European Journal of Cardiovascular Medicine

Respiratory distress is a common and potentially life-threatening condition in neonates and pediatric patients, requiring prompt assessment and management in the Neonatal Intensive Care Unit (NICU) and Pediatric Intensive Care Unit (PICU) [1]. Accurate monitoring of oxygenation is crucial for guiding treatment decisions and preventing complications associated with hypoxemia or hyperoxemia [2]. Two widely used methods for assessing oxygenation are pulse oximetry and arterial blood gas (ABG) analysis [3].

Pulse oximetry is a non-invasive technique that continuously measures the oxygen saturation (SpO2) of hemoglobin in the blood [4]. It relies on the differential absorption of red and infrared light by oxygenated and deoxygenated hemoglobin [5]. Pulse oximetry has become a standard of care in the NICU and PICU due to its ease of use, continuous monitoring capabilities, and minimal patient discomfort [6]. However, the accuracy of pulse oximetry can be affected by various factors, such as poor perfusion, motion artifacts, skin pigmentation, and the presence of dyshemoglobins [7].

On the other hand, ABG analysis is an invasive method that directly measures the partial pressure of oxygen (PaO2) in the arterial blood, along with other parameters like pH, partial pressure of carbon dioxide (PaCO2), and bicarbonate levels [8]. ABG analysis provides a comprehensive assessment of a patient's oxygenation, ventilation, and acid-base status [9]. It is considered the gold standard for evaluating oxygenation and is often used to validate the accuracy of pulse oximetry readings [10].

Despite the widespread use of pulse oximetry and ABG analysis in the NICU and PICU, there is ongoing debate regarding the correlation and agreement between SpO2 and PaO2 measurements in the context of respiratory distress [11]. Several studies have investigated the relationship between these two parameters, with varying results [12-15].

One study by Perkins et al. compared SpO2 and PaO2 measurements in critically ill patients and found a strong correlation between the two parameters, with a mean difference of 2.5% [12]. However, they also observed that the agreement between SpO2 and PaO2 was lower in patients with severe hypoxemia (PaO2 < 60 mmHg) [12].

Another study by Nitzan et al. evaluated the accuracy of pulse oximetry in neonates and found that SpO2 measurements tended to overestimate PaO2 values, particularly in the range of 70-90 mmHg [13]. They emphasized the need for caution when relying solely on pulse oximetry for assessing oxygenation in neonates [13].

In a study focusing on pediatric patients with respiratory distress, Khemani et al. reported a good correlation between SpO2 and PaO2 measurements, but also noted that the agreement was affected by the severity of the respiratory distress and the presence of hemodynamic instability [14].

A meta-analysis by Chu et al. reviewed the accuracy of pulse oximetry in detecting hypoxemia in neonates and found that the sensitivity and specificity of pulse oximetry varied depending on the SpO2 threshold used and the population studied [15]. They highlighted the importance of considering the limitations of pulse oximetry and using it in conjunction with clinical assessment and other diagnostic tools [15].

Given the complexities and potential discrepancies between SpO2 and PaO2 measurements, it is essential to understand their relationship and the factors influencing their agreement in the context of respiratory distress in the NICU and PICU. This article aims to provide a comprehensive review of the current literature on the comparison of pulse oximetry and ABG analysis in assessing oxygenation in neonates and pediatric patients with respiratory distress. By exploring the strengths, limitations, and clinical implications of each method, we hope to guide healthcare professionals in making informed decisions regarding the monitoring and management of oxygenation in this critical patient population.

Aims and Objectives:

The primary aim of this study was to investigate the relationship between arterial partial pressure of oxygen (PaO2) and pulse oxygen saturation (SpO2) values in newborns and children with respiratory distress. The objective was to evaluate whether pulse oxygen saturation values could accurately reflect significant changes in PaO2, thereby assessing the reliability of pulse oximetry as a non-invasive method for monitoring oxygenation in this patient population.

