This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Email this article to a colleague Similar articles in this journal Similar articles in Web of Science Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Permissions Citing Articles Citing Articles via HighWire Citing Articles via Web of Science (19) Citing Articles via Google Scholar Google Scholar Articles by Tobias, J. D. Articles by Meyer, D. J. Search for Related Content PubMed PubMed Citation Articles by Tobias, J. D. Articles by Meyer, D. J.

---

PEDIATRIC ANESTHESIA

Transcutaneous Monitoring of Carbon Dioxide Tension After Cardiothoracic Surgery in Infants and Children

Joseph D. Tobias, MD^*,,§, William R. Wilson, Jr., MD^*,,||, and D. Joseph Meyer, MD, PhD^*,

Departments of *Child Health, Anesthesiology, Surgery, and the Divisions of §Pediatric Critical Care/Pediatric Anesthesiology and ||Pediatric Cardiothoracic Surgery, The University of Missouri, Columbia, Missouri

Address correspondence and reprint requests to Joseph D. Tobias, MD, Department of Child Health, The University of Missouri, M658 Health Sciences Center, One Hospital Dr., Columbia, MO 65212. Address e-mail to Joseph_Tobias{at}muccmail.missouri.edu

	Abstract

Top Abstract Introduction Methods Results Discussion References

 
In this prospective investigation, we evaluated the efficacy and accuracy of transcutaneous monitoring of CO₂ (TC-CO₂) in infants and children after cardiothoracic surgery. Cardiothoracic surgery patients whose ETCO₂ and arterial CO₂ values did not correlate (gradient 5 mm Hg) during the first postoperative hour underwent placement of the TC electrode (30 of 33 patients). If the TC-CO₂ to arterial difference was 5 mm Hg, the TC-CO₂ electrode was recalibrated and reapplied on another site. If the discrepancy was still 5 mm Hg, the case was considered a clinical failure and no further data were collected (3 of 30 patients). If the arterial to TC gradient was <5 mm Hg, the patient was included in the data collection (27 of 30 patients). One to five sample sets (TC and arterial CO₂) were collected from these patients. Statistical analysis included linear regression analysis and Bland-Altman analysis. The cohort for the study included 27 patients ranging in age from 2 days to 9 yr and in weight from 3.2 to 25 kg. A total of 101 sample sets were analyzed. The mean ± SD absolute difference between the TC-CO₂ and arterial CO₂ was 1.7 ± 1.4 mm Hg (range 0–9 mm Hg). The TC-CO₂ to arterial CO₂ difference was 0–2 mm Hg in 82 of 101 values (81%), 3–5 mm Hg in 18 of 101 values (18%), and >6 mm Hg in 1 of 101 values (1%). Linear regression analysis revealed a slope of 0.90, an r value of 0.9410, and an r² value of 0.8854 (P < 0.0001). Bland-Altman analysis revealed a bias of 0.58 mm Hg with a precision of ±2.1 mm Hg when comparing the TC-CO₂ with the arterial CO₂.

Implications: We conclude that, with certain caveats in mind, including the need to correlate the transcutaneous CO₂ with an initial arterial CO₂ value, transcutaneous CO₂ monitoring can be used to estimate arterial CO₂ in most neonates and children after cardiothoracic surgery.

	Introduction

Top Abstract Introduction Methods Results Discussion References

 
The noninvasive estimation of carbon dioxide (CO₂) provides a continuous measure for documenting the adequacy of ventilation. An accurate means of monitoring arterial CO₂ tension may limit the need for repeated arterial blood gas analysis and thereby reduce costs. In patients with normal pulmonary function and matching of ventilation-perfusion, ETCO₂ provides an accurate estimation of arterial CO₂ (1–3). However, sampling errors, as well as inequalities in ventilation-perfusion matching, such as increases in dead space or shunt fraction, adversely influence this correlation (4–6). The latter are often found in children with cyanotic heart disease, as well as those undergoing cardiovascular surgery (7).

