Correlation Between End-Tidal Carbon Dioxide and Regional Cerebral Oxygen Saturation During Cardiopulmonary Resuscitation

Abstract

Background/Objectives: Near-infrared spectroscopy (NIRS) enables the non-invasive assessment of cerebral oximetry, offering insights into the efficacy of oxygen supply to the brain. NIRS, when combined with other monitoring techniques such as capnography, may play a crucial role in advanced patient monitoring during sudden cardiac arrest and post-resuscitation treatment. This research assessed the relationship between end-tidal carbon dioxide (ETCO₂) and regional cerebral oxygen saturation (rSO₂) during cardiopulmonary resuscitation. Methods: The research was performed from 11 January 2023 until 31 January 2024, at the University Hospital in Poland. The cohort of responders included patients who had in-hospital cardiac arrest (IHCA). The Rapid Response Team attached the rSO₂ and ETCO₂ monitoring devices to each patient during cardiopulmonary resuscitation (CPR). The cohort included 104 patients. Results: The correlation coefficient between ETCO₂ and rSO₂ values was 0.641 (95% CI: 0.636–0.646), and during the last 4 min of CPR before ROSC, it was 0.873 (95% CI: 0.824–0.910). Conclusions: The positive correlation between ETCO₂ and rSO₂ may suggest that concurrent monitoring of both parameters during resuscitation might serve as a valuable predictor of CPR efficacy and the likelihood of achieving recovery of spontaneous circulation in a multimodal framework. In the lack of rapid ETCO₂ monitoring capabilities, rSO₂ may function as a simple and effective alternative for assessment.

1. Introduction

Cardiac arrest refers to the interruption of cardiac activity and the disruption of blood circulation. It may arise suddenly and can be linked to several conditions. If cardiac arrest persists for a prolonged period without the heart resuming activity, all essential functions cease. Consequently, it is crucial to restart the heart to sustain life. Cardiac arrest may transpire inside a healthcare institution, such as a hospital, or in an external environment. Cardiac arrest is a significant public health challenge globally. The yearly incidence of in-hospital cardiac arrest (IHCA) in the United States is estimated to be over 300,000 cases. This varies from 1.6 to 2.85 per 1000 hospital admissions, according to statistics from the United Kingdom and the United States, respectively [1]. The yearly incidence of out-of-hospital cardiac arrest (OHCA) in Europe ranges from 67 to 170 per 100,000 population, and over 350,000 incidents occur annually in the United States [2,3]. Prioritizing high-quality chest compressions, defibrillation, and the identification and treatment of reversible causes is essential in managing both IHCA and OHCA. A key factor of effective cardiopulmonary resuscitation (CPR) is the assessment of end-tidal CO₂ (ETCO₂) [4]. Carbon dioxide is produced in tissues as a result of aerobic metabolism, diffusing from cells into the blood, is transported via the venous route to the lungs, and then is removed by ventilation. The factors that determine the ETCO₂ measurement results are CO₂ production, pulmonary perfusion, alveolar ventilation, and cardiac output. Its accurate value is 35–45 mmHg [5]. According to current knowledge, capnography provides information on the effectiveness of chest compressions due to the positive correlation of ETCO₂ with cardiac index, coronary perfusion pressure, and cerebral blood flow [6,7]. However, the value of ETCO₂ during resuscitation may be affected by various factors, such as the supply of adrenaline or other drugs from the group of vasopressors, which can lead to a temporary decrease in the value of ETCO₂, whereas the administration of sodium bicarbonate results in a marked increase in this parameter. This parameter is also markedly reduced during CPR in the presence of pulmonary embolism, and an ETCO₂ over 10 mmHg during sudden cardiac arrest (SCA) enhances the likelihood of return of spontaneous circulation (ROSC) and survival, but its utility has not yet been fully confirmed. Therefore, ETCO₂ should be considered as an element of the multimodal approach in CPR monitoring [4,8,9,10]. It is also important to recognize that to accurately evaluate this parameter, the airways must be adequately cleared, potentially causing a considerable delay in its measurement during the early phase of CPR.

