Abstract
Objective
Patients with chronic respiratory failure requiring long-term oxygen therapy (LTOT) are at a risk of CO2 retention because of excessive oxygen administration. The CapnoEye™ is a novel portable capnometer that can measure end-tidal CO2 (EtCO2) noninvasively. This retrospective study evaluated the usefulness of this device.
Methods
EtCO2 was measured using the CapnoEye™. The EtCO2 and partial pressure of venous carbon dioxide (PvCO2) were analyzed, and other clinical data were assessed.
Patients
Sixty-one consecutive patients with chronic respiratory failure receiving LTOT in the outpatient department at the Japanese Red Cross Medical Center between July 2017 and March 2018 were retrospectively reviewed.
Results
There was a significant correlation between EtCO2 and PvCO2 (r=0.63) in the total study population as well as in the COPD group (r=0.65) and ILD group (r=0.67). The PvCO2 and EtCO2 gradient was correlated with only the body mass index in a multivariate analysis (p=0.0235). The EtCO2 levels on the day of admission were significantly higher than those in the same patients when they were in a stable condition (p=0.0049). There was a significant correlation between ΔEtCO2 and ΔPvCO2 (r=0.4). A receiver-operating characteristic curve analysis revealed the optimal cut-off EtCO2 value for identifying hypercapnia to be 34 mmHg (p=0.0005).
Conclusion
The evaluation of EtCO2 by the CapnoEye™ was useful for predicting PvCO2. The body mass index was identified as a possible predictor of the PvCO2 and EtCO2 gradient. An increase in EtCO2 may indicate deterioration of the respiratory status in patients with chronic respiratory failure receiving LTOT.
Keywords: CapnoEye™, long-term oxygen therapy, end-tidal CO2, portable capnometer, chronic obstructive pulmonary disease, interstitial lung disease
Introduction
Supplemental long-term oxygen therapy (LTOT) improves the survival, exercise capacity, and quality of life in patients with chronic obstructive pulmonary disease (COPD) and hypoxemia (1) as well as in those with sequelae of tuberculosis (2). LTOT has also been reported to improve the quality of life in patients with chronic interstitial lung disease (ILD) (3). The number of patients receiving LTOT has increased in Japan (4) and is likely to continue to increase because of the aging population.
According to the Global Initiative for Chronic Obstructive Lung Disease guideline, LTOT is indicated in patients who have a PaO2 ≤55 Torr or an SaO2 ≤88% with or without hypercapnia and in those with a PaO2 of 55-60 Torr or an SaO2 of 88% if there is evidence of pulmonary hypertension suggesting congestive heart failure (1).
After oxygen therapy is started, blood gases should be checked to ensure that oxygenation is satisfactory without retention of CO2 and/or worsening acidosis. An American Thoracic Society/European Respiratory Society position paper reported that an arterial blood gas (ABG) analysis was the preferred method for determining the need for oxygen, as it includes acid-base information (5). Patients with chronic respiratory failure who need LTOT are at risk of CO2 retention as a result of the administration of excessive oxygen; therefore, it is important to monitor the blood CO2 level regularly. However, the evaluation of the blood CO2 level every time a patient visits the outpatient department is difficult because blood gas sampling is invasive and painful.
End-tidal CO2 (EtCO2), which is measured using a capnometer, is positively correlated with blood CO2 (6) and is now part of the standard of care for all mechanically ventilated patients receiving general anesthesia and routine monitoring in intensive-care settings (7). In July 2018, the CapnoEye™ MC600 (Nissei, Osaka, Japan), a novel capnometer that measures the EtCO2 level correctly in patients who are breathing spontaneously (Fig. 1), was approved for use in Japan, but its clinical value in patients receiving LTOT remains unclear.
Figure 1.
The CapnoEyeTM capnometer. The CapnoEyeTM can measure EtCO2 six times during spontaneous breathing. The EtCO2 is measured while holding the EtCO2 sensor with the hand opposite to the one with the SpO2 sensor attached and placing the mouthpiece in the mouth. EtCO2, end-tidal carbon dioxide.
The blood CO2 levels need to be measured noninvasively in outpatients with chronic respiratory failure because of the difficulties inherent in routine measurement of blood arterial CO2. The aim of this study was to evaluate the ability of the CapnoEye™ MC600 to measure EtCO2 and to assess the relationship between EtCO2 and the PCO2 level in patients receiving LTOT.
Materials and Methods
The study protocol was approved on June 1, 2018, by our institutional review board (approval number 680). The requirement for written informed consent was waived due to the use of an opt-out method (8). The study population consisted of 61 consecutive outpatients with chronic respiratory failure who received LTOT and underwent blood sampling between July 2017 and March 2018.
