Arterial versus end-tidal carbon dioxide levels in children with congenital heart disease: a prospective cohort study in patients undergoing pulmonary catheterization

Background:

Capnography has been the standard in the operating room for a long time now. When variable amounts of intrapulmonary shunt and intracardiac shunt are taken into account, arterial carbon dioxide (CO₂) and end-tidal CO₂ typically match rather well. The gap between arterial and end-tidal CO₂ widens in patients with cardiopulmonary disorders. The current study sought to determine how arterial and end-tidal CO₂ correlated with each other and with hemoglobin saturation both before and after pulmonary catheterization in a pediatric population with congenital heart disease.

Methods:

Fifty-seven children with congenital heart disease who underwent cardiopulmonary catheterization between March 2018 and April 2019 were included in a prospective cohort study at Children’s Medical Center. Arterial and end-tidal CO₂, and hemodynamic variables were assessed prior to the catheterization procedure. Then the patients underwent catheterization, and before being extubated, these variables were again assessed and compared to the baseline levels.

Results:

End-tidal CO₂ increased significantly in cyanotic patients following the catheterization procedure, and the difference between arterial and end-tidal CO₂ decreased significantly. End-tidal CO₂, arterial CO₂, and their difference did not significantly change in non-cyanotic patients following the catheterization procedure. End-tidal and arterial CO₂ were not significantly correlated in cyanotic patients (r=0.411, P=0.128), but they were correlated after the catheterization procedure (r=0.617, P=0.014).

Conclusions:

End-tidal CO₂ can estimate arterial CO₂ in non-cyanotic patients reasonably. End-tidal CO₂ cannot be used to estimate arterial CO₂ in cyanotic patients since there is no association. After cardiac defect correction, end-tidal CO₂ can be a reliable predictor of arterial CO₂.

Introduction

Highlights

• End-tidal carbon dioxide (CO₂) can estimate arterial CO₂ in acyanotic patients reasonably.

• End-tidal CO₂ cannot be used to estimate arterial CO₂ in cyanotic patients since there is no association.

• After cardiac defect correction, end-tidal CO₂ can be a reliable predictor of arterial CO₂.

Evaluation of gas exchange is a crucial instrument for monitoring and diagnosing patients suffering from cardiovascular disorders and patients undergoing anesthesia¹. The arterial blood gas analysis (ABG) is the procedure that is considered to be the gold standard for determining arterial carbon dioxide (CO₂)². This method, on the other hand, has a number of significant drawbacks: it is only capable of displaying a sectional measurement of the CO₂ level, despite the fact that it is a physiologically dynamic variable item; it has the potential to cause damage to soft tissue; it can lead to an increase in infection rates; and it requires the utilization of trained human resources in order to carry out the measurement³.

Regarding the drawbacks of the ABG procedure, noninvasive techniques have received greater attention lately. These noninvasive techniques include transcutaneous technique and capnography using end-tidal CO₂ measurement^3,4. End-tidal CO₂ measurements are tested for their effectiveness in a number of common medical procedures and conditions, such as cardiopulmonary resuscitation (CPR), airway evaluation, procedural sedation and analgesia, obstructive pulmonary disease, pulmonary embolism, heart failure, shock, metabolic disorders, diabetic ketoacidosis (DKA), gastroenteritis, and trauma^5–15.

When patients are in the pediatric age group and require safer anesthetic and surgical operations, the requirement for a noninvasive precise technology for CO₂ measurement becomes more prominent¹⁶. The effectiveness of employing the end-tidal approach for CO₂ level estimation, however, may be significantly impacted by congenital heart disease¹⁷. However, surgical correction of the defects may result in an improvement in the end-tidal method’s accuracy.

To the best of our knowledge, there are not many studies examining the relationship between arterial and end-tidal CO₂ levels in children undergoing catheterization procedure for congenital heart problems. The purpose of this study was to evaluate pulmonary catheterization’s contribution to the enhancement of this association as well as the accuracy of end-tidal CO₂ in such individuals.

Methods

This prospective cohort study was performed on all pediatric patients who underwent cardiopulmonary catheterization between March 2018 and April 2019 in Children’s Medical Center, the largest pediatric academic center in our country. The study protocol was approved by the ethics committee of our institution (Ethics Code: IR.TUMS.MEDICINE.REC.1397.470). The work has been reported in line with the STROCSS criteria¹⁸.

Inclusion criteria were: age less than 12 years, informed consent by the parents, and stable hemodynamic status. Patients who were not candidates of catheterization were excluded. Also, those with an American Society of Anesthesiologists (ASA) score of 4 or higher were excluded from the study¹⁹.

