Background: Capnometry can reduce arterial blood sampling and allow fast and noninvasive assessments using end-tidal of carbon dioxide (EtCO2).

Objectives: The aims of this study were to evaluate the correlation of partial pressure of arterial CO2 (PaCO2)levels and EtCO2, and to verify whether capnometry would be useful for noninvasive monitoring of CO2 levels in intubated premature infants with and without diffuse parenchymal lung diseases (DPLD).

Methods: This study was conducted in premature infants admitted to the Neonatal Intensive Care Unit (NICU) from August 2014 to November 2016. EtCO2 levels were compared with PaCO2 levels, in intubated premature infants with and without DPLD. Both parameters were obtained daily until tracheal extubation. The correlation coefficient and degree of bias between them was determined.

Results: Overall, 221 measurements of EtCO2 and PaCO2 levels were obtained from 51 neonates. Twenty-eight were obtained from neonates without DPLD (12.7%) and 193 from those with DPLD (87.3%). The most frequent cause of DPLD was respiratory distress syndrome (RDS) in 86.5%. There was a positive correlation between PaCO2 and EtCO2 levels (n = 221; r = 0.853; p < 0.0001) in the overall cohort. Both groups showed a good correlation between both parameters, without DPLD (mean bias = 0.21, SD, 7.05; 95% CI -13.61 – 14.05), and with DPLD (mean bias = 0.37, SD, 7.66; 95% CI -14.65 – 15.39). Conclusions: Capnometry is a useful noninvasive technique to monitor intubated premature infants. EtCO2 measurement may be a valid adjunctive parameter when titrating ventilator support.

Nakato AM, da Silva RPGVC, Rosario Filho NA (2018) Correlation of Partial Pressure of Arterial Carbon Dioxide and End-Tidal Carbon Dioxide in Intubated Pre-mature Infants. Ann Pediatr Child Health 6(1): 1140.

• Transcutaneous apnometry
• Mechanical ventilation
• Blood gas analyses
• Hypercapnia
• Premature infants
• Respiratory insufficiency

In the United States, approximately one out of nine live births occurs prematurely [1]. Preterm birth is associated with serious respiratory illnesses and often requires ventilatory support [1- 3]. Prematurity is the most important cause of death in the first month of life and contributes to increased mortality in the first year [3]. In Brazil, premature births occurred in 9.8% of the 2.913.160 births registered in 2011 [4].

Premature infants on mechanical ventilation require special attention to ventilation parameters and oxygen levels. Arterial blood gas analysis is the gold-standard test to verify both parameters. However, it is expensive, requires multiple blood sampling, and may cause distress, pain, infection, and arterial vasospasm [5,6]. Continuous noninvasive monitoring of carbon dioxide (CO2 ) levels in the neonatal intensive care unit (NICU) may protect infants from the consequences of hypocapnia and hypercapnia [7]. Capnometry may bean alternate procedure that can prevent repeated arterial blood sampling and allow a fast and noninvasive estimate of PaCO2 [5,8].

Premature infants with impending respiratory failure due to lung immaturity or poor respiratory efforts generally present with a high respiratory rate and low tidal volume in mechanical ventilation (MV), leading to considerable ventilation-perfusion imbalance, which can potentially influence the correlation of the PaCO2 and the EtCO2 [9]. Accurate, noninvasive measurements of CO2 production areparticularly troublesome in ventilated premature infants due to technical limitations [10].

Capnometry has the advantage of responding fast to changes in blood CO2 levels, and provides useful information about the efficacy of alveolar ventilation [11]. A large sample cuvette lumen is used in capnometry in order to minimize the work of breathing. Additionally, pulmonary secretions generally do not interfere with the CO2 analysis, although the airway cuvette is relatively bulky and can add to dead space [12]. In addition, accurate and noninvasive assessment of EtCO2 in premature infants has some disadvantages and remains controversial because of technical challenges. Disadvantages of capnometry include a delay between sampling and measurement of EtCO2 , falsely low EtCO2 with high aspirating flow rates, and possible contamination of the analyzer with mucus or water [6].

Another well-documented technique used in newborn infants for evaluated the CO2 levels is non-invasive transcutaneous (Tc) blood gas monitoring. The non-invasive Tc have shown agreement with the arterial measurements for PaCO2 in others studies[13,14]. This technique uses electrodes in contact with the skin surface and it is based on obtaining the balance of carbonic acid combined with the temperature that facilitates blood flow and the stabilization of the partial pressure of carbon dioxide transcutaneously (PtcCO2 ) [15].

This study aimed to correlate PaCO2 levels measured in arterial blood and EtCO2 by means of a capnometer in intubated premature infants presenting with and without DPLD, and to verify whether capnometry would be a useful method for noninvasive monitoring of effective ventilation in these infants.

