Keywords
chronic obstructive pulmonary disease - capnography - blood gas analyses - carbon
dioxide
Introduction
According to the estimates of the World Health Organization, there are approximately
65 million moderately severe cases of chronic obstructive pulmonary disease (COPD)
in the world. Although the poor recognition of COPD and its diagnosis at advanced
disease stage continue to affect epidemiological information, it is one of the leading
causes of morbidity and mortality worldwide.[1] The economic burden of COPD is estimated to be 48.4 billion euros in the European
Union.[2] In the United States of America, the direct economic burden is estimated at 32 billion
dollars, and there is an additional 20 billion cost indirectly added to this burden.
The severity of COPD is directly proportional to the cost of care, which is, in most
cases, associated with the exacerbation of the disease.[3]
Information about the metabolic and respiratory status of patients with COPD can be
obtained by examining arterial blood gas parameters, including pH, partial pressure
of oxygen (PaO2), partial pressure of carbon dioxide (PaCO2), base deficit, serum bicarbonate (HCO3), and oxygen saturation.[4] However, arterial blood gas collection is a painful procedure, and the frequency
of complications increased in the presence of repeated interventions, which has led
researchers to investigate noninvasive methods to determine the PaO2 and PaCO2 values that are considered to be the two cornerstones of treatment. Capnography,
one of the methods developed for this purpose, measures the partial pressure of carbon
dioxide from the airway during respiration.[5] With this noninvasive method, instant information can be obtained ventilation of
the patient. Capnography has been used for patient follow-up in intensive care units
for many years.[4]
[5]
[6]
In this study, we aimed to investigate whether end-tidal carbon dioxide (ETCO2) values could be used instead of PaCO2 values in guiding treatment and determining treatment benefits in patients that received
a diagnosis of COPD exacerbation at the emergency department.
Material and Method
Study Design and Population
This study was conducted prospectively and observationally at Emergency Department
of Fatih Sultan Mehmet Training and Research Hospital and Emergency Department of
Süreyyapaşa Chest Diseases and Chest Surgery Training and Research Hospital. The study
group consisted of adult patients with a previous diagnosis of COPD, who presented
to the emergency department of the two hospitals from April 1, 2015, through August
31, 2015, with the complaint of shortness of breath and received a diagnosis of COPD
exacerbation. As defined in Global Initiative for Chronic Obstructive Lung Disease
(GOLD), COPD exacerbation was accepted as an acute event characterized by the worsening
of the patient's respiratory symptoms beyond the daily observed normal variability,
resulting in changes in medication. In patients, the presence of at least one of the
symptoms of worsening dyspnea, increased amount of sputum, and increased sputum purulence
was sought.[7] Exclusion criteria included (i) no indication for arterial blood gas collection
to make clinical diagnosis and management; (ii) patients whose first place of reference
is not the emergency department (referrals) where the study was conducted; (iii) COPD
exacerbation not confirmed by a pulmonologist; (iv) concurrently with one or more
of the diagnoses of pulmonary edema, pulmonary embolism, and pneumonia; (v) mechanically
ventilated patients; and (vi) patients who did not give their consent to participate
in the study.
Data Collection
Vital signs and anamnesis were taken. Following physical examinations, the blood gas
analysis was performed in patients with relevant indications after obtaining their
consent for this procedure. At the time when the clinician was taking the first blood
gas samples, the first ETCO2 values of the patients were measured with the Masimo ISA capnograph using the sidestream
method. After five to six spontaneous respirations, three or four simultaneous capnograms
were observed on the monitor, and the ETCO2 value measured by the capnograph was recorded. Since the study was conducted observationally,
the treatments applied to the patients were recorded without any interference. The
control blood gas samples were taken by the clinician at the first hour after the
treatment was completed, and the second ETCO2 measurements were undertaken. The blood gas samples taken from the patients were
examined in the emergency laboratories of the two hospitals. The samples were taken
with a blood gas injector containing 80 IU of electrolyte balanced heparin and examined
in the ABL 700 blood gas analyzer.