Near-infrared spectroscopy (NIRS) is used to assess the efficacy of microcirculation inside a specific organ, including principles of quantum mechanics and nonlinear optics. The measurement involves positioning the sensor on the patient’s forehead using a specialized electrode. Regional cerebral saturation (rSO₂) ranges from 70% to 80% venous blood saturation. Oxygen saturation of venous blood is an indicator reflecting the balance between the demand and use of oxygen in tissues. Consequently, NIRS is used to evaluate organ perfusion and tissue oxygen consumption. Research validates its efficacy in tracking the progression of CPR; nonetheless, the findings remain unclear [11,12].

This research aimed to evaluate the association between ETCO₂ and rSO2 during cardiopulmonary resuscitation. Our results may facilitate the correlation between these parameters, thereby enhancing the efficiency of CPR quality monitoring and the prediction of ROSC. Furthermore, we aim to rectify the inadequacy in the multimodal approach. The simple installation of the rSO₂ sensor enables prompt monitoring from the beginning of CPR. This study is the only known assessment in which both measurements were simultaneously collected from a single monitoring device.

2. Materials and Methods

The research was conducted from 11 January 2023 to 31 January 2024, at the University Hospital in Krakow, Poland. The cohort of responders included individuals who had sudden cardiac arrest while hospitalized. The Ethics Committee of Kielce, Poland, authorized the research (number 11/2023). The Rapid Response Team (RRT) established a connection between the rSO₂ and end-tidal carbon dioxide concentration monitoring devices for each patient within 5 min of diagnosing in-hospital cardiac arrest.

2.1. Inclusion and Exclusion Criteria

Inclusion criteria included sudden cardiac arrest in a hospitalized patient, age over 18 years, and patients in normothermia.

Exclusion criteria included no connection of the rSO₂ and ETCO₂ sensor or malfunction (electrode or device) during CPR, patients admitted to the hospital during CPR, cardiac arrest occurring in hospitalized patients in the ICU, cardiac arrest occurring in the operating theater, placement of the rSO₂ monitoring sensor more than 5 min after the occurrence of cardiac arrest, patients suspected of internal or external hemorrhage, and information regarding palliative care.

2.2. Cardiopulmonary Resuscitation

The CPR procedure was performed in accordance with the 2021 Guidelines of the European Resuscitation Council. After receiving the report and reaching the call site, where the patient’s IHCA took place, the RRT took over the management of CPR from the nursing and medical staff of the department. The RRT comprises a specialist in anesthesiology and critical care, a physician in training in anesthesiology and intensive care, and two paramedics. Throughout the research, all team members had valid certification in Advanced Resuscitation Procedures (ALS). The electrodes for rSO₂ monitoring were positioned following the manufacturer’s guidelines by a selected trained team member, about 1 cm above the patient’s eyebrows on the forehead. The ETCO₂ was ultimately monitored by a sensor positioned between the bacteriostatic filter of the intubation tube and the self-inflating bag or ventilator. The rSO₂ and ETCO₂ measurements were recorded at 2 sec intervals using the Masimo Open Connect MOC-9 Module. Ventilation was performed asynchronously via a self-inflating bag or a CPR ventilator at a rate of 10 breaths per minute. The whole process of in-hospital cardiopulmonary resuscitation has been recorded in accordance with the Utstein Resuscitation Registry Template for the IHCA protocol [13]. The mean time from cardiac arrest recognition to initiation of basic life support (BLS) by ward staff was estimated at under 1 min. An automated external defibrillator (AED) was available in all monitored units, and defibrillation was performed when indicated prior to RRT arrival. The Rapid Response Team (RRT) arrived at the scene with a median response time of 4.0 min (IQR: 3.0–4.9 min). Advanced life support was initiated immediately upon arrival. Endotracheal intubation was performed without delay, and mechanical ventilation or self-inflating bag ventilation was provided asynchronously at 10 breaths per minute with 100% FiO₂. The ETCO₂ sensor was placed within 4 min after RRT arrival, and rSO₂ electrodes were applied simultaneously. All procedures strictly followed the 2021 ERC Guidelines for in-hospital cardiac arrest management.