This experimental study was conducted in premature infants admitted to the NICU at the University Hospital from August 2014 to November 2016. The study was approved by the Ethics Committee of Federal University of Parana, Curitiba, Brazil. All parents signed an informed consent form and authorization to include the data of their infants in the study. This is prospective study and the design is of the type diagnosis test study.

All infants were intubated and received ventilation therapy with pressure limited, time-cycled ventilators in either an assist-control mode or a synchronous, intermittent mandatory ventilation mode, Puritan Bennett™ 840®, Carlsbad, California. Premature infants with substantial hemodynamic instability, airway congenital malformation, or irregular breathing frequency were excluded.

EtCO2 / PaCO2 comparisons were obtained from all premature infants, who were separated into groups with and without DPLD. Table (1) presents the characteristics of the infants and Table (2), the clinical conditions of the infants on mechanical ventilation included in the study.

Heparinized arterial blood samples from radial artery were analyzed with the equipment GEM Premier 3000® (Instrumentation Laboratory, Lexington, MA), which was calibrated daily. Immediately before blood sampling, the EtCO2 value shown on the display of the portable capnometer was recorded simultaneously to the PaCO2 . The interval between the measurements by capnometry and arterial blood gas did not exceed 15 minutes.

The EtCO2 in exhaled air was measured by a mainstream portable capnometer (EMMA Emergency Capnometer, Phasein AB, Danderyd, Sweden) placed between the endotracheal tube and the circuit of mechanical ventilator. The capnometer has an internal calibration capability and this calibration was performed according to the manufacturer’s recommendations. Values of peak CO2 concentration were displayed breath-to-breath. Time for the signal to change from a specified low value to a specified high value was ≤ 60 ms. The mean exhaled CO2 of three consecutive measurements was used in the analysis.

Premature infants were monitored daily, starting on the first day of mechanical ventilation until the removal of the endotracheal tube. The measurements were recorded along with information on infants’ clinical diagnoses, vital parameters, and mechanical ventilation parameters. The data collection did not interfere with the routine of the ICU. Arterial blood gas analysis was performed daily per unit standard practice, but capnometry was performed only for the study purpose.

Statistical analysis

Descriptive data were analyzed with Statistica®, version 10, Pearson’s correlation and Bland-Altman technique was performed with MedCalc®, version 7.4.4.1. Student’s paired t test was used to analyze differences between EtCO2 and PaCO2 , and the Bland-Altman technique was applied to assess agreement and bias between both measurements. The Mann-Whitney test compared the independent samples of the groups. The Shapiro-Wilk test was used to analyze the normality in the groups. Data are presented as mean (± standard deviation) or median (inter quartile range). A p value< 0.05 was considered statistically significant.

The study population encompassed 51 premature infants, 26 (51%) boys. The mean gestational age of the entire cohort was 28.08 + 3.19 weeks, and the median birth weight was 880 g (700 – 1125 g). The Apgar scores were stratified, separated in groups and the score obtained at 5 minutes was used in the analysis. The premature infants with Apgar score less than three were classified at zero, between three to six at one and greater than six at two. For the temperature was used the body temperature and the incubator temperature data. The incubator was used for maintain the baby heated (Table 1).

All patients were received mechanical ventilation, but not all of them had pulmonary disease. Some premature infants were on mechanical ventilation due to hemodynamic instability (arterial hypotension / shock), perinatal asphyxia, fatigue, apnea (secondary to immaturity of the respiratory center), respiratory muscle failure or heart failure, but they did not present respiratory failure due to DPLD. Infants were grouped according to the presence or absence of DPLD, yielding 28 analyses in infants without DPLD (12.7%) and 193 analyses in infants with DPLD (87.3%). The most prevalent DPLD was RDS in 44 (86.5%) premature infants. The clinical conditions of the 7 (13.5%) infants without DPLD were perinatal asphyxia 2 (4%), heart failure 1 (1.9%), apnea 1 (1.9%), sepsis 1 (1.9%), hemodynamic instability 1 (1.9%), and fatigue 1 (1.9%), (Table 2).

Values of PaCO2 and EtCO2 from all 51 patients were similar (p=0.82) and there was no significant differences in PaCO2 and in EtCO2 between groups with (p=0.81) and without DPLD (p=0.98).

Daily arterial blood gas results yielded 221 PaCO2 / EtCO2 from 51 patients. Overall, mean PaCO2 was 44 + 14.4 mmHg and mean EtCO2 was 43.7 + 11.8 mmHg. For the group without DPLD, the mean PaCO2 was 43 + 16.06 mmHg, median 48.5 (26 – 53.5) and mean EtCO2 was 42.9 + 12.3 mmHg, median 43 (33.5 – 51.5), with 28 samples. In group with DPLD, the mean PaCO2 was 44.1 + 14.25 mmHg, median 43 (35 – 52) and mean EtCO2 was 43.8 + 11.8 mmHg, median 43 (37 – 50) with 193 samples (Figure 1).