EtCO2 measurements
The mainstream EtCO2 level was measured using the CapnoEye™. Patients performed a tidal volume (TV) maneuver with the mouthpiece at a constant flow rate and in a relaxed position six times while holding the EtCO2 sensor with the hand opposite to the one with the SpO2 sensor attached. The EtCO2 was analyzed automatically as the average of the readings obtained during the six TV maneuvers and displayed on the monitor. The measurements were supported by experienced technicians in all cases.
Data collection
The flow of the enrolled patients throughout the study is shown in Fig. 2. Between July 2017 and March 2018, 89 patients who visited our outpatient department and received LTOT were considered for enrollment. Twenty-eight patients were excluded because a lack of either EtCO2 or partial pressure of venous carbon dioxide (PvCO2) data, thus leaving 61 patients for inclusion in the study. The correlations between EtCO2, PvCO2, and pulmonary function tests were analyzed in these 61 patients (Method 1). Forty of these patients were excluded because they were not admitted for an observation period, thereby leaving 21 patients who had been admitted for additional analysis. The baseline data for these 21 patients were collected when they were in a stable condition, i.e., with stable vital signs, a normal level of consciousness, and an assessment of at least one month before the most recent admission. The EtCO2 data at baseline were compared with those obtained during the admission period and correlations between ΔEtCO2, ΔPvCO2, and pulmonary function tests were sought (Method 2).
Figure 2.
Consort flow diagram. Of the 89 patients screened, 28 were excluded because neither the EtCO2 or PvCO2 measurements were recorded, thus leaving 61 patients for enrolment in the study (Method 1). Next, a further 40 patients were excluded because they were not admitted during the observation period, thereby leaving 21 patients who had been admitted for further analysis (Method 2).
Pulmonary function tests were performed using a rolling seal-type spirometer (Fudac-77; Fukuda Denshi, Tokyo, Japan). The medical records of each patient were also reviewed.
Statistical analyses
Correlations were analyzed using the Spearman’s rank correlation coefficient. The change in the EtCO2 level according to the respiratory status in the same patients was analyzed using Wilcoxon’s signed rank test. A multiple regression analysis was used to predict the values of dependent and independent variables. Logistic regression and receiver-operating characteristic curve analyses were used to evaluate the diagnostic performance of EtCO2. The descriptive data are shown as the median, frequency, and percentage. All reported p-values are two-sided. The data were analyzed using the JMP 9 software program, version 9.0.3 (SAS Institute, Cary, USA). A p value ＜0.05 was considered statistically significant.
Results
The characteristics of the 61 patients [36 men, 25 women; median age 74 (34-96) years old] who received LTOT at our hospital during the study period are shown in Tables 1 and 2. All patients had a Glasgow Coma Scale score ＞13, indicating a clear consciousness level. None of the patients were ventilated via tracheostomy. Eight patients required noninvasive positive pressure ventilation for COPD, ILD, or sequelae of tuberculosis. Fifty-four patients required oxygen therapy for 24 hours, and 7 required it only when sleeping or on exertion.
Table 1.
Demographic and Clinical Characteristics of the 61 Patients in the Study at Baseline.
| All (n=61) | Male (n=36) | Female (n=25) | p value | |
|---|---|---|---|---|
| Age, years | 74 (34-96) | 75.5 (54-90) | 72 (34-96) | 0.7579 |
| Smoking history, pack years | 7 (0-160) | 50 (0-160) | 0 (0-88.5) | <0.0001 |
| Body mass index | 19 (12-39) | 20.5 (14-32.6) | 17.6 (12-39) | 0.1164 |
| EtCO2, mmHg | 31 (18-41) | 34.5 (18-48) | 37 (25-53) | 0.2013 |
| PvCO2, mmHg | 49 (25-75) | 50 (25-75) | 48 (33-72) | 1 |
| PvCO2- EtCO2 | 15 (-1.4, 34) | 15 (-1.4, 34) | 14.5 (8.5-26) | 0.3589 |
