|Year : 2019 | Volume
| Issue : 3 | Page : 768-772
Noninvasive assessment of arterial CO2 from end-tidal CO2 in pediatric intensive care unit of Al-Zahraa Hospital
Ragaa Abdel-Salam1, Soheir I Mohamed1, Sally A.-E.M Ezz-Elarab2
1 Department of Pediatrics, Faculty of Medicine, Al-Azhar University, Cairo, Egypt
2 Department of Pediatrics, Dar Ismail Hospital, Alexandria, Egypt
|Date of Submission||24-Nov-2019|
|Date of Decision||24-Nov-2019|
|Date of Acceptance||17-Dec-2019|
|Date of Web Publication||10-Feb-2020|
Assistant Professor Ragaa Abdel-Salam
Pediatric, Department, Faculty of Medicine, Al-Azhar University Cairo
Source of Support: None, Conflict of Interest: None
Background Arterial carbon dioxide tension (PaCO2) is considered to be the gold standard for accurate monitoring in pediatric ICU; however, it is invasive, costly, and intermittently gives snapshots about the patient status.
Objective The aim was to use end-tidal carbon dioxide (ETCO2) as an effective, persistent, and non-invasive monitor of arterial CO2.
Patients and methods This observational study was conducted on 50 mechanically ventilated children aged from 1 to 5 years in pediatric ICU of Al-Zahraa University Hospital. PaCO2 and ETCO2 were recorded at the same time, and the results were analyzed for correlation and agreement. Lung disease severity was measured by ventilation index (VI) and PaO2/FiO2 (P/F) ratio.
Results This study showed that the mean PaCO2 was higher than ETCO2 value (31.32±14.49 and 29.79±13.81 mmHg, respectively). The mean difference between PaCO2 and ETCO2 was 1.8±3.23 mmHg. A positive correlation was present between both PaCO2 and ETCO2 (correlation coefficient r=0.969, P<0.001, and 95% confidence interval=0.946–0.982). A positive correlation was found between PaCO2 and ETCO2 in mild and moderate lung disease, where P/F ratio is greater than 200 (n=39, r=0.961, P<0.001) and VI is less than 20 (n=38, r=0.885, P<0.001), and also in severe lung disease, where P/F ratio is less than 200 (n=12, r=0.991, P<0.001) and VI is greater than 20 (n=12, r=0.988, P<0.001).
Conclusion This study suggests a significant correlation between PaCO2 and ETCO2. ETCO2 monitoring exhibited a good validity to predict PaCO2 in critically ill children
Keywords: capnography, CO2, critically ill child, ETCO2, noninvasive
|How to cite this article:|
Abdel-Salam R, Mohamed SI, Ezz-Elarab SA. Noninvasive assessment of arterial CO2 from end-tidal CO2 in pediatric intensive care unit of Al-Zahraa Hospital. Sci J Al-Azhar Med Fac Girls 2019;3:768-72
|How to cite this URL:|
Abdel-Salam R, Mohamed SI, Ezz-Elarab SA. Noninvasive assessment of arterial CO2 from end-tidal CO2 in pediatric intensive care unit of Al-Zahraa Hospital. Sci J Al-Azhar Med Fac Girls [serial online] 2019 [cited 2020 Oct 24];3:768-72. Available from: http://www.sjamf.eg.net/text.asp?2019/3/3/768/278028
| Introduction|| |
Oxygenation and ventilation are different physiologic functions that must be evaluated in both intubated and spontaneously breathing patients. Pulse oximetry gives immediate feedback about oxygenation. Capnography gives immediate information about ventilation, perfusion, and metabolism .
The term capnography points to the noninvasive measurement of the partial pressure of CO2 in expired breath expressed as the CO2 concentration over time . Capnography is a helpful tool during mechanical ventilation (MV), confirming endotracheal tube position, allowing monitoring in the operating room setting and monitoring of the completeness of the ventilator circuit for early discovery of accidents such as inadvertent extubation .
ETCO2 represents the partial pressure or maximal concentration of CO2 at the end of an expired breath and is expressed as a percentage of CO2 or mmHg. The target values are 5–6% CO2, which is equal to 35–45 mmHg . ETCO2 is achieved by capnometry involving continuous quantitative measurements of exhaled air. The measured CO2 concentration (expressed in mmHg) over time is then displayed numerically (Capnometry) and/or graphically (Capnography) .
Arterial blood gas (ABG) analysis sits as the ‘gold standard’ for the assessment of ventilation in critically ill patients. However, accurate, noninvasive monitoring of arterial CO2 values (PaCO2) would be a great asset in management of these patients. Such monitoring may limit the requirement for frequent, expensive, and painful ABG analyses .