2.3. Statistical Analysis

We employed the Shapiro–Wilk test to evaluate the distribution of the gathered data for each variable (ETCO₂ and rSO₂) in both the ROSC and non-ROSC groups. Furthermore, histogram plots were examined to visually assess the distribution’s form. Both the statistical tests and visual assessments demonstrated that the data for rSO₂ and ETCO₂ in both groups conformed to a normal distribution, thereby justifying the application of parametric methods for subsequent analysis.

We employed the Student’s t-test to compare the two independent groups (ROSC vs. non-ROSC) based on the distribution assessment. A generic linear model for repeated measures was utilized to evaluate temporal variations in measurements during CPR. A post hoc analysis employing Bonferroni correction was performed to evaluate pairwise comparisons among designated time points: M4 (mean value 4 min prior to ROSC), M3, M2, and M1 (1 min prior to ROSC).

We evaluated the correlations between ETCO₂ and rSO₂ using the Pearson linear correlation coefficient, classifying them as follows: 0–0.19 = destitute; 0.20–0.39 = poor; 0.40–0.59 = moderate; 0.60–0.79 = strong; 0.80–1.0 = forceful correlation.

Missing data (≤10% per patient) were addressed by linear interpolation, utilizing the mean of the neighboring values preceding and succeeding the gap, based on the assumption of missing at random (MAR).

Descriptive statistics were presented as means (standard deviation, SD), medians (interquartile range, IQR), and 95% confidence intervals (CIs) when applicable. A p-value less than 0.05 was deemed statistically significant. All statistical analyses were conducted utilizing IBM SPSS Statistics, version 29.0 for macOS.

2.4. Confounding Variables

The supply of sodium bicarbonate (NaHCO₃) was considered a factor that could have a significant impact on the ETCO₂ result [5,14]. A period of 7 min after drug administration was excluded from the analyses. The drug was administered to 21 patients.

3. Results

The total duration of the analyzed signals was 1776 min. The mean value of rSO2 (SD) during CPR for patients who achieved ROSC was 63.8% (7.4), and for patients who did not achieve ROSC, it was 35.6% (5.6) (p < 0.001). The mean (SD) ETCO2 value in the ROSC group compared to the non-ROSC group was 26 mmHg (4.9) vs. 17 mmHg (4.8) (p < 0.001). The characteristics of the patients are presented in

Table 1. Characteristics of patients divided into study groups (non-ROSC, ROSC).

Variables		IHCA (n = 104)
		Non-ROSC (n = 50)	ROSC (n = 54)
Sex	Male (%)	34 (68%)	32 (59%)
Age (years)	Me (IQR)	69 (62; 75)	70 (63; 75)
	<65 (years)	17 (34%)	15 (28%)
	≥65 (years)	33 (66%)	39 (72%)
Duration of resuscitation (min)	Me (IQR)	25 (21; 30)	15 (10; 20)
	Min	15	5
	Max	40	31
rSO2 (%)	M (SD)	35.6 (5.6)	63.8 (7.4)
	Min	19	38
	Max	56	84
ETCO2 (mmHg)	M (SD)	17 (4.8)	26 (4.9)
	Min	4	11
	Max	40	64
Initial rhythm	Asystole (%)	20 (40%)	13 (24%)
	PEA (%)	24 (48%)	30 (56%)
	VF (%)	5 (10%)	7 (13%)
	pVT (%)	1 (2%)	4 (7%)

Legend: ROSC—return of spontaneous circulation; n—group size; SD—standard deviation; M—mean; Me—median; Min—minimum value; Max—maximum value; IQR—25 quartile, 75 quartile; rSO2—regional cerebral oxygen saturation; ETCO₂—end-tidal carbon dioxide; PEA—pulseless electrical activity; pVT—pulseless ventricular tachycardia; VF—ventricular fibrillation.