| FEV1, L | 1.32 (0.43-3.01) | 1.15 (0.43-3.01) | 1.4 (0.55-1.99) | 0.2674 |
| %FEV1, % | 54 (25-132) | 53.9 (25-131) | 66.5 (34.9-132) | 0.1964 |
| FVC, L | 2.14 (0.6-4.7) | 2.58 (0.6-4.7) | 1.56 (0.79-2.96) | 0.0057 |
| %FVC, % | 77 (31-134) | 79 (31.2-134) | 64.5 (39-129) | 0.2184 |
| FEV1%, % | 66.1 (23-97) | 62 (23-88) | 73.1 (61-97) | 0.0084 |
| TV, L | 0.66 (0.21-1.4) | 0.72 (0.21-1.4) | 0.48 (0.37-1.17) | 0.176 |
| Oxygen flow rate at rest | 2 (0-5) | 2 (0-5) | 2 (0-4) | 0.8577 |
| NPPV, n (%) | 8 | 6 | 2 | 0.6363 |
| IPAP, cmH2O | 9 (0-12) | 8 (0-12) | 11 (8-11) | 0.4478 |
| EPAP, cmH2O | 4 (4-8) | 4 (4-8) | 4 | 0.3291 |
| Admission, n (%) | 21 | 14 | 7 | 0.3606 |
| ∆EtCO2, mmHg) | 4 (-8, 16) | 0 (-8, 16) | 7.5 (-3, 13) | 0.0543 |
| ∆PvCO2, mmHg) | 5 (-5, 21) | 4 (-5, 21) | 7 (-1.7, 12) | 0.6926 |
| Underlying diseases, n (%) | ||||
| COPD | 31 (50.8) | 23 (37.7) | 8 (13.1) | 0.0143 |
| ILD | 17 (27.9) | 8 (13.1) | 9 (14.8) | 0.2379 |
| TBsq | 6 (9.8) | 5 (8.2) | 1 (1.6) | 0.2021 |
| Other disease | 7 (11.5) | 0 | 7 (11.5) | 0.0007 |
The data are presented as the median (range) or number (percentage). COPD: chronic obstructive pulmonary disease, EPAP: expiratory positive airway pressure, EtCO2: end-tidal carbon dioxide, FEV1: forced expiratory volume in one second, FVC: forced vital capacity, ILD: interstitial lung disease, IPAP: inspiratory positive airway pressure, NPPV: noninvasive positive pressure ventilation, PvCO2: partial pressure of venous carbon dioxide, TBsq: sequelae of tuberculosis, TV: tidal volume
Table 2.
Baseline Characteristics of Patients with COPD, ILD and Other Diseases.
| COPD (n=31) | ILD (n=17) | Other diseases (n=13) | p value | |
|---|---|---|---|---|
| Age, years | 75 (41-96) | 72 (54-87) | 74 (34-90) | 0.4985 |
| Sex, n (%) | ||||
| Male | 23 | 8 | 5 | 0.0444 |
| Female | 8 | 9 | 8 | |
| Smoking history, pack years | 50 (0-150) | 0 (0-160) | 0 | <0.0001 |
| Body mass index | 21 (14-28.2) | 18.9 (12-39) | 16.2 (14.1-19.5) | 0.0079 |
| EtCO2, mmHg | 34 (18-46) | 36 (27-50) | 42 (30-53) | 0.0083 |
| PvCO2, mmHg | 47 (25-75) | 46 (37-72) | 62 (44-73) | 0.0064 |
| PvCO2- EtCO2 | 13 (-1.4, 34) | 8.5 (8-28) | 19.5 (14-26) | 0.075 |
| FEV1, L | 1.14 (0.43-2.84) | 1.81 (0.65-3.01) | 0.77 (0.43-1.99) | 0.0128 |
| %FEV1, % | 54 (25-132) | 100 (34.9-131) | 46.9 (37.7-73) | 0.0026 |
| FVC, L | 2.6 (0.82-4.7) | 2.08 (0.98-3.39) | 0.93 (0.6-3.06) | 0.0015 |
| %FVC, % | 80.4 (40-134) | 85.4 (41-111) | 41.2 (31.2-65) | 0.0006 |
| FEV1%, % | 60.5 (23-84) | 78.5 (64.3-97) | 74.7 (53.2-88.9) | <0.0001 |
| TV, L | 0.72 (0.37-1.4) | 0.81 (0.28-1.18) | 0.45 (0.21-0.81) | 0.0052 |
| Oxygen flow rate at rest | 2 (0-5) | 2 (0-4) | 2 (1-4) | 0.8734 |
| NPPV, n (%) | 3 | 1 | 4 | 0.013 |
| IPAP | 10 (0-12) | 0 | 9.5 (8-11) | 0.2336 |
| EPAP | 4 (4-8) | 7 | 4 | 0.2401 |
| Admission, n (%) | 10 | 3 | 8 | 0.1815 |
| ∆EtCO2, mmHg | 1 (-8, -14) | -4 (-6, -4) | 7.5 (-2, -16) | 0.0755 |
| ∆PvCO2, mmHg | 6 (0-12) | -1.7 (-5, -3) | 7 (-1, -21) | 0.0716 |
The data are presented as the median (range) or number (percentage). COPD: chronic obstructive pulmonary disease, EPAP: expiratory positive airway pressure, EtCO2: end-tidal carbon dioxide, FEV1: forced expiratory volume in one second, FVC: forced vital capacity, ILD: interstitial lung disease, IPAP: inspiratory positive airway pressure, NPPV: noninvasive positive pressure ventilation, PvCO2: partial pressure of venous carbon dioxide, TV: tidal volume
The underlying diseases were COPD in 31 patients, ILD in 17, sequelae of tuberculosis in 6, and other lung disease in 7. No patient had chronic kidney disease. The median EtCO2 was 31 (18-41) mmHg and the median PvCO2 was 49 (25-75) mmHg. On pulmonary function testing, the patients had a median FEV1 of 1.32 (0.43-3.01) L, a %FEV1 of 54% (25-132%), an FVC of 2.14 (0.6-4.7) L, and a %FVC of 77% (31-134%). The patient characteristics are shown according to sex in Table 1. There were significant differences in the smoking history, FVC, FEV1%, and underlying diseases between the study groups.