Statistical Analysis
For the statistical analyses of the data obtained from the study, IBM SPSS Statistics
v. 22 was used. The conformity of the parameters to the normal distribution was evaluated
with the Shapiro–Wilk test. In addition to descriptive statistical methods (mean,
standard deviation, and frequency), the paired-samples t-test was used in the pre- and post-treatment comparisons of quantitative parameters
in the presence of a normal distribution. The Mann–Whitney U test was conducted to
compare the non-normally distributed parameters between the treatment groups. The
Pearson correlation analysis was used to examine the relationships between PaCO2 and ETCO2 parameters, which conformed to the normal distribution. In addition, the Bland–Altman
plot was constructed to determine the agreement between the two parameters. Significance
was evaluated at the p-value less than 0.05 level.
Ethical Considerations
Ethical approval for the study was obtained from the ethics committee of Fatih Sultan
Mehmet Training and Research Hospital (date: 12.03.2015, number: FSM EAH-KAEK/18 SAYI
2015/7). The content and purpose of the study were explained to the participants,
and their consent was obtained. The researchers adhered to the tenets of the Declaration
of Helsinki throughout the study.
Results
The study was carried out with a total of 121 cases, 78 (64.5%) male and 43 (35.5%)
female, aged 33 to 89 years. The median age of the patients was 67 (25th and 75th
percentiles: 60–75) years. Baseline characteristics of the patients are presented
in [Table 1].
Table 1
Baseline characteristics of the patients
|
Parameters
|
|
Age (median, 25th and 75th percentiles)
|
67 (60–75)
|
|
Male (n, %)
|
78 (64.5%)
|
|
Female (n, %)
|
43 (35.5%)
|
|
Vital parameters (mean/standard deviation)
|
|
Systolic arterial pressure (mm Hg)
|
134.13 ± 22.92
|
|
Diastolic arterial pressure (mm Hg)
|
78.55 ± 12.67
|
|
Pulse rate (/minute)
|
102.93 ± 21.56
|
|
Respiratory rate (/minute)
|
27.54 ± 5.98
|
|
Body temperature (°C)
|
36.61 ± 0.59
|
|
SpO2 levels (%)
|
85.21 ± 9.5
|
|
Treatments applied in the emergency department (
n
, %)
|
|
Intravenous steroid
|
104 (86 %)
|
|
Beta2 agonist
|
120 (99.2 %)
|
|
Anticholinergic drug
|
116 (98.9 %)
|
|
Inhaled steroid
|
35 (28.9 %)
|
|
Theophylline
|
31 (25.6 %)
|
|
Noninvasive mechanical ventilation
|
8 (6.6 %)
|
Abbreviation: SpO2, oxygen saturation.
Combinations of short-acting inhaled beta2-agonist, anticholinergic, inhaler steroid, and systemic steroid treatments were applied
to 64 of the cases (52.9%), and methylxanthine treatment and noninvasive mechanical
ventilation were applied to 57 (47.1%) cases in addition to these treatments. Beta-agonists
were used in 120 (99.2%) patients, anticholinergics in 116 (98.9%), intravenous steroids
in 104 (86%), inhaled steroids in 35 (28.9%), theophylline in 31 (25.6%), and noninvasive
mechanical ventilation in 8 (6.6%) patients ([Table 1]).
[Table 2] presents the comparison of the changes in the pH, PaO2, PaCO2, arterial oxygen saturation (SaO2), HCO3, and ETCO2 values before and after treatment. The correlation between PaCO2 and ETCO2 was also evaluated. A positive statistically significant correlation was found between
the pre-treatment PaCO2 and ETCO2 values, with the percentage of correlation being determined as 73.6% (p = 0.001; p < 0.01, Pearson correlation analysis) ([Fig. 1]). There was a positive statistically significant correlation between the post-treatment
PaCO2 and ETCO2 values, with a correlation percentage of 88.3% (p = 0.001; p < 0.01, Pearson correlation analysis).