. A statistically significant strong positive correlation was observed between ETCO2 and rSO2 values (p < 0.001). The correlation coefficient was 0.641 (95% CI: 0.636–0.646). When the ETCO2 value rises, the rSO2 value also increases. After eliminating data after NaHCO₃ administration (n = 21), the correlation coefficient increased to 0.648 (95% CI: 0.643–0.652, p < 0.001). For the group of patients without ROSC, the correlation coefficient was 0.317 (95% CI: 0.307–0.327; p < 0.001), indicating a slight positive correlation. After excluding data after NaHCO₃ administration (n = 6), the correlation coefficient increased to 0.319 (95% CI: 0.309–0.329) (p < 0.001). In the ROSC group, the correlation coefficient was 0.228 (95% CI: 0.215–0.240) (p < 0.001), indicating a slight positive correlation. After excluding the growth phases for ETCO2 resulting from the administration of NaHCO3 (n = 15), the correlation increased to 0.253 (95% CI 0.241–0.266) (p < 0.001). The distribution of initial rhythms between the ROSC and non-ROSC groups revealed a higher proportion of shockable rhythms (VF or pVT) in the ROSC group compared to the non-ROSC group (20% vs. 12%, respectively). Conversely, non-shockable rhythms (asystole and PEA) were predominant in both groups but significantly more common in the non-ROSC group (88%) than in the ROSC group (80%).

Correlation Between rSO2 and ETCO2 Prior to ROSC

The last 4 min before the ROSC in the patient group were evaluated to determine whether a rapid elevation in both measures might indicate ROSC. The average values of rSO₂ and ETCO₂ were assessed over four intervals. The values are designated as M4, M3, M2, and M1. Statistically significant variations were observed between consecutive rSO₂ and ETCO₂ observations. No variation in ETCO₂ was seen between the mean values at the 2nd and 3rd minutes of resuscitation before the onset of ROSC. The data indicate that both rSO₂ and ETCO₂ levels increased as they neared the ROSC finding. The rSO₂ value between M4 and M1 (64.6% vs. 70.4%) exhibited a variation of 5.8 percentage points, representing a 9% increase. The ETCO₂ measurement between M4 and M1 (26.2 mmHg vs. 32.2 mmHg) showed a variation of 6 mmHg, corresponding to a change of 23%. ETCO₂ exhibited a more significant variation than rSO₂ in the last 4 min before the achievement of ROSC. The correlation coefficient of ETCO₂ to rSO₂ in patients who achieved ROSC in the last 4 min of CPR was 0.873 (95% CI: 0.824–0.910), indicating a robust correlation and demonstrating that both parameters rise as the return of circulation, confirmed by the pulse in the central arteries, is approached (

Table 2. Average values of rSO₂ and ETCO₂ from the last 4 min of resuscitation.

ROSC n = 54
Variable	M4 (SD)	M3 (SD)	M2 (SD)	M1 (SD)	p for Post Hoc Test
rSO2 (%)	64.6 (6.9)	65.8 (6.4)	68 (6.3)	70.4 (5.6)	M4–M3 p < 0.001 M3–M2 p < 0.001 M2–M1 p = 0.003
ETCO2 (mmHg)	26.2 (4.6)	28.2 (5.4)	28.8 (5.8)	32.2 (5.1)	M4–M3 p < 0.001 M3–M2 p = 1.000 M2–M1 p < 0.001

Legend: rSO₂—regional cerebral oxygen saturation; ETCO₂—end-tidal carbon dioxide; ROSC—return of spontaneous circulation; SD—standard deviation; M—mean; M4—indicates the average from 4th minute before the occurrence of ROSC; M3—indicates the average from 3rd minute before the occurrence of ROSC; M2—indicates the average from the 2nd minute before the occurrence of ROSC; M1—indicates the average from the last minute before the occurrence of ROSC.

