$	95% (CI)		$p$ -value	SE
RR (breaths/min)	21.26 $\pm$ 4.57	17.16 $\pm$ 3.04	0.71	0.57	0.83	$<$ 0.05	0.06
EtCO ${}_{2}$ (mmHg)	33.76 $\pm$ 2.09	36.51 $\pm$ 2.50	0.77	0.64	0.90	$<$ 0.05	0.06
HA	152.87 $\pm$ 85.89	248.35 $\pm$ 58.62	0.83	0.73	0.94	$<$ 0.05	0.05

Figure 3.

Illustration of ROC curves to identify the best-suited features for differentiating asthma and non-asthma conditions. The black square (HA) possesses the higher sensitivity and specificity, followed by red circle (EtCO ${}_{2}$ ), and upward triangle (RR).

Table 4 lists the measured physiological parameter in terms of mean $\pm$ SD, along with $A_{z}$ with 95% confidence interval, $p$ -value, and standard error (SE) that help in identifying the features capabilities for discrimination of asthma and non-asthma. The $p$ -values reveal that all the features are statistically significant ( $p<$ 0.05) whose mean values significantly differ for asthmatic and non-asthmatic. However, $A_{z}$ and corresponding SE values indicate that HA of the complete breath cycle ( $A_{z}=$ 0.83 with SE $=$ 0.05), and end-tidal carbon dioxide (EtCO ${}_{2}$ , $A_{z}=$ 0.77 with SE $=$ 0.06), are highly statistically considerable than the RR, for the discrimination asthma and non-asthma condition. But, it should be noted that RR also exhibits the good as the $A_{z}$ value was 0.71, which can be used for asthma and non-asthma differentiation. Figure 3 depicts the feature capabilities based on their corresponding $A_{z}$ values.

Furthermore, Fig. 4 illustrates the direct comparison of EtCO ${}_{2}$ , RR and HA between non-asthma and asthma using box plot. These figures exhibit an average normalized data point for each subject, where the x and y-coordinate, represent a number of subjects and EtCO ${}_{2}$ , RR and HA values for non-asthmatic and asthmatic subjects, respectively. The graph elucidates that most of the mean EtCO ${}_{2}$ values were found to be lower for asthma compared to non-asthmatic. Additionally, average RR values were higher for most of the asthmatic subjects, in comparison to those without asthma, as can be seen in Fig. 4. Furthermore, the average HA values were higher in asthmatic comparing with non-asthmatic subjects.

Figure 4.

Box plot of EtCO ${}_{2}$ , RR and HA for the asthmatic and non-asthmatic subject. EtCO ${}_{2}$ _NA, EtCO ${}_{2}$ _A, RR_NA, RR_A, HA_NA, and HA_A represent the asthmatic and non-asthmatic condition for end-tidal carbon dioxide, respiratory rate and Hjorth Activity, respectively.

4. Discussion

This study demonstrated that concurrent measurement of EtCO ${}_{2}$ , RR and HA using a newly developed handheld device [26], are feasible to differentiate asthmatic and non-asthmatic condition in among young adults in and outside of the hospital. Table 3 shows that the normalized mean $\pm$ SD values of EtCO ${}_{2}$ , RR and HA were found lower in asthmatic subjects compared with healthy subjects. In addition, non-parametric t-test shows that HA is highly statistically significant ( $p<$ 0.05) in order to differentiate asthmatic and non-asthmatic conditions, comparing with EtCO ${}_{2}$ and RR in agreement with Kean and Malarvili [19]. In addition, although, all the features extracted from our study exhibits good discriminate capabilities with asthma and no-asthma as presented in Fig. 3 and Table 4. However, the features, namely, HA and EtCO ${}_{2}$ possess higher $A_{z}$ values (ranges: 0.77–0.83) and low standard error (ranges: 0.06–0.05) compared to RR. Therefore, HA and EtCO ${}_{2}$ could be the more appropriate feature for the classification of asthma and no-asthma, which may possibly maximize the accuracy, provided further implantation with the classifier. Furthermore, measurements of these parameters using the newly developed device are quite easy and are well tolerated, even with young adult and children in agreement with earlier studies reported for EtCO ${}_{2}$ by Langhan et al. [13]. Thus EtCO ${}_{2}$ , RR and HA can be considered as objective variables that can be assessed noninvasively and which, may be useful for evaluating the asthmatic condition in young adults outside of the hospital.

Figure 4 implies the qualitative view of the measured parameters for asthma and non-asthma. The EtCO ${}_{2}$ values were found lower for an asthmatic subject in agreement with an earlier study reported by Langhan et al. [13]. This also implies that the measurements of the EtCO ${}_{2}$ are sometimes greater, but most of the EtCO ${}_{2}$ values are less for asthmatic than those of non-asthmatic. In other words, there is a systematic difference in the measurements of the EtCO ${}_{2}$ compared to the healthy. Furthermore, RR was found higher in most of the asthmatic subjects which coincide with the finding of Kassabian et al. [17]. Furthermore, the HA values were lower in more than 80% of asthmatic subjects compared with non-asthmatic in agreement with an earlier study reported by Kean and Malarvili [19]. Thus, these results show that the measured parameters were radically difference between asthma and non-asthma. Hence, the developed device can be used as an early screening and managing tool outside of a hospital in order to reduce the mortality rate and admission in hospital.