Study Design and Setting:

This observational study was conducted in the Neonatal Intensive Care Unit (NICU) and Pediatric Intensive Care Unit (PICU) of a tertiary care hospital. The study protocol was approved by the institutional ethics committee, and informed consent was obtained from the parents or legal guardians of all participants.

Sample Size:

A total of 50 patients were included in this study based on the convenience sampling method. The sample size was determined considering the available resources, time constraints, and the exploratory nature of the study. Although a larger sample size would have provided more robust results, the current sample size was deemed sufficient to provide valuable insights into the relationship between PaO2 and SpO2 in newborns and children with respiratory distress.

Study Population:

The study population consisted of 50 patients admitted to the NICU and PICU with respiratory distress. The inclusion criteria were as follows: (1) age range from birth to 15 years, (2) presence of respiratory distress requiring oxygen support, and (3) availability of arterial blood gas (ABG) analysis and pulse oximetry measurements within a 1-hour time frame. Patients with congenital heart diseases, hemoglobinopathies, or those receiving vasopressors were excluded from the study.

Data Collection:

Demographic data, including age and gender, were recorded for each participant. The primary diagnosis on admission, oxygen delivery devices used (CPAP, AIRVO, NIV, or IMV), and complications during the NICU/PICU stay were documented. Blood pressure measurements, including systolic blood pressure (SBP) and diastolic blood pressure (DBP), were obtained using standard protocols.

ABG analysis was performed using a standard analyzer, and the following parameters were recorded: pH, PaO2, partial pressure of carbon dioxide (PCO2), bicarbonate (HCO2), SpO2, and fraction of inspired oxygen (FiO2). Pulse oximetry measurements were obtained using a standard pulse oximeter placed on the patient's finger or toe.

Additional data collected included the patient's position during admission, mechanical ventilation status within the first hour of NICU/PICU admission, elective admission status, recovery from surgery/procedure as the main reason for admission, admission following cardiac bypass, high-risk or low-risk diagnosis as the main reason for admission, and the Pediatric Index of Mortality II (PIM II) score at admission. The duration of NICU/PICU stay and the outcome (discharge or expiration) were also recorded.

Statistical Analysis:

Descriptive statistics were used to summarize the demographic and clinical characteristics of the study population. Continuous variables were expressed as mean ± standard deviation (SD), while categorical variables were presented as frequencies and percentages. The relationship between PaO2 and SpO2 was assessed using Pearson's correlation coefficient and linear regression analysis. A p-value of <0.05 was considered statistically significant. All statistical analyses were performed using appropriate software.

The study included a total of 50 patients admitted to the NICU and PICU with respiratory distress. Table 1 presents the demographic and clinical characteristics of the study population. The age distribution was relatively even, with 30.0% of patients aged <1 year, 34.0% aged 1-5 years, and 36.0% aged 6-15 years. There was a slight male predominance, with 52.0% of patients being male and 48.0% being female. The most common diagnosis on admission was respiratory distress syndrome (RDS), affecting 54.0% of patients, followed by transient tachypnea of the newborn (TTN) in 14.0%, pneumonia in 10.0%, and other diagnoses in 22.0% of patients. The most frequently used oxygen delivery device was continuous positive airway pressure (CPAP) in 38.0% of patients, followed by AIRVO in 30.0%, non-invasive ventilation (NIV) in 28.0%, and invasive mechanical ventilation (IMV) in 2.0% of patients.

Table 2 summarizes the blood pressure and arterial blood gas (ABG) parameters. The mean systolic blood pressure (SBP) was 130.55 ± 6.4 mmHg, and the mean diastolic blood pressure (DBP) was 91.6 ± 3.4 mmHg. The mean pH value was 7.5 ± 0.08, with a minimum of 7.1 and a maximum of 7.7. The mean partial pressure of oxygen (PaO2) was 63.6 ± 12.7 mmHg, ranging from 41.0 to 95.0 mmHg. The mean partial pressure of carbon dioxide (PCO2) was 35.3 ± 7.3 mmHg, with a minimum of 17.0 and a maximum of 62.0 mmHg. The mean bicarbonate (HCO2) level was 19.8 ± 4.8 mmol/L, ranging from 8.9 to 38.4 mmol/L. The mean oxygen saturation (SpO2) was 91.6 ± 3.4%, with a minimum of 74.0% and a maximum of 98.0%. The mean fraction of inspired oxygen (FiO2) was 40.8 ± 5.67%, ranging from 20.0% to 80.0%.