Fig. 1 Correlation graphs of partial pressure of carbon dioxide (PaCO2) and end-tidal carbon dioxide (ETCO2) before (A) and after (B) treatment.
Table 2
Comparison of changes in the pH, PaO2, PaCO2, SaO2, HCO3, and ETCO2 values before and after treatment
|
Before treatment (Mean ± SD)
|
After treatment
(Mean ± SD)
|
p-Value
|
|
pH
|
7.4 ± 0.06
|
7.42 ± 0.05
|
0.009**
|
|
PaO2
|
62.42 ± 18.45
|
71.31 ± 24.77
|
0.001**
|
|
PaCO2
|
46.04 ± 13.51
|
44.78 ± 11.96
|
0.011*
|
|
SaO2
|
87.61 ± 9.14
|
91.52 ± 6.39
|
0.001**
|
|
HCO3
|
29.1 ± 9.73
|
28.33 ± 5.74
|
0.248
|
|
ETCO2
|
32.79 ± 8.39
|
36.38 ± 10.12
|
0.001**
|
Abbreviations: ETCO2, end-tidal carbon dioxide; HCO3, serum bicarbonate; PaCO2, partial pressure of carbon dioxide; PaO2, partial pressure of oxygen; SaO2, arterial
oxygen saturation; SD, standard deviation.
Paired-samples t-test; *p < 0.05, **p < 0.01.
When the Bland–Altman method was used to compare the agreement between the PaCO2 and ETCO2 parameters before treatment, most of the measurements were found within the limits
of agreement (−4.9 and +31.4), and the mean difference was 13.2. The same analysis
of the post-treatment values also indicated most measurements were within the limits
of agreement (−2.6 and +9.4), with the mean difference being calculated as 8.4. The
Bland–Altman plots showing the agreement between the PaCO2 and ETCO2 parameters before and after treatment are shown in [Fig. 2].
Fig. 2. Bland–Altman plot showing the agreement between showing partial pressure of carbon
dioxide (PaCO2) and end-tidal carbon dioxide (ETCO2) before (A) and after (B) treatment. SD, standard deviation.
Discussion
This study investigated whether ETCO2 could be used instead of PaCO2, guiding treatment, and determining treatment benefits in patients with COPD exacerbation
at the emergency department. We found a statistically significant positive correlation
between the PaCO2 values obtained by arterial blood gas analysis and ETCO2 values obtained by sidestream capnography; nonetheless, we consider that it would
not be appropriate to use these values interchangeably in clinical practice.
In a study evaluating 118 patients presenting to the emergency department with COPD
exacerbation, Kartal et al found the mean ETCO2 value measured by the sidestream method as 33.7 ± 10.45 and the mean PaCO2 value as 42.2 ± 13.5. The authors reported that the agreement between ETCO2 and PaCO2 was low and ETCO2 could not be used instead of PaCO2 in clinical practice.[8] In a study by Delerme et al, evaluating 48 measurements performed with the sidestream
method in 43 patients with shortness of breath and in another study by Jabre et al,
in which the authors evaluated the measurements taken by the sidestream method during
the ambulance transportation of 49 patients with respiratory distress, the common
conclusion was that ETCO2 could not clinically replace PaCO2.[9]
[10] This inconsistency in the measurements made with the sidestream method may be due
to various factors, such as air leaks and fluid secretions clogging the measuring
tube. Similarly, in both the study of Kartal et al and our study, the difference between
the ETCO2 and PaCO2 values may have arisen due to the cohort consisting of COPD cases. A plausible explanation
for this may be the reduced exhalation of CO2 in COPD due to both increased dead space and impaired ventilation. During the exacerbation
of the disease, CO2 excretion is even more limited.