Before the catheterization procedure, a pulse oximeter was used to check the patients’ hemoglobin saturation for monitoring. The patient received oral premedication in the form of 0.5 mg/kg of midazolam before being preoxygenated with 100% oxygen; 0.2 mg/kg of the cisatracurium relaxant and 8% sevoflurane were used by inhalation to induce anesthesia; and 1.5% isoflurane was used to keep the anesthesia continued.

The Draeger model Fabius-plus volume control equipment was used to provide controlled ventilation for the patients while measuring and recording arterial variables, hemodynamic variables, and hemoglobin concentration. In parallel, the arterial sample of end-exhalation CO₂ was measured, recorded, and compared with the arterial sample collected in the operation room using the GemePremier 3000 model.

The patient underwent catheterization after arterial variables, expiratory CO₂, and hemodynamic factors were measured prior to the procedure. Before the extubation, variables were also measured again, and the results were compared to the data from before the procedure. Due to the study’s blind design, the data collector was unable to distinguish between patients who were and were not cyanotic when gathering the data.

Statistical analysis

Mean, standard deviation (SD), frequency, and percentage were used to describe the data. Kolmogorov–Smirnov (KS) test was used to test the normality of the data. Nonparametric Wilcoxon test or paired t-test was used to compare the results before and after the catheterization procedure. All analyses were performed by SPSS 25.0. A P value less than 0.05 was considered significant.

Results

Forty-nine participants in total – 24 girls (49%) and 25 boys (51%) – participated in this study. Table 1 displays the details regarding the patients’ weight and age. The gender distribution, age, and weight of those with and without cyanotic diseases did not differ significantly (P>0.05).

Table 1.

Age and weight distribution

	Mean	SD	Min	Max
Age (months)	24.34	23.29	0	96
Weight (kg)	11.05	6.77	2.3	38

Figure 1 illustrates the frequency of underlying diseases. The most prevalent disease among the participants was patent ductus arteriosus (PDA), which was present in 22 of the participants (44.9%). In the study, 15 patients (30.6%) had cyanosis and 34 people (69.4%) did not experience cyanosis after the catheterization procedure. Figure 2 shows the frequency of the patients’ surgical procedures. The most frequent procedure done on patients was PDA closure, which 23 patients (46.7%) underwent. In cyanotic patients, atrial septostomy (13.3%), PDA stenting (46.7%), pulmonary artery stenting (20%), pulmonary valve balloon dilation (13.3%), and right pulmonary artery (RPA) stenting (6.7%) were performed.

Figure 1.

Frequency of underlying diseases. AS, aortic stenosis; ASD, atrial septal defect; AVSD, atrioventricular septal defect; CoA, coarctation of the aorta; PA, pulmonary atresia; PDA, patent ductus arteriosus; PS, pulmonary stenosis; TGA, transposition of the great arteries; ToF, tetralogy of Fallot; VSD, ventricular septal defect.

Figure 2.

Frequency of each surgical treatment type. Device removal: stent removal. ASD, atrial septal defect; PDA, patent ductus arteriosus; RPA, right pulmonary artery.

With the exception of arterial blood pH (P=0.05) and oxygen pressure (P=0.001), the distribution of all data was normal, and parametric tests were employed to assess the differences between groups.

Table 2 provides an overview of the patients’ vital signs both before and after the catheterization procedure. Before and after the catheterization procedure, there was no appreciable difference in heart rate, systolic blood pressure, or diastolic blood pressure between patients both with and without cyanosis or between males and females (P>0.05).

Table 2.

Mean (SD) of vital signs and arterial blood gas findings before and after catheterization procedure

	All patients (N=49)	Non-cyanotic (N=34)	Cyanotic (N=15)	P
Pre-op heart rate	121.24 (16.52)	119.21 (17.7)	125.87 (12.83)	0.41
Post-op heart rate	119.51 (18.24)	115.5 (17.96)	128.6 (15.89)	0.689
Pre-op systolic blood pressure	81.61 (14.53)	85.41 (12.56)	73 (15.42)	0.095
Post-op systolic blood pressure	83.16 (15.26)	87.82 (13.44)	72.6 (14.15)	0.650
Pre-op diastolic blood pressure	43.67 (10.59)	46.68 (8.71)	36.87 (11.6)	0.332
Post-op diastolic blood pressure	45.49 (12.26)	49.06 (11.24)	37.4 (10.76)	0.765
Pre-op pH	7.35 (0.06)	7.36 (0.06)	7.33 (0.07)	0.244
Post-op pH	7.34 (0.08)	7.35 (0.06)	0.12	0.769
Pre-op HCO3	22.81 (2.64)	22.61 (2.4)	23.25 (3.16)	0.223
Post-op HCO3	21.48 (3.07)	21.63 (2.7)	21.14 (3.88)	0.622
Pre-op O2	151 (120.57)	193.35 (118.51)	55 (50.05)	<0.001
Post-op O2	149.14 (106.295)	183.44 (109.68)	71.4 (32.35)	<0.001