There are two forms of capnographs: the first is sidestream capnograph and the second is mainstream capnograph. In sidestream capnography, the sensor is remote from the patient and is not within the ventilator circuit, which makes it useful for both intubated and nonintubated patients. In mainstream systems, air sample from the expired breath is aspirated from the circuit through a sampling line and then delivered to the sensor. The line can be connected by a T-piece to an ETT .
| Patients and methods|| |
This observational study was carried out in the pediatric ICU in Al-Zahraa University Hospital. Informed consent was obtained from the participating parents in adherence with the Ethical Committee of Al Zahraa Hospital Al-Azhar University. A total of 50 children aged from 1 to 5 years, all tracheally intubated and MV, were included, whereas patients with air leak syndromes, cyanotic congenital heart disease, or right to left intracardiac shunt were excluded.
Diagnostic and demographic data were recorded for each patient. Full history taking was done for all patients, including, sex, age, weight, underlying disease diagnosis, and full clinical examination. Moreover, laboratory diagnosis, including complete blood count, liver and renal functions, and ABG was done. Overall, 22 males and 28 females were recorded in the study, with an age range between 1 and 5 years. Each patient received standard critical care monitoring including ETCO2 measurement. This value was obtained via dedicated flow/CO2 sensor (Philips Capnostat with Flow/CO2 sensor; Philips Capnostat with Flow/CO2 sensor; Philips Medical Sysrtems, 3000 minuteman Road Andover, MA, USA). The Y-shaped adaptor is placed at the end of endotracheal tube and joined to the monitor. ETCO2 was recorded simultaneously while ABG was obtained. Data collected at the time of ABG analysis included peak inspiratory pressure (PIP), positive end expiratory pressure (PEEP), ventilation index (VI), mean pressure, and vital signs, including respiratory rate (RR), heart rate (HR), temperature, and systemic blood pressure. From the data collected, VI was calculated by the equation, (PaCO2×PIP×RR)/1000. PaO2/FiO2 was defined as P/F ratio. Oxygenation index (OI) was calculated by equation (FiO2×mean airway pressure/PaO2).
Data were fed to the computer and analyzed using IBM SPSS software package version 20.0. (IBM Corp., Armonk, New York, USA). Qualitative data were described using number and percent. Quantitative data were expressed as range (minimum and maximum), mean, SD, median, and interquartile range (IQR). Significance of the obtained results was judged at the 5% level. Analyses of ETCO2 and PaCO2 were done by computing Spearman test, and correlation coefficient (r), 95% confidence intervals (CI), and coefficient of determination (R2) were calculated. A linear regression analysis was done to find the equation between PaCO2 and ETCO2.
| Results|| |
The study group consisted of 50 Egyptian children aged between 1 and 5 years with median HR and RR of 144 beats/min and 25 cycle/min, respectively, whereas the median systolic blood pressure was 100, with IQR of 90–100, and that of diastolic blood pressure was 60, with IQR of 55–70. There were 22 males and 28 females. In this study, it was found that the z score for weight ranged from −2.36 to 1.96. The commonest diagnosis at admission was respiratory diseases (40%) followed by neurological diseases (30%); cardiovascular diseases (10%); others (10%) in the form of burn, DKA, and acute myeloid leukemia; then toxicological and road traffic accidents (8% for each); and sepsis (2%) ([Table 1]). [Table 2] shows no statistical difference between both PaCO2 and ETCO2 regarding respiratory and nonrespiratory cases. Findings of ABG and ETCO2 and the difference between PaCO2 and ETCO2 are shown in [Table 3]. We observed an excellent correlation between PaCO2 and ETCO2 in studied children. On the contrary, regarding the severity of lung disease, we found an excellent correlation between ETCO2 and PaCO2 ([Table 4] and [Table 5]). [Table 6] shows that for cases with hypocarbia (PaCO2 <30 mmHg), the correlating ETCO2 according to linear regression equation was also 30 mmHg, with sensitivity of 80%, a specificity of 100%, positive predictive value (PPV) of 100%, and negative predictive value (NPV) 87.88%, and with hypercarbia (PaCO2 >50 mmHg), the corresponding ETCO2 was greater than 45 mmHg by linear regression, with sensitivity of 88%, specificity 95%, PPV 86%, and NPV of 97%.
|Table 2 Relation between the diagnosis and both PaCO2 and ETCO2 regarding respiratory or nonrespiratory cases|
Click here to view
|Table 3 Arterial blood gas results, ETCO2, and the difference between PaCO2 and ETCO2 of the studied cases|
Click here to view
|Table 5 The relation between ETCO2 and PaCO2 regarding the severity of lung disease|
Click here to view
|Table 6 Sensitivity and specificity for ETCO2 in cases of hypocarbia and hypercarbia|
Click here to view
| Discussion|| |
CO2 is produced in the tissues as a byproduct of aerobic metabolism and then transported to the lungs by the venous circulation and is finally eliminated during the expiratory phase of ventilation. End-tidal CO2 (ETCO2) is defined as the peak CO2 value during expiration and is dependent on adequate pulmonary blood flow to the ventilating areas of the lung. ETCO2 in healthy subjects differs normally by less than 5 mmHg from the arterial CO2 .