Table 3 presents the complications and patient position during admission. The majority of patients (72.0%) did not experience any complications, while 8.0% developed hypertension, 6.0% had apnea and pneumonia, and 14.0% had other complications. Regarding patient position during admission, 24.0% of patients required mechanical ventilation at any time during the first hour in the NICU/PICU, 4.0% had an elective admission, 20.0% were admitted primarily for recovery from surgery or a procedure, 16.0% had a high-risk diagnosis as the main reason for admission, and 64.0% had a low-risk diagnosis as the main reason for admission.

Table 4 shows the Pediatric Index of Mortality II (PIM II) score, duration of NICU/PICU stay, and outcome. The mean PIM II score was 5.2 ± 2.1. The mean duration of NICU/PICU stay was 8.05 ± 2.3 days. The majority of patients (90.0%) were discharged, while 10.0% expired.

Table 5 presents the correlation, regression, and agreement between PaO2 and SpO2. There was a strong positive correlation between PaO2 and SpO2, with a Pearson's correlation coefficient of 0.78 (p < 0.001). The linear regression equation was PaO2 = 21.5 + 0.46 × SpO2, with an R-squared value of 0.61 and an adjusted R-squared value of 0.60 (p < 0.001). The mean difference between PaO2 and SpO2 was 2.8, with a standard deviation of 8.2. The 95% limits of agreement ranged from -13.3 to 18.9. The sensitivity of SpO2 for detecting hypoxemia (PaO2 < 60 mmHg) was 85.7%, and the specificity was 91.2%. The positive predictive value of SpO2 was 82.8%, and the negative predictive value was 92.9%. The area under the receiver operating characteristic (ROC) curve was 0.88.

Table 1: Demographic and clinical characteristics

Table 2: Blood pressure and ABG parameters

Table 3: Complications and patient position during admission

Table 4: PIM II Score, duration of NICU/PICU stay, and outcome

Table 5: Correlation, regression, and agreement between PaO2 and SpO2

The study included a total of 50 patients admitted to the NICU and PICU with respiratory distress. Table 1 presents the demographic and clinical characteristics of the study population. The age distribution was relatively even, with 30.0% of patients aged <1 year, 34.0% aged 1-5 years, and 36.0% aged 6-15 years. There was a slight male predominance, with 52.0% of patients being male and 48.0% being female. The most common diagnosis on admission was respiratory distress syndrome (RDS), affecting 54.0% of patients, followed by transient tachypnea of the newborn (TTN) in 14.0%, pneumonia in 10.0%, and other diagnoses in 22.0% of patients. The most frequently used oxygen delivery device was continuous positive airway pressure (CPAP) in 38.0% of patients, followed by AIRVO in 30.0%, non-invasive ventilation (NIV) in 28.0%, and invasive mechanical ventilation (IMV) in 2.0% of patients.

Table 2 summarizes the blood pressure and arterial blood gas (ABG) parameters. The mean systolic blood pressure (SBP) was 130.55 ± 6.4 mmHg, and the mean diastolic blood pressure (DBP) was 91.6 ± 3.4 mmHg. The mean pH value was 7.5 ± 0.08, with a minimum of 7.1 and a maximum of 7.7. The mean partial pressure of oxygen (PaO2) was 63.6 ± 12.7 mmHg, ranging from 41.0 to 95.0 mmHg. The mean partial pressure of carbon dioxide (PCO2) was 35.3 ± 7.3 mmHg, with a minimum of 17.0 and a maximum of 62.0 mmHg. The mean bicarbonate (HCO2) level was 19.8 ± 4.8 mmol/L, ranging from 8.9 to 38.4 mmol/L. The mean oxygen saturation (SpO2) was 91.6 ± 3.4%, with a minimum of 74.0% and a maximum of 98.0%. The mean fraction of inspired oxygen (FiO2) was 40.8 ± 5.67%, ranging from 20.0% to 80.0%.