The reason to test the newly handheld and light-weight developed device with asthmatic subjects is that the proper care of asthmatic patients, age-independent and objective measurement of airway obstruction may be helpful outside of the hospital. Furthermore, Kerem et al. [31] reported that respiratory rate, heart rate and physical examination findings were not correlated with severity of asthma by pulmonary function testing. On the other hand, Yaron et al. [23] and Kunkov et al. [25] disclosed that both the plateau angle and slope of the CO ${}_{2}$ signal are linked to severity of exacerbations of asthma. However, on contrary, Langhan et al. divulged that the analysis of CO ${}_{2}$ waveform cannot be easily quantified in an Emergency Department and are not clinically usable [13].

Furthermore, we can easily measure, monitor, and save the EtCO ${}_{2}$ , RR and HA over time and are not limited by inter- or intra-observer variability as the developed device has been tested-retested and validated with a standard CO ${}_{2}$ measurement device (CapnostreamTM20 Model CS08798) [26]. Besides, the proposed device can be used for all age groups or severity of their respiratory diseases in effort independent way. In addition, measurement of these parameters using the newly developed device has a significant advantage over traditional lung function test such as peak flow meter and spirometer that are patient dependent, required training manpower in order to operate, and difficult to interpret as well [32]. Furthermore, results are not reliable if the patient’s non-cooperative, particularly in patients with more severe degrees of airway obstruction [8, 33]. Thus, the developed device can be used over the traditional device to measure the asthmatic conditions in and outside of the hospital.

However, this study design is limited by the facts that this has a small number of populations and recruitment of patients was based on a convenience sample that makes difficult to generalize our finding. It might be argued by someone that measurement of EtCO ${}_{2}$ , RR and HA would be more useful parameters in patients with the mild and/or moderate stage of asthma. In young adult and children suffering from more mild and moderate disease, this device may provide a better way regarding treatment and admission versus patients reported with a severe condition of asthma in which decisions are obvious for the treatment protocol.

5. Conclusion

This study demonstrates that newly developed human respiration CO ${}_{2}$ measurement device that measures EtCO ${}_{2}$ , RR and newly introduced feature (HA) may possibly be used as an early screening and asthma management device. Here, we collected 30 asthmatic and non-asthmatic data and measured parameters are presented in terms of normalized mean $\pm$ SD. In addition, the non-parametric t-test and area ( $A_{z}$ ) under ROC curves were performed to confirm the finding from the statistical point of view. Finding reveals that normalized mean and SD values of EtCO ${}_{2}$ , RR and HA were lower with asthma comparing with non-asthmatics. Furthermore, these parameters were also statistically significant with a $p$ -value ( $p<$ 0.05) and $A_{z}$ values of 0.71 to 0.83. Thus, the newly developed device is found capable of differentiating asthma and non-asthma conditions and may possibly be used in and outside of the hospital for the management of asthmatic conditions. In future, the developed device will be tested on many asthmatic subjects suffering from a different severity level to generalize the finding.

Footnotes

Acknowledgments

This research is supported by the Ministry of Higher Education under the prototype research grant scheme (PRGS), grant no. R.J130000.7845.4L669. The authors would also like to express a deep gratitude towards the Universiti Teknologi Malaysia for providing facilities and laboratory equipment.

Conflict of interest

None to report.

References

Network

. The global asthma report 2014. Auckland, New Zealand. 2014, 769.

Stanaway

Flaxman

Naghavi

Fitzmaurice

Vos

Abubakar

, et al. The global burden of viral hepatitis from 1990 to 2013: Findings from the global burden of disease study 2013. The Lancet 2016; 388(10049): 1081-1088.

Anandan

Nurmatov

Van Schayck

Sheikh

. Is the prevalence of asthma declining? Systematic review of epidemiological studies. Allergy 2010; 65(2): 152-167.

Marks

Pearce

Strachan

Asher

. Global burden of disease due to asthma. Global Asthma Network. The Global Asthma Report 2014.

Centers for Disease Control and Prevention (CDC). Vital signs: Prevalence, treatment, and control of hypertension – United States, 1999–2002 and 2005–2008. MMWR. Morbidity and Mortality Weekly Report 2011; 60(4): 103.

Lenfant

. Global initiative for asthma, global strategy for asthma management and prevention. In: NHLBI/WHO Workshop Report 1995.

Goepp

, inventor; Goepp

, assignee. Effort-Independent, portable, user-operated capnograph devices and related methods. United States patent application US 12/265,574 2008.