, Figure 1).

In contrast to the ROSC group, the non-ROSC group exhibited only minimal increases in rSO₂ and ETCO₂ values across the M4 to M1 time points, with a weak correlation (r = 0.317; 95% CI: 0.307–0.327). These findings are summarized in

Table 3. Mean rSO₂ and ETCO₂ values for ROSC and non-ROSC groups over the last 4 min of CPR.

Group	Parameter	M4 (Mean ± SD)	M3	M2	M1	ΔM4–M1	Pearson r
ROSC (n = 54)	rSO2 (%)	64.6 ± 6.9	65.8	68.0	70.4	+5.8	0.873
	ETCO2 (mmHg)	26.2 ± 4.6	28.2	28.8	32.2	+6.0	0.873
non-ROSC (n = 50)	rSO2 (%)	34.7 ± 5.2	34.9	35.2	35.3	+0.6	0.317
	ETCO2 (mmHg)	16.3 ± 4.1	16.5	16.6	16.8	+0.5	0.317

Legend: ΔM4–M1—change from 4th minute to 1st minute before ROSC assessment; Pearson r—correlation coefficient for M4–M1 in each group.

and reinforce the hypothesis that a dynamic rise in both rSO₂ and ETCO₂ is associated with the successful return of spontaneous circulation.

4. Discussion

The results underscore that the correlation of rSO₂ with ETCO₂ during resuscitation may serve as an essential tool for enhancing the resuscitation process, monitoring patient condition, and increasing the chance of successful resuscitation and favorable neurological outcomes. In the context of IHCA, when CPR is performed by a nurse–physician team, the use of NIRS monitoring enables the tracking of rSO₂ values from the first minutes of CPR, prior to the availability of ETCO₂ measurements [15]. Monitoring physiological signals during CPR may improve the assessment of resuscitation quality and help predict ROSC and long-term neurological outcomes. Although comparisons with out-of-hospital cardiac arrest (OHCA) research provide useful insights into the general physiology of resuscitation, it is crucial to interpret these data cautiously. IHCA and OHCA markedly differ regarding response time, initial rhythm prevalence, pre-existing comorbidities, and monitoring capacity. Consequently, the extension of OHCA findings to the IHCA group in our investigation is intended just to furnish physiological context, rather than to establish direct clinical equivalency.

Capnography, although a non-invasive technique to measure ETCO₂ during CPR, necessitates an invasive procedure such as endotracheal intubation or using the epiglottic approach. A recent secondary analysis of a large OHCA trial indicated that ETCO₂ trends over time provide prognostic information. ETCO₂ values were significantly higher in patients who achieved ROSC compared to those who did not, as resuscitation progressed, and an increasing ETCO₂ slope was independently linked to ROSC [16]. These data confirm that capnography offers immediate feedback on systemic perfusion and can direct resuscitative measures. Nonetheless, ETCO₂ possesses limitations as a predictive measure. Its measurements may be distorted by alterations in ventilation (rate and tidal volume), airway condition, pulmonary disease, or metabolic influences [7,17]. Excessive ventilation may reduce ETCO₂ levels despite sufficient perfusion, while CO₂ accumulation during extended downtimes or bicarbonate administration might temporarily increase ETCO₂ irrespective of circulation. Consequently, although ETCO₂ is essential for monitoring CPR, it offers restricted prognostic accuracy when utilized alone [17]. Moreover, the sensor may often get contaminated with respiratory secretions, perhaps resulting in inaccurate readings or temporary failures. Extreme values or distinct trends (consistent increase or abrupt decrease) provide more insights than any singular absolute measurement, and additional research is required to develop reliable baselines for clinical decision-making. Experimental studies have shown that during CPR, ETCO2 correlates well with the Cardiac Index (0.79; p < 0.001) [16] with coronary perfusion pressure (0.78; p < 0.001) [16] and blood flow through the brain (0.64; p  = 0.01) [7]. The published document of the American Heart Association recommends monitoring ETCO₂ as the basic indicator during CPR and suggests that a target ETCO₂ > 20 mmHg should be pursued [18].