There were significant correlations between EtCO2 and PvCO2 (r=0.63), FEV1 (r=-0.44), %FEV1 (r=-0.36), FVC (r=-0.54), and %FVC (r=-0.64) and between %FEV1 and %FVC (r=0.52; Fig. 3). A multivariate linear regression analysis was performed to determine if PvCO2 could be predicted on the basis of EtCO2, %FEV1, and %FVC. The findings were statistically significant for EtCO2 [regression coefficient beta, 0.63; 95% confidence interval (CI) 0.58-1.48, p＜0.001] and %FEV1 (regression coefficient beta, -0.3; 95% CI -0.22--0.02, p=0.0189) but not for %FVC (regression coefficient beta, -0.11; 95% CI -0.07-0.16, p=0.458). There was a significant correlation between EtCO2 and PvCO2 in patients with COPD (r=0.5) and in those with ILD (r=0.63; Fig. 4).
Figure 3.
Correlations of EtCO2 with pulmonary function test results. EtCO2 was measured in outpatients with chronic respiratory failure who were receiving long-term oxygen therapy. (A) There was a significant positive correlation of EtCO2 with PvCO2 (r=0.63) and (B) a significant negative correlation of EtCO2 with FEV1 (r=-0.44), (C) %FEV1 (r=-0.36), (D) FVC (r=-0.54), and (E) %FVC (r=-0.64). (F) There was a significant positive correlation of %FEV1 with %FVC (r=0.52). COPD: chronic obstructive pulmonary disease, EtCO2: end-tidal carbon dioxide, FEV1: forced expiratory volume in one second, FVC: forced vital capacity, ILD: interstitial lung disease, PvCO2: partial pressure of venous carbon dioxide
Figure 4.
Correlations of EtCO2 with PvCO2 in outpatients with COPD or ILD. There was a significant positive correlation of EtCO2 with PvCO2 in (A) outpatients with COPD (r=0.5) and (B) with ILD (r=0.63). COPD: chronic obstructive pulmonary disease, EtCO2: end-tidal carbon dioxide, ILD: interstitial lung disease, PvCO2: partial pressure of venous carbon dioxide
There were significant correlations between the PvCO2 and EtCO2 gradient and the body mass index (BMI; r=-0.35) and %FEV1 (r=-0.33); however, there was no significant correlation with the TV (r=-0.14) or %FVC (r=-0.08; Fig. 5). A multivariate linear regression analysis was performed to predict the PvCO2 and EtCO2 gradient based on the BMI and %FEV1. The results were statistically significant for the BMI (regression coefficient beta, -0.34; 95% CI -1.16, -0.09, p=0.0235) but not for %FEV1 (regression coefficient beta, -0.021; 95% CI -0.15, 0.02, p=0.1476).
Figure 5.
Correlations of the PvCO2 and EtCO2 gradient with the BMI and pulmonary function test results. (A) There was a significant negative correlation of the PvCO2 and EtCO2 gradient with the BMI (r=-0.35) and (B) %FEV1 (r=-0.33) and (C) no significant correlation of the PvCO2 and EtCO2 gradient with the TV (r=-0.14) or (D) %FVC (r=-0.08). EtCO2: end-tidal carbon dioxide, FEV1: forced expiratory volume in one second, FVC: forced vital capacity, PvCO2: partial pressure of venous carbon dioxide, TV: tidal volume
The median time interval between the first visit and the day of admission was 2 (1-24) months. The EtCO2 levels on the day of admission were significantly higher than those in the same patients when they were in a stable condition (p=0.0049; Fig. 6). Furthermore, there were significant correlations between ΔEtCO2 and ΔPvCO2 (r=0.4), TV (r=0.43), %FEV1 (r=-0.4), and %FVC (r=-0.45; Fig. 7) in the study population overall. However, there was no statistically significant difference between the EtCO2 level on the day of admission and that in a stable condition in the COPD and ILD groups.