Significant positive correlation was noted between PaCO2 and EtCO2 (n = 221; r = 0.853; p < 0.0001) (Figure 1), and there was a longer strong concordance [16] with mean difference of 0.352 ± 7.57 with 95% confidence interval of 14.5 – 15.2 (Figure 2).

We also observed a significant correlation between values of PaCO2 and EtCO2 in patients without DPLD, (mean bias = 0.21, SD, 7.05; 95% CI -13.61 – 14.05), and with DPLD (mean bias = 0.37, SD, 7.66; 95% CI -14.65 – 15.39), variations within acceptance limits of approximately 2 mmHg (Figure 2).

Monitoring of both oxygenation and CO2 is essential in neonates submitted to mechanical ventilation. EtCO2 is an alternate noninvasive method to monitor CO2 values, although this measurement in infants is not well accepted, particularly in preterm infants [5,6,17]. EtCO2 values can be influenced by factors such as arterial hypotension or hypertension, temperature variations, changes in respiratory rate or ventilation-perfusion ratio [18]. In our study, DPLD did not influence the results of EtCO2 as an estimate of PaCO2 . The correlations between EtCO2 and PaCO2 were similar in infants with and without DPLD.

Blood gas monitoring is critical in premature infants with respiratory insufficiency and is a priority in NICU [15]. Arterial blood gas analysis is the gold-standard method to assess the adequacy of ventilation [19]. In our study, concordant values were observed in both groups, the mean difference between EtCO2 and PaCO2 in infants with DPLD was small, which can be interpreted as a successful result. Thus, EtCO2 monitoring can be recommended as an alternative and may be helpful in adjusting mechanical ventilation settings [20].

PaCO2 and EtCO2 were positively correlated in all premature infants and corroborating to different authors [5]. Wu et al. (2003), [6], showed a good correlation in premature infants with RDS. Contrariwise Watkins and Weindling in 1987 [9], found a poor correlation (r = 0.387, p<0.01, n = 62), between both parameters in sick preterm neonates attributed to increased physiologic dead space, different ventilation-perfusion and the severity of pulmonary disease. EtCO2 monitoring has not had much support in most premature infants with DPLD, considering that these infants have higher respiratory rate, lower tidal volume, and mismatched ventilation-perfusion [6].

The most prevalent DPLD in our cohort was RDS, which can result primarily from a pulmonary surfactant deficiency [21]. This complication is common in premature infants and is associated with pulmonary hypertension, as well as abnormalities of postnatal alveolarization and neovascularization [22].

Our study attempted to show the possibility of a less invasive technique to monitor CO2 levels, considering that premature infants require careful monitoring through frequent blood sampling, which sometimes conflicts with the nonaggressive, gentle approach of infant care in the NICU [15]. Thus, there is a need for monitoring through noninvasive techniques for the occurrence of hypercapnia in infants, minimizing blood sampling to reduce the risk of anemia, and protecting these children from discomfort and pain [23].

Our study had some limitations. One of them was the size of the sample that was small with only 51 premature infants enrolled and analyzed in a period of 2 years. The study included only intubated premature infants without neurological impairment and without hemodynamic instabilities, which also contributed to the difficulty in enrolling more infants. Further, with increasing use of non-invasive ventilation there are less infants remain intubated. Additionally hospital employees strike and ICU reform may have affected the data collection for this study.

Table 1: Characteristics of the infants included in the study (n = 51).

Table 2: Clinical conditions of patient on mechanical ventilation (n = 51).

The results of our study suggest that EtCO2 is a valid adjunctive parameter when titrating ventilator support and may be useful for real-time analyses of abnormal PaCO2 levels both premature infants with and without pulmonary disease. Measuring both EtCO2 in combination with other clinical parameters allows detection of lung disease progression, appropriateness strategies of treatment and check the ventilation-perfusion inefficiency.

We thank the team of the Neonatal Intensive Care Unit, Hospital de Clínicas and Ms. Lilian Messias for statistical support and assistance.

1. Pryhuber GS, Maitre NL, Ballard R A, Cifelli D, Davis SD, Ellenberg JH, et al. Prematurity and respiratory outcome program (PROP): study protocol of a prospective multicenter study of respiratory outcomes of preterm infants in the United States. BMC Pediatr. 2015; 15: 1-14.

2. Katz PJ, Anne CC Lee, Kozuki N, Lawn JE, Cousens S, Blencowe H, et al. Europe PMC Funders Group Mortality risk in preterm and small-for-gestational-age infants in low-income and middle-income countries?: a pooled country analysis. Lancet. 2013; 382: 417-425.