NIRS is commonly used to optimize general anesthesia during open-heart surgery [19,20] for patients with head traumas [21] and during carotid endarterectomy [22]. The efficacy of NIRS has been validated in clinical scenarios with increased risk of cerebral ischemia, and in recent years, rSO₂ has been used to monitor venous–arterial and venous–venous extracorporeal membrane oxygenation [23]. Numerous observational studies indicate that the mean rSO₂ in patients who achieve ROSC is significantly elevated compared to those who do not [18,19]. In the study by Parnia et al., the average rSO₂ during CPR was approximately 52% in patients who achieved ROSC, compared to around 41% in those who did not. Moreover, Parnia et al. indicated that patients exhibiting favorable neurological status at discharge (Cerebral Performance Category 1–2) demonstrated a significantly elevated mean rSO₂ during CPR (~56%) in contrast to those with adverse outcomes (~44%). Additionally, the proportion of CPR duration exceeding a crucial rSO₂ threshold is also indicative: In that study, sustaining rSO₂ > 50% for ≥60% of the resuscitation correlated with a 98% negative predictive value for unfavorable outcomes [20]. These data indicate that sufficient cerebral perfusion during CPR correlates with less neurological damage. Conversely, ETCO₂ has not been demonstrated to forecast neurological outcomes directly beyond its function in attaining ROSC, as the restoration of circulation is a precondition for brain recovery. It is essential to recognize that post-resuscitation treatment and additional factors eventually influence neurological outcomes. A review of 26 studies indicated that the pooled mean rSO₂ in patients with ROSC was approximately 41%, in contrast to about 30% in those without ROSC (p = 0.009) [12]. Importantly, cerebral oximetry trends appear to carry prognostic value. Patients who achieve ROSC exhibit a more significant increase in rSO₂ during continuous CPR compared to those who do not. In a substantial cohort of OHCA cases, the median increase in rSO₂ was 17% in cases with a ROSC, compared to 8% in other cases. An increase greater than 15% from baseline was associated with an odds ratio of approximately 4.9 for predicting ROSC [20]. A sudden increase in rSO₂, such as from the 20–30% range to over 50% within a few minutes, may indicate ROSC, similar to the spike in ETCO₂, and can be identified without the necessity of an arterial pulse or advanced airway [20,21]. The findings align with these trends: Successful resuscitation cases exhibited elevations in rSO₂ toward the conclusion of the process, while consistently low rSO₂ indicated a failure in resuscitation efforts. Our data, along with previous studies, indicate that rSO₂ may serve as a more sensitive early marker of perfusion return compared to ETCO₂.

In a study conducted in an emergency department on cardiac arrest, cerebral oximetry demonstrated superior performance compared to ETCO₂ in predicting ROSC, with an area under the curve of 0.89 versus 0.77 in the final minute of CPR [22]. Our IHCA cohort exhibited enhanced predictive performance of rSO₂ compared to ETCO₂, evidenced by a significantly higher area under the curve for ROSC. The findings suggest that cerebral saturation serves as a direct measure of end-organ (brain) perfusion, which is influenced by cardiac output and arterial oxygen content. If CPR produces any forward flow, rSO₂ will increase, assuming that the oxygen delivered is adequate, thereby reflecting the sufficiency of both circulation and oxygenation [23,24]. ETCO₂ indicates the delivery of CO₂ to the lungs. Even with improved perfusion, it may not rise if ventilation is poor or metabolic CO₂ production is low. Combining both metrics provides a more comprehensive understanding: In our study, every patient who achieved ROSC exhibited improvements in both ETCO₂ and rSO₂, whereas those without ROSC demonstrated corresponding declines. The result suggests a strong relationship between systemic and cerebral perfusion during CPR, consistent with findings from experimental models that demonstrate an increasing correlation between ETCO₂ and rSO₂ values over the course of resuscitation, nearing r ≈ 1.0 in the later stages of CPR [25]. NIRS, when combined with other monitoring techniques like capnography, may play a crucial role in the advanced monitoring of critically ill patients and those experiencing SCA, offering insights into hemodynamic state and cerebral perfusion [11]. Moreover, the simplicity of sensor installation might be significantly crucial when ETCO₂ measurement is not easily accessible.