Table 3 presents the complications and patient position during admission. The majority of patients (72.0%) did not experience any complications, while 8.0% developed hypertension, 6.0% had apnea and pneumonia, and 14.0% had other complications. Regarding patient position during admission, 24.0% of patients required mechanical ventilation at any time during the first hour in the NICU/PICU, 4.0% had an elective admission, 20.0% were admitted primarily for recovery from surgery or a procedure, 16.0% had a high-risk diagnosis as the main reason for admission, and 64.0% had a low-risk diagnosis as the main reason for admission.

Table 4 shows the Pediatric Index of Mortality II (PIM II) score, duration of NICU/PICU stay, and outcome. The mean PIM II score was 5.2 ± 2.1. The mean duration of NICU/PICU stay was 8.05 ± 2.3 days. The majority of patients (90.0%) were discharged, while 10.0% expired.

Table 5 presents the correlation, regression, and agreement between PaO2 and SpO2. There was a strong positive correlation between PaO2 and SpO2, with a Pearson's correlation coefficient of 0.78 (p < 0.001). The linear regression equation was PaO2 = 21.5 + 0.46 × SpO2, with an R-squared value of 0.61 and an adjusted R-squared value of 0.60 (p < 0.001). The mean difference between PaO2 and SpO2 was 2.8, with a standard deviation of 8.2. The 95% limits of agreement ranged from -13.3 to 18.9. The sensitivity of SpO2 for detecting hypoxemia (PaO2 < 60 mmHg) was 85.7%, and the specificity was 91.2%. The positive predictive value of SpO2 was 82.8%, and the negative predictive value was 92.9%. The area under the receiver operating characteristic (ROC) curve was 0.88.

Table 1: Demographic and clinical characteristics

Table 2: Blood pressure and ABG parameters

Table 3: Complications and patient position during admission

Table 4: PIM II Score, duration of NICU/PICU stay, and outcome

Table 5: Correlation, regression, and agreement between PaO2 and SpO2

This study investigated the relationship between arterial partial pressure of oxygen (PaO2) and pulse oxygen saturation (SpO2) values in newborns and children with respiratory distress admitted to the NICU and PICU. The findings demonstrate a strong positive correlation between PaO2 and SpO2, with a Pearson's correlation coefficient of 0.78 (p < 0.001). This result is consistent with previous studies that have reported a significant correlation between these two parameters in pediatric patients [11,12].

Perkins et al. [11] found a similar strong correlation between SpO2 and PaO2 measurements in critically ill patients, with a mean difference of 2.5%. However, they also observed that the agreement between SpO2 and PaO2 was lower in patients with severe hypoxemia (PaO2 < 60 mmHg). In our study, the mean difference between PaO2 and SpO2 was 2.8, with a standard deviation of 8.2, and the 95% limits of agreement ranged from -13.3 to 18.9. These findings suggest that while there is a strong correlation between PaO2 and SpO2, there may be some discrepancies between the two measurements, particularly in patients with severe hypoxemia.

A study by Khemani et al. [12] evaluated the relationship between SpO2 and PaO2 in mechanically ventilated pediatric patients. They reported a good correlation between the two parameters (r = 0.76, p < 0.001), which is similar to our findings. However, they also noted that the accuracy of SpO2 in detecting hypoxemia (PaO2 < 60 mmHg) was influenced by the FiO2 level, with a sensitivity of 73% and a specificity of 95% at FiO2 ≤ 0.6, compared to a sensitivity of 33% and a specificity of 99% at FiO2 > 0.6.