The results of our study were also examined in terms of whether ETCO2 values could be used to guide treatment. While the 26 patients with an initial ETCO2 value of 40 and above had a mean ETCO2 value of 44.5 before treatment, the PaCO2 values of the same patients were all above 45, with a mean value of 64.15. Since
there was a statistically significant positive correlation between PaCO2 and ETCO2, we consider that high ETCO2 values can be used in the decision to start noninvasive mechanical ventilation therapy
in patients with COPD exacerbation. When the data were examined in terms of the PaCO2 values, 49 PaCO2 values measured in the pre-treatment period were 45 and above, while the ETCO2 values of 23 of the same patients were found to be lower than 40. Therefore, we consider
that a low ETCO2 value will not be helpful in guiding treatment. In a study conducted by Doğan et
al to evaluate the success of ETCO2 in showing the severity of COPD exacerbation in 102 patients, it was reported that
the ETCO2 levels obtained with the mainstream method were higher in those with severe COPD
exacerbation. The authors suggested that these high values could be useful in the
prehospital setting to determine the need for noninvasive mechanical ventilation or
hospitalization, but ETCO2 values would provide limited benefits in evaluating patients with COPD exacerbation.[11] In this study, when the Bland–Altman analysis was undertaken to compare the agreement
between the PaCO2 and ETCO2 parameters before treatment, we determined that most measurements were within the
limits of agreement (−4.9 and +31.4), and the mean difference was 13.2. Although this
is considered to be statistically consistent, due to the mean differences between
ETCO2 and PaCO2 being significantly high, we do not think that the ETCO2 value can be used instead of the PaCO2 value in the clinical setting.
The percentage of correlation between ETCO2 and PaCO2 increased after treatment. When the post-treatment agreement between the two parameters
was evaluated with the Bland–Altman method, most measurements were found to be within
the limits of agreement (−2.6 and +9.4), and the mean difference was 8.4. However,
despite the results being statistically significant and the mean difference being
reduced compared to the pre-treatment measurements, we consider that ETCO2 value cannot be clinically used to replace the PaCO2 value because there is still a large mean difference between the two parameters.
In this study, 36 of the 40 patients whose post-treatment ETCO2 values were 40 or higher had a PaCO2 value of 45 and higher. This suggests that high ETCO2 values measured after treatment seem to be effective in deciding whether to continue
noninvasive mechanical ventilation therapy. When we evaluated the usability of ETCO2 values in the decision to discontinue treatment, we determined that 12 of the 48
patients with a post-treatment PaCO2 value of 45 and above had an ETCO2 value of below 40. In light of these data, we consider that it is not appropriate
to terminate treatment based on lower or decreased ETCO2 values compared to the pre-treatment measurements.
There are certain limitations to our study. The first concerns the limited number
of patients we were able to reach over the determined study period and the small number
of measurements obtained per patient. Due to the observational design of our study,
blood gas samples were not taken from the patients without the clinician making this
decision, which resulted in a low number of measurement values for comparison. Second,
PaCO2 values measured in blood gas were not corrected for body temperature. Lastly, we
could not classify the cases included in the study according to the severity of COPD
exacerbation according to objective data. Therefore, the values of mild and severe
cases were evaluated together. On the other hand, our sample with a mean oxygen saturation
of 85 and pH of 7.4 seem to be mild COPD exacerbation cases. This may limit the generalizability
of the results of our study to patients with severe COPD exacerbations. Studies with
subgroup analysis may be better to determine how disease severity affects the relationship
between PaCO2 and ETCO2.
In conclusion, high ETCO2 values may indicate that noninvasive mechanical ventilation should be included in
the treatment of patients with COPD without waiting for the results of blood gas analysis.
Considering the high hospitalization rate of patients who require noninvasive mechanical
ventilation during the exacerbation period, high ETCO2 values can be used to predict the need for inpatient treatment. Despite these benefits,
ETCO2 has limited value in evaluating patients during the COPD exacerbation process and
guiding treatment.