Figure 6.
A comparison of EtCO2 between patients who were admitted to the hospital according to their respiratory status. The EtCO2 was significantly higher in patients with respiratory failure who were admitted immediately after measurement of EtCO2 than in the same patients when they were in a stable condition (p=0.0049). EtCO2: end-tidal carbon dioxide
Figure 7.
Correlations of ΔEtCO2 with ΔPvCO2 and pulmonary function test results. (A) There was a significant positive correlation of ΔEtCO2 with ΔPvCO2 (r=-0.4) and a significant negative correlation of ΔEtCO2 with the (B) TV (r=-0.43), (C) %FEV1 (r=-0.4), and (D) %FVC (r=-0.45). EtCO2: end-tidal carbon dioxide, FEV1: forced expiratory volume in one second, FVC: forced vital capacity, PvCO2: partial pressure of venous carbon dioxide, TV: tidal volume
The receiver-operating characteristic curve for EtCO2 predicting hypercapnia as PvCO2＞45 mmHg is shown in Fig. 8A. The area under the curve was 0.849 for EtCO2. The optimum EtCO2 cut-off point for identifying hypercapnia was 34 mmHg (sensitivity 79.1%, specificity 88.9%). The receiver-operating characteristic curve for EtCO2 predicting hypercapnia as PvCO2 ＞70 mmHg is shown in Fig. 8B. The area under the curve was 0.806, and the optimum cut-off point was 38 mmHg for EtCO2 (sensitivity 100%, specificity 69.1%).
Figure 8.
Ability of the receiver-operating characteristic curve of EtCO2 to predict hypercapnia. (A) A plot of the EtCO2 curve for the prediction of hypercapnia defined as PvCO2 >45 mmHg with an area under the curve of 0.849. (B) A plot of the EtCO2 curve for the prediction of hypercapnia defined as a PvCO2 of >70 mmHg with an area under the curve of 0.806.
Discussion
This is the first report on the value of EtCO2 as measured by the CapnoEye™ in patients with chronic respiratory failure receiving LTOT. There was a significant correlation of EtCO2 with PvCO2 and an association of the PvCO2 and EtCO2 gradient with BMI. The EtCO2 cut-off level of 34 mmHg was useful for predicting CO2 retention and deterioration of respiratory status in the outpatient department.
EtCO2 measurements obtained by a capnometer have been widely accepted as a sensitive method for reflecting the PaCO2 level in intubated and mechanically ventilated patients. However, few studies have investigated the usefulness of EtCO2 in patients who are breathing spontaneously (9-11). One of the reported advantages of capnometry is its ability to obtain a reliable estimate of EtCO2 during a vital capacity maneuver in patients with chronic respiratory disease who are breathing spontaneously (9). In the present study, the EtCO2 value was measured using the CapnoEye™ during a TV maneuver and compared with that obtained during a vital capacity maneuver; the TV maneuver is easy for patients with respiratory failure to perform because it needs only spontaneous breathing.
It is well known that PaCO2 is the gold standard for the evaluation of hypercapnia; however, PvCO2 has also been reported to have good concordance with PaCO2 and to be a reliable, feasible, and safe alternative to repeated ABG analyses in patients with severe hypoxemic and/or hypercapnic respiratory failure (12). Therefore, PvCO2 was thought to be a useful surrogate marker of PaCO2 for the evaluation of hypercapnia in the present study.
The correlation of EtCO2 and PaCO2 has been reported to be unreliable in some clinical situations, and no correlation was found between EtCO2 and PaCO2 when the physiologic dead space was substantially elevated (13-19). Physiologic dead space ventilation is the sum of the anatomic dead space from the conducting airways and the alveolar dead space arising from a disease process and/or therapy.
The EtCO2 level is normally 5 mmHg lower than that of PaCO2 because of the mixing of CO2 containing alveolar gas and gas devoid of CO2 from the anatomic dead space (20). In patients with lung disease, the additional alveolar dead space further dilutes the EtCO2 relative to the PaCO2 (20). For example, the normal ratio of the physiologic dead space to TV (VD/VT) is 0.2-0.35 (21) but has been found to be 0.4-0.55 in patients with acute lung injury as a result of the additional alveolar dead space (22).