Table 2 lists the outcomes of blood gas analysis for patients both before and after the catheterization procedure. People with cyanotic heart disease had considerably lower arterial oxygen pressure before and after the operation than those without cyanotic heart disease (P=0.001). Following the catheterization procedure, both those with and without cyanosis experienced a substantial decrease in bicarbonate levels (P=0.05). In patients with cyanosis, the amount of arterial blood oxygen increased significantly after the catheterization procedure (P=0.02), while in patients without cyanosis, there was no significant difference between arterial blood oxygen levels before and after the catheterization procedure (P=0.158). Patients with and without cyanosis did not significantly differ in their pH readings before and after the catheterization procedure (P>0.05).

Table 3 displays the arterial and end-tidal CO₂ measurements in patients before and after the catheterization procedure. People with cyanotic heart disease had considerably lower end-tidal CO₂ levels than those without the condition (P=0.05). End-tidal CO₂ levels increased significantly, and the difference between arterial and end-tidal CO₂ decreased significantly in patients with cyanosis following the catheterization procedure (P=0.05). In patients with cyanosis following the catheterization procedure, there was no significant change in arterial CO₂ levels (P=0.134). The arterial and end-tidal CO₂ levels as well as their difference did not significantly alter after the catheterization procedure in those without cyanosis (P>0.05).

Table 3.

Mean (SD) of arterial and end-tidal CO₂ before and after catheterization procedure

	All patients (N=49)	Non-cyanotic (N=34)	Cyanotic (N=15)	P
Pre-op arterial CO2	41.24 (7.16)	40.06 (6.91)	43.93 (7.22)	0.878
Post-op arterial CO2	39.84 (6.72)	39.68 (6.05)	40.2 (8.27)	0.113
Pre-op end-tidal CO2	30.59 (5.8)	32.68 (4.81)	25.87 (5.13)	0.004
Post-op end-tidal CO2	32.35 (4.85)	32.94 (4.92)	31 (4.53)	0.002
Pre-op PaCO2- end-tidal CO2	10.65 (7.82)	7.38 (5.71)	18.07 (6.93)	0.118
Post-op PaCO2- end-tidal CO2	7.49 (6.69)	6.74 (6.72)	9.2 (6.53)	0.705

End-tidal CO₂ and arterial CO₂ were significantly correlated in patients without cyanosis prior to the catheterization procedure (r=0.576, P=0.001), but they were not significantly correlated following the catheterization procedure (r=0.264, P=0.131). Before the catheterization procedure, there was no statistically significant correlation between end-tidal CO₂ and arterial CO₂ in cyanotic patients (r=0.411, P=0.128), whereas there was a significant correlation (r=0.617, P=0.014) after the catheterization procedure.

Discussion

In this study, individuals with congenital heart disease who underwent cardiopulmonary catheterization had their arterial and end-tidal CO₂ levels evaluated before and after the operation. Despite the fact that similar trials have been carried out in a variety of heart diseases, little is known about this particular patient population and this particular kind of operation.

Before the catheterization procedure, in this study, individuals without cyanosis had considerably lower arterial CO₂ and end-tidal CO₂ differences than cyanotic patients (7.38 vs. 18.07). This difference is clinically important even if it was not statistically significant (P=0.118). The results of this investigation are in agreement with those of Burrows^20, which indicated that end-tidal CO₂ measurements provided a reduced estimate of arterial CO₂ levels in children with cyanotic heart disease. Also, children with congenital heart disease who had cyanosis or who were non-cyanotic and had increased pulmonary blood flow, compared to other congenital heart patients, showed a greater differential between arterial CO₂ and end-tidal CO₂ in Choudhury et al.’s study²¹. One of the causes of the rise in the differential between arterial and end-tidal CO₂ in cyanotic individuals is venous mixing. The amount of blood that enters the systemic blood stream without absorbing oxygen is calculated using venous admixture. The pulmonary blood flow that flows in the lung’s alveoli with little ventilation may be the cause of the discrepancy between the quantity of venous mixing expected and observed. The difference between arterial CO₂ and end-tidal CO₂ is widened by venous mixing^20,22.

End-tidal CO₂ is more closely predicted in patients with extensive defects and severe shunts than in those without underlying disease, where the shunt is less significant than the ventilation-to-blood-supply ratio, which is present in many cyanotic cardiac patients^20,23,24.