Capnography or ETCO2 assessment has a variety of uses in pediatric ICU setting. The ability to regularly measure exhaled CO2 can give vital information about airway breathing and circulation in pediatric patients who are critically ill .
E CO2 is a reliable tool for continuous and noninvasive monitoring of CO2, avoiding frequent ABG samples and avoiding blood loss from the patient, and thereby decreasing the complications, including needle sticks, and decreasing the cost .
A total of 22 males and 28 females were involved in this study, with an age range between 1 and 5 years. In the current study, it was found that the z score for weight ranged from −2.36 to 1.96.
Although the commonest diagnosis at admission was respiratory diseases (40%), there was no statistical significance regarding the diagnosis between the p ETCO2 and PaCO2 (P=0.256 and 0.268, respectively). these results come in agreement with Machdonald et al.  and Goonasekera et al. , who demonstrated that PaCO2-pETCO2 difference was not significantly influenced by disorders leading to MV (P<0.6), and it was individual specific.
The analysis of ABG showed that the mean PaCO2, PaO2, and O2 saturation were 31.32±14.49, 84.92±33.47, and 95.04±4.50, respectively, whereas the mean value of ETCO2 was 29.79±13.81, and the difference between PaCO2 and ETCO2 ranged between −5 and 8 mmHg, with mean difference of 1.8±3.23 mmHg. These results were in agreement with Panigrahi et al. , who showed that the mean PaCO2 was higher than mean ETCO2 (28.9±9.7 and 27.8±9.6 mmHg, respectively), and the mean difference between PaCO2 and ETCO2 was 1.6±3.23 mmHg. Moreover, Wu et al. , showed that ETCO2 was less than the corresponding PaCO2 in infants (2–3 mmHg) who demonstrated a good correlation between ETCO2 and paCO2.
In contrast to our study, Amuchou and Singhal  reported that ETCO2 was significantly lower than the corresponding PaCO2 values (37.3±7.8 and 43.5±9.6 mmHg, respectively, P<0.001).
In this study, an excellent correlation was found between PaCO2 and ETCO2 (r=0.969, narrow 95% CI=0.946 to 0.982, and R2=0.939; P=0.001). These results were in agreement with Mehta et al. , who demonstrated that there is an excellent correlation between PaCO2 and ETCO2, with r=0.914, CI=0.87–0.94, and R2=0.83.
Trevisanuto et al. , observed that on very low birth weight infants a positive correlation between PaCO2 and ETCO2, the correlation was r=0.78, P less than 0.0001 with coefficient=0.84±0.07, P less than 0.0001 which showed that any change in PaCO2 was associated with positive change in ETCO2.
Lin et al.  reported in their study, which was done on preterm infants, that there was a moderate correlation between ETCO2 and PaCO2 values (r=0.603, P<0.01). The correlation was higher in the group that received the surfactant treatment (r=0.786, P<0.01) than those before surfactant treatment (r=0.235).
In contrast to our results, Jacob et al. reported a poor correlation in neonates with pulmonary disease. This might be explained by that the patient’s tidal volume is too small to deliver alveolar gas to the capnograph and so the ETCO2 will be falsely low.
In this study, a positive correlation between lung disease severity and PaCO2 and ETCO2 was found in mild to moderate lung disease patients with P/F>200, and the correlation coefficient between PaCO2 and ETCO2 was r=0.961 (95% CI=932–977), and in severe lung disease with P/F was less than 200, the correlation coefficient was r=0.991 (95% CI=0.985–0.994).
Panigrahi et al.  found that in cases of mild to moderate lung disease with P/F greater than 200, the correlation coefficient between ETCO2 and PaCO2 was r=0.861 and in cases of severe lung disease with P/F less than 200, the correlation coefficient was r=0.932.
Similarly, in a study done by Mehta et al. , a strong correlation was shown in patients with P/F ratio greater than 200, with r=0.094, 95%CI=0.091–0.95, and R2=0.88, whereas in patients with severe lung disease, where P/F less than 200, the correlation was good (r=0.782, R2=0.68). This difference can be explained as ETCO2 measurements are affected by alveolar CO2 levels and dead space and pulmonary perfusion.
In this study, a statistical significant positive correlation between the VI and ETCO2 and PaCO2, with p value of 0.001, with normal VI less than 20 having a strong correlation (r=0.885, 95% CI=0.805–0.933) and with abnormal VI greater than 20 having an excellent correlation (r=0.988, 95%CI=0.979–0.993). These results come in agreement with Macdonald et al. , who showed a strong positive correlation (r=0.716) between ETCO2 and PaCO2, with influence of both P/F ratio and VI on correlation of PaCO2 and ETCO2.