In the present study, the BMI was the only factor found to have a negative effect on the PvCO2 and EtCO2 gradient in a multivariate analysis, with no significant effect of TV. These findings are similar to those of Ickx et al., who found that the PaCO2 and EtCO2 gradient was negatively correlated with weight in small children and attributed this finding to the high ratio of dead space to TV (VD/VT) (23). Therefore, in the present study, the PvCO2 and EtCO2 gradient may have been correlated with dead space.
In general, in patients with COPD, progressive airflow limitations and emphysematous destruction were considered to increase the alveolar dead space, which might have caused the inequality in the ventilation/perfusion (V/Q) ratio. However, Sandek et al. found no significant correlation between the air flow obstruction measured by spirometry and the V/Q ratio (24). In the present study, %FEV1 was not a significant independent predictor of the PvCO2 and EtCO2 gradient. This finding is consistent with a previous speculation that inequality in the V/Q ratio might be buffered by underlying pathophysiological processes, including hypoxic vasoconstriction and active collateral ventilation (24, 25).
The positive mechanical pressure used when measuring EtCO2 in intubated and mechanically ventilated patients might increase the ventilation of the atelectatic lobe and reduce intravascular pulmonary fluid iatrogenically, leading to a V/Q mismatch and additional alveolar dead space (26). The mechanical dead space added by the endotracheal tube might further increase the PvCO2 and EtCO2 gradient. In our study, the exclusion of mechanically ventilated patients resulted in a reduction in dead space ventilation and might explain the strong correlation found between PvCO2 and EtCO2 even in patients with chronic respiratory failure.
EtCO2 was correlated with both FEV1 and FVC in this study. There has been a similar report of a significant correlation of FEV1 with EtCO2 measured using a capnometer (11, 12); however, to our knowledge, this is the first report of a correlation between EtCO2 and FVC. Although the mechanism underlying the elevation of EtCO2 in response to decreases in FEV1 remains unclear, a reduced FEV1 value is associated with the severity of obstructive lung disease, and obstruction of the terminal bronchi has been cited as a reason for hypoventilation of the alveoli and a cause of CO2 retention (27, 28). In the present study, FVC was strongly correlated with FEV1, so FEV1 may be a confounding factor that influenced the correlation between EtCO2 and FVC.
EtCO2 has previously been reported to be correlated with PCO2 in patients with COPD (9); however, there has been no study of the correlation between EtCO2 and PCO2 in patients with ILD. In this study, EtCO2 was confirmed to be correlated with PvCO2 not only in patients with COPD but also in those with ILD. This finding highlights the value of EtCO2 for predicting hypercapnia in both obstructive and restrictive lung disease.
EtCO2 on the day of admission was significantly higher than that recorded when the patients were in a stable condition. The ΔEtCO2 was correlated with the ΔPvCO2, suggesting that elevation of EtCO2 may be a sign of an increasing PvCO2 level over time. The patients in our study were in an unstable respiratory condition when they were admitted and unable to undergo respiratory function testing. However, EtCO2 was measured successfully in these patients using the CapnoEye™ because the protocol required for the TV maneuver is much easier to perform than that for the VC maneuver, regardless of the respiratory status. Furthermore, the FEV1, FVC, and TV values measured when the patients were in a stable condition were negatively correlated with the ΔEtCO2 and ΔPvCO2. Therefore, we assume that the elevation of the CO2 level was easier to observe in patients with an impaired lung function than in those with a preserved lung function. The regular measurement of EtCO2 on an outpatient basis using the CapnoEye™ may help detect deterioration of the respiratory status and the need for hospital admission in patients with respiratory failure.
It is well known that CO2 retention depresses awareness, even to the point of total loss of consciousness, so monitoring the blood CO2 level is an important component of the patient assessment. Previous studies that used PvCO2 to screen for potential hypercapnia found that it had a sensitivity of 79% when a cut-off point of ＞45 mmHg was used (29). Therefore, the screening cut-off in the present study was defined as a PvCO2 of ＞45 mmHg, and the optimum cut-off point for hypercapnia was set at an EtCO2 of ＞34 mmHg. Furthermore, the elevation of PaCO2 to ＞70-75 mmHg has been reported to reduce the level of awareness (30). An EtCO2 of ＞38 mmHg was shown to be a possible biomarker of a PvCO2＞70 mmHg. By determining the cut-off point for EtCO2, the evaluation of the increase in EtCO2 over time may be able to predict the exacerbation of respiratory failure, thereby allowing for the early treatment and avoidance of admission.