Transcutaneous CO₂ (TC-CO₂) monitoring is accurate in toddlers and infants with respiratory failure (8), but it has not been studied in patients with alterations in cardiovascular function or those requiring pharmacologic support of cardiovascular function. We therefore prospectively investigated the efficacy and accuracy of TC-CO₂ monitoring in infants and children after cardiothoracic surgery.

	Methods

Top Abstract Introduction Methods Results Discussion References

 
The study was approved by our institutional review board. The study population included patients <8 yr of age who had undergone cardiothoracic surgery. Patient demographics included age, weight, gender, reason for cardiothoracic surgery, and type and dose of vasoactive medications.

The study population included consecutive patients who had undergone cardiothoracic surgical procedures. ETCO₂ was determined by infrared spectroscopy with a sidestream aspirator using a flow rate of 150 mL/min. If the ETCO₂ and arterial CO₂ values did not correlate (gradient 5 mm Hg) during the first postoperative hour, a TC-CO₂ electrode was placed. If the TC-CO₂ to arterial difference was 5 mm Hg, the TC-CO₂ electrode was recalibrated and reapplied on another site. If the discrepancy was still 5 mm Hg, the case was considered a clinical failure and no further data were collected. If the arterial to TC-CO₂ gradient was <5 mm Hg, the patient’s data were included.

TC-CO₂ was measured using a standard TC-CO₂/O₂ device (Radiometer, Copenhagen, Denmark). The protocol for TC-CO₂ monitoring included placement and maintenance of the monitor by the respiratory therapy staff. Before placement, the electrode membrane was cleaned and calibrated. The monitoring site was changed every 3–4 h to avoid thermal injury, and the electrode was recalibrated before placement at a new site. The working temperature of the electrode was kept at 43–43.5°C.

When clinically indicated, arterial blood gas tensions were measured at 37°C and corrected to the patient’s temperature. TC and arterial CO₂ values were recorded, with one to five sample sets (arterial CO₂ and TC-CO₂) collected from each patient. To avoid biasing the data, no more than five sample sets were collected from any single patient. The absolute difference between the TC and arterial CO₂ was calculated. No negative numbers were used, because this could have artificially lowered the mathematical mean of the differences between the TC-CO₂ and the arterial CO₂. For example, if the TC reading was 7 mm Hg higher or 7 mm Hg lower than arterial CO₂, 7 was used, not -7 or +7. Using the raw data (positive and negative numbers), a linear regression analysis and a Bland-Altman (10) analysis were performed to compare the TC and arterial CO₂ values. The Bland-Altman analysis included a bias, defined as the mean difference between values and precision, defined as the standard deviation of the bias. All data are expressed as the mean ± SD; P < 0.05 was considered significant.

	Results

Top Abstract Introduction Methods Results Discussion References

 
Thirty-three consecutive patients were enrolled in the study. In three patients, the ETCO₂ to arterial CO₂ gradient was <5 mm Hg. In the remaining 30 patients, the gradient was 5 mm Hg, and the TC-CO₂ electrode was placed. In 3 of these 30 patients, the TC-CO₂ to arterial CO₂ gradient was 5 mm Hg, despite recalibrating the TC-CO₂ monitoring and placing the probe on another body location. These three patients all exhibited cardiovascular instability and received dopamine 20 µg · kg^-1 · min^-1 and epinephrine 0.3–0.5 µg · kg^-1 · min^-1. The remaining 27 patients formed the cohort for the study. The data presented were collected from these 27 patients.