There is a very limited number of studies that determine the correlation between ETCO₂ and rSO₂. This may be due to the limited availability of devices that have the ability to measure both of these parameters at the same time. The device used in our research has two modules, enabling the simultaneous monitoring of these two indicators and the collection of data from the same moment of measurement. We observed a strong correlation between rSO₂ and ETCO₂ during CPR, suggesting that both metrics reflect the underlying hemodynamic quality. Research using a porcine model established a significant correlation between ETCO₂ and cardiac output during resuscitation, with a value of 0.83 (p < 0.001) (95% CI: 0.67–0.92), and a positive correlation between rSO2 and cardiac output during resuscitation, with a value of 0.50 (p = 0.004) (95% PU: 0.16–0.73). In this study, ETCO₂ demonstrated a more robust association than rSO₂. Researchers underscored the need to monitor both values in a hemodynamically unstable patient for the prompt detection of cardiac arrest, hence facilitating the early commencement of life-saving interventions [26]. Moreover, there are pivotal situations in which one monitor may identify problems that the other fails to recognize. For instance, if an endotracheal tube is dislodged or the patient experiences hyperventilation, ETCO₂ will sharply decrease despite the continuity of chest compressions (and consequently cerebral perfusion); rSO₂ could inform clinicians that brain oxygenation is still sufficient or is declining gradually during that period. Conversely, an elevation in ETCO₂ without a corresponding increase in rSO₂ may raise concerns regarding inadequate oxygen delivery or cerebral perfusion; for example, CO₂ may be circulating from peripheral tissues, but the brain is not receiving sufficient blood flow or oxygen content. Using both methods together helps confirm that brain and systemic blood flow are adequate. Resuscitation experts have proposed a combination technique as part of “physiology-guided” CPR, in which ETCO₂, arterial pressure, and, if accessible, cerebral oximetry are adjusted in real time to enhance perfusion [7]. Significantly, in contrast to capnography, NIRS monitoring does not necessitate the cessation of chest compressions or the establishment of an advanced airway, allowing for its implementation from the initiation of resuscitation, including during bag-mask breathing [5,21]. The strategy may be especially beneficial in hospital environments where NIRS devices are more readily available (such as operating rooms and ICUs) and where qualified workers can analyze various data streams. By monitoring both ETCO₂ and rSO₂, the team can more reliably detect ROSC, as a simultaneous increase in both metrics serves as a robust signal of ROSC, hence minimizing premature interruptions for pulse assessments. In our IHCA instances, we saw that dual monitoring offered confidence when both levels improved concurrently and indicated possible issues when they varied. It is important to acknowledge that while the qualitative integration of ETCO₂ and rSO₂ enhances monitoring, the additional prognostic significance of their combined use remains under investigation. In a porcine arrest model, the integration of rSO₂ and ETCO₂ did not markedly improve predictive accuracy for favorable neurologic outcomes compared to rSO₂ or ETCO₂ individually [25], mostly because both metrics provided analogous information at the conclusion of CPR. Similarly, in clinical investigations, the measure that first attains an extreme threshold (either extremely low or very high) frequently prompts clinical actions, complicating the demonstration that the second parameter influences outcomes. Nonetheless, a multimodal approach is advantageous for personalized care, as it enables responders to identify and rectify certain deficiencies in CPR. For instance, if ETCO₂ is diminished despite satisfactory rSO₂, clinicians may prioritize optimizing ventilation or assessing the airway; conversely, if rSO₂ is reduced despite sufficient ETCO₂, attention may be directed towards enhancing cerebral perfusion (e.g., improved chest recoil, vasopressor administration, or head positioning to augment cerebral blood flow). This personalized approach shows the future of resuscitation: moving from a standard protocol to goal-directed CPR using real-time data [7].