Several limitations associated with the present study warrant mention, including its retrospective design, single-institution setting, small sample size, and inclusion of patients who needed differing oxygen flow rates. Furthermore, PvCO2 was measured instead of PaCO2, and the difference between the venous and arterial CO2 levels may be influenced by the cardiac output and tissue consumption; therefore, our ability to evaluate the correlation between EtCO2 and PaCO2 accurately was limited. Other limitations that restricted our ability to evaluate the usefulness of EtCO2 in patients with a deteriorating respiratory status included the lack of clinical data besides EtCO2 and PvCO2 in patients who required hospital admission. Finally, the CapnoEye™ can only evaluate EtCO2 in patients who are alert and breathing spontaneously and is thus unsuitable for use with patients who are unconscious.
Conclusion
This is the first study to demonstrate the correlation of EtCO2 and PvCO2 with the pulmonary function in spontaneously breathing patients with chronic respiratory failure. Repeated EtCO2 measurements were obtained noninvasively by the CapnoEye™. The CapnoEye™ was convenient and useful for estimating PvCO2. The BMI was identified as a possible predictor of the PvCO2 and EtCO2 gradient. An increase in EtCO2 to ＞34 mmHg may indicate the deterioration of the respiratory status in patients with chronic respiratory failure receiving LTOT.
The authors state that they have no Conflict of Interest (COI).
References
- 1. Vogelmeier CF, Criner GJ, Martinez FJ, et al. Global strategy for the diagnosis, management and prevention of chronic obstructive lung disease 2017 report: GOLD executive summary. Respirology 22: 575-601, 2017. [DOI] [PubMed] [Google Scholar]
- 2. Miyamoto K, Aida A, Nishimura M, Aiba M, Kira S, Kawakami Y. Gender effect on prognosis of patients receiving long-term home oxygen therapy. The Respiratory Failure Research Group in Japan. Am J Respir Crit Care Med 152: 972-976, 1995. [DOI] [PubMed] [Google Scholar]
- 3. Visca D, Mori L, Tsipouri V, et al. Effect of ambulatory oxygen on quality of life for patients with fibrotic lung disease (AmbOx): a prospective, open-label, mixed-method, crossover randomised controlled trial. Lancet Respir Med 6: 759-770, 2018. [DOI] [PubMed] [Google Scholar]
- 4. Kida K, Motegi T, Ishii T, Hattori K. Long-term oxygen therapy in Japan: history, present status, and current problems. Pneumonol Alergol Pol 81: 468-478, 2013. [PubMed] [Google Scholar]
- 5. Celli BR, MacNee W, Agusti A, et al. Standards for the diagnosis and treatment of patients with COPD: a summary of the ATS/ERS position paper. Eur Respir J 23: 932-946, 2004. [DOI] [PubMed] [Google Scholar]
- 6. Dogan NO, Sener A, Gunaydin GP, et al. The accuracy of mainstream end-tidal carbon dioxide levels to predict the severity of chronic obstructive pulmonary disease exacerbations presented to the ED. Am J Emerg Med 32: 408-411, 2014. [DOI] [PubMed] [Google Scholar]
- 7. Yamanaka MK, Sue DY. Comparison of arterial-end-tidal PCO2 difference and dead space/tidal volume ratio in respiratory failure. Chest 92: 832-835, 1987. [DOI] [PubMed] [Google Scholar]
- 8. Vellinga A, Cormican M, Hanahoe B, Bennett K, Murphy AW. Opt-out as an acceptable method of obtaining consent in medical research: a short report. BMC Med Res Methodol 11: 40, 2011. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9. Takano Y, Sakamoto O, Kiyofuji C, Ito K. A comparison of the end-tidal CO2 measured by portable capnometer and the arterial PCO2 in spontaneously breathing patients. Respir Med 97: 476-481, 2003. [DOI] [PubMed] [Google Scholar]
- 10. Corbo J, Bijur P, Lahn M, Gallagher EJ. Concordance between capnography and arterial blood gas measurements of carbon dioxide in acute asthma. Ann Emerg Med 46: 323-327, 2005. [DOI] [PubMed] [Google Scholar]
- 11. Cinar O, Acar YA, Arziman I, Kilic E, Eyi YE, Ocal R. Can mainstream end-tidal carbon dioxide measurement accurately predict the arterial carbon dioxide level of patients with acute dyspnea in ED. Am J Emerg Med 30: 358-361, 2012. [DOI] [PubMed] [Google Scholar]