The 27 patients included 16 boys and 11 girls ranging in age from 2 days to 9 yr (1.7 ± 2.4 yr) and in weight from 3.2 to 25 kg (8.9 ± 4.4 kg). There were 6 neonates (1 mo of age), 15 infants (1 mo to 2 yr of age), and 6 children (2 yr of age). The cardiothoracic procedures included 3 thoracotomies for mediastinal masses, 4 closed heart procedures (Blalock-Taussig shunt, repair of coarctation of the aorta, and coarctation repair with pulmonary artery banding), and 20 open heart procedures with cardiopulmonary bypass. The open heart procedures included repair of arterial septal defect (n = 6), repair of ventricular septal defect (n = 3), repair of tetralogy of Fallot (n = 3), and repair of atrial septal defect with patch angioplasty of the pulmonary artery (n = 1). They also included repair of atrial septal defect and partial anomalous pulmonary venous return (n = 1), aortic valvotomy (n = 1), pulmonary valvotomy (n = 1), mitral valve annuloplasty and repair of left ventricular outflow tract obstruction (n = 1), repair of anomalous left coronary artery (n = 1), Norwood procedure (n = 1), and arterial switch (n = 1). Sixteen patients received infusions of vasoactive medications, including dopamine (0–5 µg · kg^-1 · min^-1 [n = 7], 6–10 µg · kg^-1 · min^-1 [n = 3], 11 µg · kg^-1 · min^-1 [n = 1]); dobutamine (0–5 µg · kg^-1 · min^-1 [n = 8], 6–10 µg · kg^-1 · min^-1 [n = 5], 11 µg · kg^-1 · min^-1 [n = 1]); milrinone 0.3–1 µg · kg^-1 · min^-1 (n = 6); nicardipine (n = 1); and nitroglycerin (n = 1).

A total of 101 sample pairs of TC and arterial CO₂ data were analyzed. The mean absolute difference (see Methods) between the TC-CO₂ and arterial CO₂ was 1.7 ± 1.4 mm Hg (range 0–9 mm Hg). The TC to arterial CO₂ difference was 0–2 mm Hg in 82 of 101 values (81%), 3–5 mm Hg in 18 of 101 values (18%), and >6 mm Hg in 1 of 101 values (1%). Linear regression analysis revealed a slope of 0.90, an r value of 0.9410, and an r² value of 0.8854 (P < 0.0001) (Figure 1). Bland-Altman analysis revealed a bias of 0.58 mm Hg with a precision of ±2.1 mm Hg when comparing the TC-CO₂ and the arterial CO₂ (Figure 2).

View larger version (16K): [in this window] [in a new window]   Figure 1. Linear regression analysis of transcutaneous CO2 versus arterial CO2. The y intercept = 3.4 ± 1.4 when x = 0.

View larger version (16K): [in this window] [in a new window]   Figure 2. Differences of transcutaneous CO2 minus arterial CO2 plotted against the arterial CO2. There were 101 sample sets included in the analysis; however, some of the points may overlap on the diagram.

	Discussion

Top Abstract Introduction Methods Results Discussion References

Although previous studies have validated the accuracy of TC-CO₂ monitoring in the neonatal population (9), its application outside of the neonatal population has been limited (13). The current study included only 16 of 27 patients older than 6 mo of age, and additional studies are needed to determine its accuracy in older children and adolescents. Because previous studies have demonstrated that TC-O₂ monitoring in this age group is not reliable (14), we did not attempt to use it in the current study.

The technique for TC-CO₂ monitoring has been described previously (9). Proper use and accuracy of this technique is dependent on appropriate training and expertise of personnel. Several factors may affect the accuracy of TC-CO₂ monitoring. Technical variables, such as trapped air bubbles, improper placement technique, damaged membranes, and inappropriate calibration, can be avoided by careful use of the equipment and training of the respiratory therapy staff. Aside from technical problems, patient problems may affect the accuracy of TC-CO₂ monitoring. The latter may include variations in skin thickness, the presence of edema, tissue hypoperfusion, or the administration of vasoconstricting drugs. All of these factors affect the ability of CO₂ to diffuse from the capillary bed to the membrane of the monitor. As no technique can be expected to be 100% reliable, periodic calibration with arterial values is recommended. Although the current study included only tracheally intubated and mechanically ventilated patients, TC-CO₂ monitoring can be used in spontaneously breathing patients without an artificial airway.