The current AHA and European guidelines do not endorse cerebral oximetry for regular neuroprognostication due to inadequate data [23]. Nevertheless, near-infrared monitoring continues to serve as a research instrument for the early detection of brain perfusion, and exceedingly low rSO₂ values during CPR (e.g., consistently < 20–25%) may assist in identifying patients with a significant probability of severe neurological impairment or futility [19]. Future procedures may integrate rSO₂ with existing post-arrest prognostic evaluations, including neurological examinations and EEG, in a multimodal approach.

The positive correlation we have shown indicates that both of these parameters are congruent with one another. However, ETCO₂ can be affected by many confounding variables [5]. We have shown that in the group of patients achieving ROSC, both these parameters maintain elevated levels throughout the resuscitation period, in contrast to the group without ROSC. In the research conducted by Kämäräinen et al., cerebral saturation was assessed alongside resuscitation metrics, including the depth and frequency of chest compressions and ventilation rate. It was observed that the rSO₂ value remained diminished, even during high-quality CPR [27,28,29]. Our study also verifies that rSO2 levels are decreased in patients who have not attained ROSC and elevated throughout the CPR period in the ROSC group. The assessment of rSO₂ seems to be a more stable measure than ETCO₂ and is unresponsive to pharmacological interventions or ventilation; nonetheless, more study is required [30]. A crucial result is that a substantial simultaneous rise in both measures may serve as a predictor of ROSC during the last period of CPR and might significantly enhance the multimodal monitoring method.

Limitations

This research offers significant insights; nonetheless, it is crucial to recognize numerous limitations. The restricted sample size and single-center nature of this research are significant limitations that could limit the generalizability of the findings and diminish the statistical power to detect deeper differences. Furthermore, the research assumed that CPR performed by the RRT was carried out with adequate chest compressions and ventilation. Occasionally, there may have been unrecorded temporary disruptions in chest compressions or ventilation caused by intubation or other resuscitative interventions during ALS. NIRS mainly assesses the superficial layers of the brain, and when comparing studies with other researchers, it is essential to identify the kind of equipment used, since different models may provide disparate findings. The intervals of adrenaline administration, which may influence ETCO₂ levels without impacting rSO2, were also not taken into account. Another restriction is the exclusive focus on ROSC, neglecting neurological prognosis and long-term prognostic evaluation. It is crucial to highlight that both rSO₂ and ETCO₂ are surrogate indicators influenced by various interrelated and independent physiological processes. These encompass cardiac output, pulmonary blood flow, alveolar-capillary gas exchange, CO₂ generation at the tissue level, and regional cerebral perfusion dynamics. All of these measures are affected by breathing techniques, metabolic condition, vasopressor administration, and underlying pathology. While we concentrated on the direct relationship between rSO₂ and ETCO₂, a more extensive multiparametric model that includes simultaneous measurements—such as cardiac output (e.g., through echocardiography or invasive pressure monitoring), arterial oxygen saturation (SpO₂), FiO₂, lactate concentrations, or cerebral perfusion indices—might provide enhanced predictive capability for ROSC and neurological outcomes. Future research should incorporate these characteristics to enhance real-time prognostication and improve physiology-guided CPR techniques.

5. Conclusions

The strong correlation between ETCO₂ and rSO₂ may suggest that simultaneous monitoring of both parameters during resuscitation might serve as a valuable predictor of CPR efficacy and the likelihood of achieving ROSC in a multimodal approach. In the lack of rapid ETCO₂ monitoring capabilities, rSO₂ may function as a simple and effective alternative for monitoring.