- 12. Horvath CM, Brutsche MH, Baty F, Rudiger JJ. Transcutaneous versus blood carbon dioxide monitoring during acute noninvasive ventilation in the emergency department - a retrospective analysis. Swiss Med Wkly 146: w14373, 2016. [DOI] [PubMed] [Google Scholar]
- 13. Grenier B, Verchere E, Mesli A, et al. Capnography monitoring during neurosurgery: reliability in relation to various intraoperative positions. Anesth Analg 88: 43-48, 1999. [DOI] [PubMed] [Google Scholar]
- 14. Short JA, Paris ST, Booker PD, Fletcher R. Arterial to end-tidal carbon dioxide tension difference in children with congenital heart disease. Br J Anaesth 86: 349-353, 2001. [DOI] [PubMed] [Google Scholar]
- 15. Russell GB, Graybeal JM. Reliability of the arterial to end-tidal carbon dioxide gradient in mechanically ventilated patients with multisystem trauma. J Trauma 36: 317-322, 1994. [DOI] [PubMed] [Google Scholar]
- 16. Kerr ME, Zempsky J, Sereika S, Orndoff P, Rudy EB. Relationship between arterial carbon dioxide and end-tidal carbon dioxide in mechanically ventilated adults with severe head trauma. Crit Care Med 24: 785-790, 1996. [DOI] [PubMed] [Google Scholar]
- 17. Tobias JD, Meyer DJ. Noninvasive monitoring of carbon dioxide during respiratory failure in toddlers and infants: end-tidal versus transcutaneous carbon dioxide. Anesth Analg 85: 55-58, 1997. [DOI] [PubMed] [Google Scholar]
- 18. Berkenbosch JW, Lam J, Burd RS, Tobias JD. Noninvasive monitoring of carbon dioxide during mechanical ventilation in older children: end-tidal versus transcutaneous techniques. Anesth Analg 92: 1427-1431, 2001. [DOI] [PubMed] [Google Scholar]
- 19. McDonald MJ, Montgomery VL, Cerrito PB, Parrish CJ, Boland KA, Sullivan JE. Comparison of end-tidal CO2 and PaCO2 in children receiving mechanical ventilation. Pediatr Crit Care Med 3: 244-249, 2002. [DOI] [PubMed] [Google Scholar]
- 20. Sullivan KJ, Kissoon N, Goodwin SR. End-tidal carbon dioxide monitoring in pediatric emergencies. Pediatr Emerg Care 21: 327-332. [DOI] [PubMed] [Google Scholar]
- 21. Sun XG, Hansen JE, Garatachea N, Storer TW, Wasserman K. Ventilatory efficiency during exercise in healthy subjects. Am J Respir Crit Care Med 166: 1443-1448, 2002. [DOI] [PubMed] [Google Scholar]
- 22. Kallet RH, Alonso JA, Pittet JF, Matthay MA. Prognostic value of the pulmonary dead-space fraction during the first 6 days of acute respiratory distress syndrome. Respir Care 49: 1008-1014, 2004. [PubMed] [Google Scholar]
- 23. Jacques Olivier BI. Arterial to end-tidal carbon dioxide tension differences in infants and children. J Anesth Clin Res 6: 2, 2015. [Google Scholar]
- 24. Sandek K, Bratel T, Lagerstrand L, Rosell H. Relationship between lung function, ventilation-perfusion inequality and extent of emphysema as assessed by high-resolution computed tomography. Respir Med 96: 934-943, 2002. [DOI] [PubMed] [Google Scholar]
- 25. Rodriguez-Roisin R, Drakulovic M, Rodriguez DA, Roca J, Barbera JA, Wagner PD. Ventilation-perfusion imbalance and chronic obstructive pulmonary disease staging severity. J Appl Physiol 106: 1902-1908, 2009. [DOI] [PubMed] [Google Scholar]
- 26. Kim CK, Heyman S. Ventilation/perfusion mismatch caused by positive pressure ventilatory support. J Nucl Med 30: 1268-1270, 1989. [PubMed] [Google Scholar]
- 27. Krauss B, Deykin A, Lam A, et al. Capnogram shape in obstructive lung disease. Anesth Analg 100: 884-888, 2005. [DOI] [PubMed] [Google Scholar]
- 28. Kodali BS. Capnography outside the operating rooms. Anesthesiology 118: 192-201, 2013. [DOI] [PubMed] [Google Scholar]
- 29. Kelly AM, Kerr D, Middleton P. Validation of venous pCO2 to screen for arterial hypercarbia in patients with chronic obstructive airways disease. J Emerg Med 28: 377-379, 2005. [DOI] [PubMed] [Google Scholar]
- 30. Refsum HE. Relationship between state of consciousness and arterial hypoxaemia and hypercapnia in patients with pulmonary insufficiency, breathing air. Clin Sci 25: 361-367, 1967. [PubMed] [Google Scholar]