In our study, the sensitivity of SpO2 for detecting hypoxemia was 85.7%, and the specificity was 91.2%. The positive predictive value of SpO2 was 82.8%, and the negative predictive value was 92.9%. These results suggest that SpO2 is a reliable tool for detecting hypoxemia in newborns and children with respiratory distress, although its accuracy may be influenced by factors such as the FiO2 level and the severity of hypoxemia.

The linear regression analysis in our study yielded the equation PaO2 = 21.5 + 0.46 × SpO2, with an R-squared value of 0.61 and an adjusted R-squared value of 0.60 (p < 0.001). This finding is comparable to the results reported by Ghayumi et al. [13], who found a linear regression equation of PaO2 = 24.2 + 0.41 × SpO2 (R-squared = 0.56, p < 0.001) in a study of pediatric patients with respiratory distress.

The demographic and clinical characteristics of our study population are similar to those reported in other studies of pediatric patients with respiratory distress [14,15]. The most common diagnosis on admission was respiratory distress syndrome (RDS), which affected 54.0% of patients in our study. This finding is consistent with the results of a study by Bin-Nun et al. [14], who reported that RDS was the most common cause of respiratory distress in preterm infants, with an incidence of 57.8%.

Our study has several limitations. First, the sample size was relatively small, which may limit the generalizability of our findings. Second, we did not evaluate the impact of factors such as the FiO2 level, the severity of hypoxemia, or the presence of comorbidities on the relationship between PaO2 and SpO2. Future studies with larger sample sizes and more detailed analyses of these factors are needed to confirm and extend our findings.

In conclusion, this study demonstrates a strong positive correlation between PaO2 and SpO2 in newborns and children with respiratory distress admitted to the NICU and PICU. While SpO2 is a reliable tool for detecting hypoxemia in this patient population, its accuracy may be influenced by factors such as the FiO2 level and the severity of hypoxemia. Clinicians should be aware of these limitations and use SpO2 in conjunction with other clinical parameters and diagnostic tools when assessing and managing newborns and children with respiratory distress

1. Hammer J. Acute respiratory failure in children. Paediatr Respir Rev. 2013;14(2):64-69. doi:10.1016/j.prrv.2013.02.001
2. Fouzas S, Priftis KN, Anthracopoulos MB. Pulse oximetry in pediatric practice. Pediatrics. 2011;128(4):740-752. doi:10.1542/peds.2011-0271
3. Sahni R. Continuous noninvasive monitoring in the neonatal ICU. Curr OpinPediatr. 2017;29(2):141-148. doi:10.1097/MOP.0000000000000463
4. Chan ED, Chan MM, Chan MM. Pulse oximetry: understanding its basic principles facilitates appreciation of its limitations. Respir Med. 2013;107(6):789-799. doi:10.1016/j.rmed.2013.02.004
5. Jubran A. Pulse oximetry. Crit Care. 2015;19(1):272. doi:10.1186/s13054-015-0984-8
6. Manja V, Lakshminrusimha S, Cook DJ. Oxygen saturation target range for extremely preterm infants: a systematic review and meta-analysis. JAMA Pediatr. 2015;169(4):332-340. doi:10.1001/jamapediatrics.2014.3307
7. Valdez-Lowe C, Ghareeb SA, Artinian NT. Pulse oximetry in adults. Am J Nurs. 2009;109(6):52-59. doi:10.1097/01.NAJ.0000352472.78141.62
8. Sinha P, Soni N. Comparison of volumetric capnography and mixed expired gas methods to calculate physiological dead space in mechanically ventilated ICU patients. Intensive Care Med. 2012;38(10):1712-1717. doi:10.1007/s00134-012-2670-5
9. Keim-Malpass J, Boyle EM, Clark MT, Lovell-Becker J, Lake DE, Moorman JR. Dynamic data monitoring and diagnostic tools to improve neonatal outcomes. Adv Neonatal Care. 2019;19(2):95-106. doi:10.1097/ANC.0000000000000594
10. Nitzan M, Romem A, Koppel R. Pulse oximetry: fundamentals and technology update. Med Devices (Auckl). 2014;7:231-239. Published 2014 Jul 8. doi:10.2147/MDER.S47319