Howe

Jaalam

Ahmad

Sheng

Ab Rahman

. The use of end-tidal capnography to monitor non-intubated patients presenting with acute exacerbation of asthma in the emergency department. J Emerg Med 2011; 41(6): 581-589.

Gibson

Powell

Wilson

Abramson

Haywood

Bauman

, et al. Self-management education and regular practitioner review for adults with asthma. Cochrane Database Syst Rev 2000; (2): CD001117.

10.

Pruitt

. Asthma self-management education programs: The key to good control. RT: The Journal for Respiratory Care Practitioners 2011; 24(5): 14-21.

11.

Hilger

Krull

. Controlling asthma: Self-management education for young adults. [cited 2017 February 14]; [Available from: rtmagazinecom/2014/05/controlling-asthma-self-management-education-young-adults].

12.

Guthrie

Adler

Powell

. End-tidal carbon dioxide measurements in children with acute asthma. Acad Emerg Med 2007; 14(12): 1135-1140.

13.

Langhan

Zonfrillo

Spiro

. Quantitative end-tidal carbon dioxide in acute exacerbations of asthma. J Pediatr 2008; 152(6): 829-832.

14.

Nagurka

Bechmann

Gluckman

Scott

Compton

Lamba

. Utility of initial prehospital end-tidal carbon dioxide measurements to predict poor outcomes in adult asthmatic patients. Prehosp Emerg Care 2014; 18(2): 180-184.

15.

Lamba

Gluckman

Nagurka

Rosania

Bechmann

Langley

, et al. 165: Initial out-of-hospital end-tidal carbon dioxide measurements in adult asthmatic patients. Ann Emerg Med 2009; 54(3): S51.

16.

Kesten

Maleki-Yazdi

Sanders

Wells

McKillop

Chapman

, et al. Respiratory rate during acute asthma. Chest 1990; 97(1): 58-62.

17.

Kassabian

Miller

Lavietes

. Respiratory center output and ventilatory timing in patients with acute airway (asthma) and alveolar (pneumonia) disease. Chest 1982; 81(5): 536-543.

18.

Azab

El Mahalawy

El Aal

Taha

. Breathing pattern in asthmatic patients during exercise. Egypt J Chest Dis Tuberc 2015; 64(3): 521-527.

19.

Kean

Malarvili

. Analysis of capnography for asthmatic patient. In: Proceeding of IEEE International Conference on Signal and Image Processing Applications 2009; 464-467.

20.

Mieloszyk

Verghese

Deitch

Cooney

Khalid

Mirre-Gonzalez

, et al. Automated quantitative analysis of capnogram shape for COPD-normal and COPD-CHF classification. IEEE Trans Biomed Eng 2014; 61(12): 2882-2890.

21.

Malik

Singh

Nurifhan

Malarvili

. Portable respiratory CO2 monitoring device for early screening of asthma. In: Proc of the Fourth International Conference on Advances in Computing, Electronics and Communication, 2016; 90-94.

22.

Brown

Gough

Seim

. Can quantitative capnometry differentiate between cardiac and obstructive causes of respiratory distress? Ann Emerg Med 1997; 30(3): 389-390.

23.

Yaron

Padyk

Hutsinpiller

Cairns

. Utility of the expiratory capnogram in the assessment of bronchospasm. Ann Emerg Med 1996; 28(4): 403-407.

24.

Nik Hisamuddin

Rashidi

Chew

, et al. Correlations between capnographic waveforms and peak flow meter measurement in emergency department management of asthma. Int J Emerg Med 2009; 283-89.

25.

Kunkov

Pinedo

Silver

Crain

. Predicting the need for hospitalization in acute childhood asthma using end-tidal capnography. Pediatr Emerg Care 2005; 21(9): 574-577.

26.

Singh

Howe

Balakrishnan

. Real-time human respiration carbon dioxide measurement device for cardiorespiratory assessment. J Breath Res 2017; 12(2): 026003.

27.

Smith

Baty

Hodge

. Validation of the pulmonary score: An asthma severity score for children. Acad Emerg Med 2002; 9(2): 99-104.

28.

Galia

Brimioulle

Bonnier

Vandenbergen

Dojat

Vincent

, et al. Use of maximum end-tidal CO2 values to improve end-tidal CO2 monitoring accuracy. Respir Care 2011; 56(3): 278-283.

29.

Shapiro

Wilk

. An analysis of variance test for normality (complete samples). Biometrika 1965; 52(3/4): 591-611.

30.

Galton

. Regression towards mediocrity in hereditary stature. Man (Lond) 1886; 15: 246-263.

31.

Kerem

Canny

Reisman

Bentur

Levison

Tibshirani

Schuh

. Clinical-physiologic correlations in acute asthma of childhood. Pediatrics 1991; 87(4): 481-486.

32.

Gorelick

Stevens

Schultz

Scribano

. Difficulty in obtaining peak expiratory flow measurements in children with acute asthma. Pediatr Emerg Care 2004; 20(1): 22-26.

33.

Eid

Yandell

Howell

Eddy