Inadequate tissue perfusion from cardiovascular depression or vasoconstriction from vasoactive drugs can be a problem immediately after cardiovascular surgery. Although some investigators have suggested that shock does not affect TC-O₂ and -CO₂ values (15), others have confirmed the clinical impression that the gradient between TC-CO₂ and arterial CO₂ increases as tissue perfusion decreases.¹² We found a poor correlation of the TC and arterial CO₂ in the three patients with cardiovascular instability requiring dopamine 20 µg · kg^-1 · min^-1 and epinephrine 0.3–0.5 µg · kg^-1 · min^-1. Using these vasoactive drugs at these doses can lead to peripheral and subcutaneous vasoconstriction. These factors, combined with diminished cardiac output, lead to decreased tissue perfusion and most likely explain the inaccuracies of TC-CO₂ monitoring in these patients. However, in other patients, even with the use of moderate doses of vasoactive drugs including dobutamine (20 µg · kg^-1 · min^-1), dopamine (10 µg · kg^-1 · min^-1), and milrinone (0.3–1 µg · kg^-1 · min^-1), there was a clinically acceptable correlation of TC and arterial CO₂.

Martin et al. (16) also demonstrated that the TC to arterial CO₂ gradient may widen as the arterial CO₂ increases because of an imbalance between local tissue CO₂ production and removal. We noted no such inaccuracy in our patients with hypercarbia; however, only 7 of the 101 arterial CO₂ values were 50 mm Hg.

A consideration of the limitations of the current study must include the number of sample sets (n = 101) and the limited patient population (n = 27 patients). Based on the number of patients, no statistical analysis to evaluate the effect of patient age, weight, type of surgical procedure, or type of cardiac lesion (left to right versus right to left) on the accuracy of ETCO₂ monitoring was possible. Additionally, because the arterial blood gas analyses were performed at various times after application of the TC electrode, we cannot comment on variations of its accuracy over time. However, the overall accuracy of the TC-CO₂ monitor in the current and previous studies suggests that there is not a significant decrease in accuracy over time provided the monitoring site is changed every 4 h. We did note a poor correlation in patients with cardiovascular instability who required pharmacologic support with epinephrine or dopamine in doses >10 µg · kg^-1 · min^-1.

In summary, in the current study, we demonstrated that TC-CO₂ measurement provides a clinically acceptable estimate of arterial CO₂ in infants and children after cardiothoracic surgery. The technique may be inaccurate in patients with cardiovascular dysfunction requiring vasoactive medications, especially at large doses.

	Footnotes

 
¹ Peabody JL, Emery JR. Does shock affect the transcutaneous-arterial P_CO2 gradient? [abstract]. Pediatr Res 1982;16:302A.

² Kashyap S, Stefanski M, Schulze K, James LS. Effects of changes in circulation on transcutaneous P_CO2 [abstract]. Pediatr Res 1981;15:666.

	References

Top Abstract Introduction Methods Results Discussion References

 