Alveolar dead space is typically linked to larger values of the gap between arterial and end-tidal CO₂ in cyanotic patients and greater and more intense venous admixture. Additionally, a number of earlier investigations have demonstrated the impact of elevated pulmonary alveolar proteinosis (PAP) and pulmonary blood flow on gas exchange. These studies showed that the rise in PAP may play a significant role in the imbalance between blood supply and ventilation. Because of this, the right-to-left shunt or increased pulmonary blood flow may not be the only cause of the discrepancy between arterial and end-tidal CO₂ in cyanotic children; an increase in PAP may also have a negative impact on the balance and uniformity of blood supply in ventilation process, which can cause a disruption in the hemostasis of CO₂ ^20,25,26.

The gap between arterial and end-tidal CO₂ was smaller after the catheterization procedure in cyanotic heart disease patients than it was before (9.2 vs. 18.07). It appears that the ventilation and blood supply in these patients have been significantly improved, and the exchange of oxygen with the environment is more favorable, given the increase in arterial blood oxygen in these patients after the catheterization procedure compared to before (55 vs. 71.4) and also the decrease in arterial CO₂ after the catheterization procedure in these patients compared to before (40.2 vs. 43.93). Additionally, given that these patients’ congenital cardiac abnormalities have been treated, it is anticipated that their venous admixture will greatly improve. Venous admixture and the ventilation-to-blood-supply ratio both affect the difference between arterial and end-tidal CO₂ ^20,27, and the improvement of these factors may be the cause of the decrease in this difference after treating patients’ valvular defects. End-tidal CO₂ will be a more accurate indicator of arterial CO₂ after this flaw has been fixed.

Before the catheterization procedure, there was no statistically significant link between end-tidal and arterial CO₂ in cyanotic patients (r=0.411, P=0.128), whereas there was a statistically significant link (r=0.617, P=0.014) after the catheterization procedure. Therefore, it would appear that end-tidal CO₂ does not provide a good estimate of the arterial CO₂ level prior to catheterization procedure in patients with cyanotic heart disease, but that by correcting the background defect, it can be a trustworthy signal in the evaluation of CO₂. In this group of individuals, arterial CO₂ should be utilized.

There was a substantial connection between end-tidal and arterial CO₂ in patients who did not have cyanosis prior to catheterization procedure (r=0.576, P=0.001), indicating that end-tidal CO₂ can be used as an accurate indicator of arterial CO₂ in these patients. This result is in line with other researches like the paper by Lazzell and Burrows^28, who also noted a respectable relationship between end-tidal and arterial CO₂ in acyanotic patients. Because the alveolar dead space is frequently small and gas exchange occurs optimally in these individuals^20,29, the difference between end-tidal and arterial CO₂ is typically quite minimal in these patients.

This study is among the few studies evaluating the relation between end-tidal and arterial CO₂ in patients with congenital heart diseases, particularly those who underwent surgical procedures with a prospective method. However, our survey was not without limitations. We only checked the end-tidal and arterial CO₂ levels two times, before and after the catheterization procedure. Additionally, PAP and the volume of pulmonary blood flow – which may have an impact on patients’ end-tidal CO₂ – were not examined in this study. Further investigations including these factors may be helpful in the future.

Conclusion

In non-cyanotic patients, end-tidal CO₂ can provide a reasonable estimate of arterial CO₂. There is no significant correlation between end-tidal and arterial CO₂ in cyanotic patients, indicating that end-tidal CO₂ cannot be a good estimate of arterial CO₂. However, after the repair of the cardiac defect, the difference between end-tidal and arterial CO₂ decreases, and end-tidal CO₂ can be a good predictor of arterial CO₂.

Ethical approval

The study protocol was approved by the ethics committee of Tehran University of Medical Sciences (Ethics Code: IR.TUMS.MEDICINE.REC.1397.470).

Consent

Informed consent by the parents was obtained prior to entering the study.

Sources of funding

This study was not funded.

Author contribution

All authors made substantial contributions to the conception and design of the study, and drafting the article and revising it critically for important intellectual content and performed final approval of the version.

Conflicts of interest disclosure

The authors declare that they have no conflicts of interest to declare.

Research registration unique identifying number (UIN)

1. Name of the registry: A. Research Proposal Information System. B. Iran National Committee for Ethics in Biomedical Research.

2. Unique identifying number or registration ID: A. 36132. B. IR.TUMS.CHMC.REC.1397.026.

3. Hyperlink to your specific registration (must be publicly accessible and will be checked): A. https://rpis.research.ac.ir/Researcher.php?id=607215. B. https://ethics.research.ac.ir/PortalProposalList.php?code=IR.TUMS.CHMC.R EC.1397.026&title=&name=&stat=&isAll=&GlobalBackPage=https%3A%2 F%2Fwww.google.com%2F

Guarantor

Mehrdad Goudarzi, MD.

Data availability statement

Datasets generated and analyzed during the current study are available upon reasonable request from the corresponding author, MG.

Provenance and peer review

Not commissioned, externally peer-reviewed.