In this study, for hypocarbia, represented as PaCO2 less than or equal to 30 mmHg, the correlating ETCO2 according to linear regression equation was also 30 mmHg. The data showed that ETCO2 of less than or equal to 30 mmHg had sensitivity of 80%, a specificity of 100%, PPV of 100%, and NPV of 87.88%. Moreover, for hypercarbia represented with PaCO2 level of greater than 50 mmHg, the corresponding ETCO2 was 45 mmHg by linear regression, and ETCO2 of greater than or equal to 45 mmHg has sensitivity of 88%, specificity of 95%, PPV of 86%, and NPV of 97%.
McDonald et al. , defined hypocarbia as a PaCO2 of less than 30 mmHg, and the correlating ETCO2 was 20 mmHg, with sensitivity of 67% a specificity of 90%, and a PPV of 43%, and when they chose hypercarbia, with PaCO2 greater than 50 mmHg, the corresponding ETCO2 was 45 mmHg, with a sensitivity of 83%, specificity of 88%, PPV 77%, and NPV 91%.
| Conclusion|| |
There was a positive correlation between ETCO2 and PaCO2 in ventilated children admitted in the pediatric ICU, so the use of ETCO2 monitoring exhibited a good validity to predict PaCO2 in critically ill children and to decrease the number of ABGs.
Financial support and sponsorship
Conflicts of interest
There are no conflicts of interest.
| References|| |
Truchot J, Gayet A-R, Philippon A-L, Chauvin A, Malka J, Vicaut E, Plaisance P. End tidal carbon dioxide monitoring in acute asthma. Eur J Emerg Med
Heather B, Melissa L. Capnography In The Pediatric Emergency Department: clinical applications. Pediatr Emerg Med Pract
McSwain S, Hamel DS, Smith B, Gentile MA, Srinivasan S, Meliones JN, Cheifetz IM. End-tidal and arterial carbon dioxide measurements correlate across all levels of physiologic dead space. Respir Care
Coté CJ, Wilson S. Guidelines for monitoring and management of pediatric patients before, during, and after sedation for diagnostic and therapeutic procedures: update 2016. Pediatr Dent
Selby ST, Abramo T, Hobart-Porter N. An update on end-tidal CO2 monitoring. Pediatr Emerg Care
Baudin F, Bourgoin P, Brossier D, Essouri S, Emeriaud G, Wysocki M, Jouvet P. Noninvasive estimation of arterial CO2
from end-tidal CO2
in mechanically ventilated children: the GRAeDIENT Pilot Study. Pediatr Crit Care Med
Totapally B. Utility of end-tidal carbon dioxide monitoring in critically ill children. Indian J Crit Care Med
Bailly DK, Reeder RW, Zabrocki LA, Hubbard AM, Wilkes J, Bratton SL, Thiagarajan RR. Development and validation of a score to predict mortality in children undergoing extracorporeal membrane oxygenation for respiratory failure. Crit Care Med
2017; 45: e58–e66.
Goonasekera C, Wang Y, Goodwin A, Goodman J, Deep A. Arterial and end-tidal carbon dioxide difference in pediatric intensive care. Indian J Crit Care Med
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
Panigrahi N, Ahmad M, Mishra M, Basavaraja GV. Predictive capability of end tidal carbon dioxide and its correlation with arterial carbon dioxide in mechanically ventilated children. J Pediatr Crit Care
Wu CH, Chou HC, Hsieh WS, Chen WK, Huang PY, Tsao PN. Good estimation of arterial carbon dioxide by end-tidal carbon dioxide monitoring in the neonatal intensive care unit. Pediatr Pulmonol
Amuchou Singh S, Singhal N. Dose end-tidal carbon dioxide measurement correlate with arterial carbon dioxide in extremely low birth weight infants in the first week of life? Indian Pediatr
Mehta H, Trivedi S, Kashyap R. Correlation of end tidal and arterial carbon dioxide levels in critically Ill neonates and children. Indian J Crit Care Med
Trevisanuto D, Giuliotto S, Cavallin F, Doglioni N, Toniazzo S, Zanardo V. End-tidal carbon dioxide monitoring in very low birth weight infants: Correlation and agreement with arterial carbon dioxide. Pediatr Pulmonol
Lin H-J, Huang C-T, Hsiao H-F, Chiang M-C, Jeng M-J. End-tidal carbon dioxide measurement in preterm infants with low birth weight. PloS One
Jacob R, Nelkenbauma A, Merrick, Brik R. Comparison of end-tidal CO2 and Paco2 in children receiving mechanical ventilation. Pediatr Crit Care Med
[Table 1], [Table 2], [Table 3], [Table 4], [Table 5], [Table 6]