1. Hillier SC, Badgwell JM, McLeod ME, et al. Accuracy of end-tidal PCO2 measurements using a sidestream capnometer in infants and children ventilated with the Sechrist infant ventilator. Can J Anaesth 1990;37:318–21.[Web of Science][Medline]
2. Tobias JD, Flanagan JFK, Wheeler TJ, et al. Non-invasive monitoring of end-tidal CO₂ via nasal cannulas in spontaneously breathing children during the perioperative period. Care Med 1994;22:1805–8.
3. Campbell FA, McCleod ME, Bissonnette B, Swartz JS. End-tidal carbon dioxide measurement in infants and children during and after general anesthesia. Can J Anaesth 1994;41:107–10.[Web of Science][Medline]
4. Whitesell R, Asiddao C, Gollman D, et al. Relationship between arterial and peak expired carbon dioxide pressure during anesthesia and factors influencing the difference. Anesth Analg 1981;60:508–12.[Abstract/Free Full Text]
5. Liu SY, Lee TS, Bongard F. Accuracy of capnography in nonintubated patients. Chest 1992;102:1512–5.[Abstract/Free Full Text]
6. Yamanaka MK, Sue DY. Comparison of arterial-end tidal PCO2 difference and dead space/tidal volume ratio in respiratory failure. Chest 1987;92:832–5.[Abstract/Free Full Text]
7. Burrows FA. Physiologic dead space, venous admixture, and the arterial to end-tidal carbon dioxide difference in infants and children undergoing cardiac surgery. Anesthesiology 1989;70:219–25.[Web of Science][Medline]
8. Tobias JD, Meyer DJ. Nonivasive monitoring of carbon dioxide during respiratory failure in toddlers and infants: end-tidal versus transcutaneous carbon dioxide. Anesth Analg 1997;85:55–8.[Abstract]
9. Cassady G. Transcutaneous monitoring in the newborn infant. J Pediatr 1983;103:837–48.[Web of Science][Medline]
10. Bland JM, Altman DG. Statistical methods for assessing agreement between two methods of clinical measurement. Lancet 1986;12:307–10.
11. Rozycki H, Sysyn GD, Marshall MK, et al. Mainstream end-tidal carbon dioxide monitoring in the neonatal intensive care unit. Pediatrics 1998;101:648–53.[Abstract/Free Full Text]
12. Wahba RWM, Tessler MJ. Misleading end-tidal CO₂ tensions. Pediatrics 1996;43:862–88.
13. Bhavani-Shankar K, Steinbrook RA, Mushlin PS, Freiberger D. Transcutaneous PCO₂ monitoring during laparoscopic cholecystectomy in pregnancy. Can J Anaesth 1998;45:164–9.[Web of Science][Medline]
14. Knill RL, Clement JL, Kieraszewicz LK, Dodgson BG. Assessment of two non-invasive monitors of arterial oxygenation in anesthetized man. Anesth Analg 1982;61:582–6.[Abstract/Free Full Text]
15. Hansen TN, Tooley WH. Skin surface carbon dioxide tension in sick infants. Pediatrics 1979;64:942–6.[Abstract/Free Full Text]
16. Martin RJ, Beoglos A, Miller MJ, et al. Increasing arterial carbon dioxide tension: influence on transcutaneous carbon dioxide tension measurements. Pediatrics 1988;81:684–7.[Abstract/Free Full Text]

This article has been cited by other articles:

				E J Molloy and K Deakins Are carbon dioxide detectors useful in neonates? Arch. Dis. Child. Fetal Neonatal Ed., July 1, 2006; 91(4): F295 - F298. [Abstract] [Full Text] [PDF]

				J. Wilson, P. Russo, J. Russo, and J. D. Tobias Noninvasive Monitoring of Carbon Dioxide in Infants and Children With Congenital Heart Disease: End-Tidal Versus Transcutaneous Techniques J Intensive Care Med, September 1, 2005; 20(5): 291 - 295. [Abstract] [PDF]

				J. W. Berkenbosch, J. Lam, R. S. Burd, and J. D. Tobias Noninvasive Monitoring of Carbon Dioxide During Mechanical Ventilation in Older Children: End-Tidal Versus Transcutaneous Techniques Anesth. Analg., June 1, 2001; 92(6): 1427 - 1431. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Email this article to a colleague Similar articles in this journal Similar articles in Web of Science Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Permissions Citing Articles Citing Articles via HighWire Citing Articles via Web of Science (19) Citing Articles via Google Scholar Google Scholar Articles by Tobias, J. D. Articles by Meyer, D. J. Search for Related Content PubMed PubMed Citation Articles by Tobias, J. D. Articles by Meyer, D. J.