3. Glass HC, Costarino AT, Stayer SA, Brett CM, Cladis F, Davis PJ. Outcomes for extremely premature infants. Anesth Analg. 2015; 120: 1337-1351.

4. Orsi KC, Llaguno NS, Avelar AF, Tsunemi MH, Pedreira Mda L, Sato MH, et al. Effect of reducing sensory and environmental stimuli during hospitalized premature infant sleep. Rev Esc Enferm USP. 2015; 49: 550-555.

5. Bhat R, Abhishek N. Mainstream end-tidal carbon dioxide monitoring in ventilated neonates Mainstream end-tidal carbon dioxide monitoring in ventilated neonates. Singapore Med J. 2008; 49: 199- 203.

6. Wu C, Chou H, Hsieh W, Chen W, Huang PY, Tsao PN. Good Estimation of Arterial Carbon Dioxide by End-Tidal Carbon Dioxide Monitoring in the Neonatal Intensive Care Unit. Pediatr Pulmonol. 2003; 30: 292- 295.

7. Kugelman A, Zeiger-Aginsky D, Bader D, Shoris I, Riskin A. A novel method of distal end-tidal CO2 capnography in intubated infants: comparison with arterial CO2 and with proximal mainstream end-tidal CO2 . Pediatrics. 2008; 122: 1219-1224.

8. Belpomme V, Ricard-Hibon A, Devoir C, Dileseigres S, Devaud ML, Chollet C , et al. Correlation of arterial PCO2 and PETCO2 in prehospital controlled ventilation. Am J Emerg Med. 2005; 23: 852-859.

9. Watkins AM, Weindling AM. Monitoring of end tidal CO2 in neonatal intensive care. Arch Dis Child. 1987; 62: 837-9.

10. Kingdon CC, Mitchell F, Bodamer OA, Williams AF. Measurement of carbon dioxide production in very low birth weight babies. Arch Dis Child Fetal Neonatal Ed. 2000; 83: 50-55.

11. Caroline C, Miguel P, Maria A, Domingues T, Faria F. The effect of body temperature on the accuracy of arterial and end-tidal carbon dioxide measurement q. Measurement. 2011; 44: 60-64.

12. Anderson CT, Breen PH. Carbon dioxide kinetics and capnography during critical care. Crit Care. 2000; 4: 207-215.

13. Marsden D, Chiu MC, Paky F, Helms P. Transcutaneous oxygen and carbon dioxide monitoring in intensive care. Arch Dis Child. 1985; 60: 1158-1161.

14. Berkenbosch JW, Tobias JD. Transcutaneous carbon dioxide monitoring during high-frequency oscillatory ventilation in infants and children. Crit Care Med. 2002; 30: 1024-1027.

15. Sandberg KL, Brynjarsson H, Hjalmarson O. Transcutaneous blood gas monitoring during neonatal intensive care. Acta Paediatr. 2011; 100: 676-679.

16. Bland JM, Altman DG. Statistical methods for assessing agreement between two methods of clinical measurement. Lancet. 1986; 1: 307- 310.

17. Rüdiger M, Töpfer K, Hammer H, Schmalisch G, Wauer RR. A survey of transcutaneous blood gas monitoring among European neonatal intensive care units. BMC Pediatr. 2005; 6: 3-8.

18. Rasera CC, Gewehr PM, Domingues AM, Junior FF. Measurement of End-Tidal Carbon Dioxide in Spontaneously Breathing Children After Cardiac Surgery. Am J Crit Care. 2011; 20: 388-394.

19. Tai C, Lu FL, Chen P, Jeng S, Chou H, Chen C, et al. Noninvasive Capnometry for End-tidal Carbon Dioxide Monitoring via Nasal Cannula in Nonintubated Neonates. Pediatr Neonatol. Taiwan Pediatric Association; 2010; 51: 330-335.

20. Ashwood ER, Kost G, Kenny M. Temperature Correctionof BloodGasand pH Measurements. Clin Chem. 1983; 29: 1877-1885.

21. Chung EH, Ko SY, Kim IY, Chang YS, Park WS. Changes in dead space/tidal volume ratio and pulmonary mechanics after surfactant replacement therapy in respiratory distress syndrome of the newborn infants. J Korean Med Sci. 2001; 16: 51-56.

22. Schreiber MD, Gin-Mestan K, Marks JD, Huo D, Lee G, Srisuparp P. Inhaled nitric oxide in premature infants with the respiratory distress syndrome. N Engl J Med. 2003; 349: 2099-2107.

23. Kugelman A, Riskin A, Shoris I, Ronen M, Stein IS, Bader D. Continuous Integrated Distal Capnography in Infants Ventilated With High Frequency Ventilation. Pediatr Pulmonol. 2012; 47: 876-883.