Introduction
At present, no effective intervention is available for resuscitating patients under general anesthesia. The recovery process relies primarily on the natural elimination of anesthetics from the body, particularly from the brain, which plays a pivotal role in regaining consciousness. Expediting clearance of anesthetics from the brain facilitates prompt emergence of patients from anesthesia.
Hypercapnia exerts dual perioperative physiological effects through distinct mechanisms: (1) cerebrovascular effects mediated by CO2-induced vasodilation, which reduces cerebrovascular resistance and enhances cerebral perfusion [1]; (2) cardiovascular effects driven by sympathetic activation, manifesting as increased heart rate and augmented cardiac output [2]. Some studies have further explored the clinical implications of hypercapnia on the time to emergence from inhaled anesthesia [3–6]. Clinical studies and experimental models have consistently demonstrated that intraoperative hypercapnia can reduce the emergence time after use of volatile anesthetics [7–10].
Nevertheless, few clinical studies have investigated the effects of hypercapnia on emergence after total intravenous anesthesia (TIVA) [11,12]. The findings of those few studies remain controversial and may have been partially influenced by the presence of moderate-to-severe acute pain and opioid analgesics, which can prevent the recovery of patients even when intravenous anesthetics have been cleared. Moreover, intraoperative hypercapnia may enhance cerebral drug clearance through CO2-mediated vasodilation [13], potentially that could necessitate higher anesthetic requirements to maintain target depth of anesthesia.
Previous research has indicated that fluctuations in blood pressure and carbon dioxide levels within the physiological range minimally influence the diameter of the cerebral arteries [14]. Therefore, monitoring changes in cerebral blood flow (CBF) velocity before and after hypercapnia may indirectly reflect the impact of hypercapnia on CBF.
Building on previous findings that mild hypercapnia can enhance CBF and cardiac output (CO), which could potentially affect emergence time from inhaled anesthesia [1,2,4], we hypothesized that mild hypercapnia during the recovery period following TIVA could similarly shorten emergence time. To test this hypothesis, we conducted a randomized controlled trial comparing emergence times—specifically, extubation time, spontaneous breathing-onset time, and voluntary eye-opening time—between a mild hypercapnia group and a normocapnia control group. In addition, changes in CBF velocity were monitored throughout the perioperative period to assess the physiological impact of hypercapnia on cerebral hemodynamics. By exploring these parameters, we sought to gain further insight into whether mild hypercapnia could enhance post-anesthetic recovery in the context of TIVA.
Materials and Methods
Study design
The principles outlined in the Declaration of Helsinki of 2013 were followed throughout this prospective, double-blind, randomized study. This study was approved by the Ethics Review Committee of the Second Affiliated Hospital of Zhejiang University School of Medicine in May 2022 (no. 2022-0403) and registered at ClinicalTrials.gov (no. NCT05401266 [2022-0389]).
We recruited adult patients aged between 18 and 70 years who met the American Society of Anesthesiologists (ASA) class I or II criteria and were scheduled for transurethral lithotripsy under TIVA with endotracheal intubation. The study design and objectives were thoroughly explained to the participants who subsequently provided written voluntary informed consent. The participants were randomly allocated to either the normocapnia or the mild hypercapnia group.
We excluded the following patients: patients with psychiatric, pulmonary, cardiac, endocrine, neuromuscular, hepatic, and neurological conditions, or a history of such conditions; patients with atelectasis, pulmonary infection, or pleural effusion on lung CT; patients taking sedatives, analgesics, or other medications for mental disorders; patients with alcohol- or drug-dependence issues; patients who had undergone general anesthesia within the past 1 month; patients with a body mass index ≥ 30 kg/m2; patients with a difficult airway; and patients whose expected anesthesia time was less than 30 min.
Anesthesia protocol
Prior to anesthesia induction, no sedative or analgesic medications were administered. Upon entering the operating room, the following vital signs of the patients were monitored: noninvasive blood pressure, electrocardiogram, pulse oxygen saturation (SpO2), ETCO2 (USE1503A monitor, GE), and anesthesia depth (ConView YY-106 bispectral index [BIS] monitor, Aibei). Changes in blood flow velocity in the middle cerebral artery (MCA) were monitored using a transcranial Doppler sonography (TCD) blood flow analyzer (EMS–9PB, Delica).
Anesthesia was induced with 100% oxygen preoxygenation at a fresh gas flow rate of 6 L/min, followed by sequential intravenous administration of midazolam (20 μg/kg), sufentanil (0.3–0.6 μg/kg), propofol (1–2 mg/kg), cisatracurium (0.1–0.3 mg/kg), and tracheal intubation using 7.0-mm internal diameter (ID) tubes for males and 6.5-mm ID tubes for females. The anesthesia machine (Datex-Ohmeda Aespire, GE) was connected to a breathing circuit to allow volume-controlled mechanical ventilation with a fresh gas flow rate set at 2 L/min. The fraction of inspired oxygen was maintained at 40%. The respiratory rate ranged from 10 to 12 breaths/min, tidal volume ranged from 6 to 10 ml/kg, and ETCO2 levels were maintained at 35–45 mmHg.
Anesthesia was maintained using propofol infusion at a rate of 60–120 μg/kg/min and remifentanil infusion at a rate of 0.05–0.15 μg/kg/min, while the BIS value was kept within the 40–60 range. Vasoactive agents, such as ephedrine, phenylephrine, or urapidil hydrochloride, were administered as needed to maintain mean arterial pressure (MAP) and heart rate (HR) fluctuations within ± 20% of their pre-induction levels.
Recovery management
According to our institution's routine protocol, patients were transferred to the post-anesthesia care unit (PACU) with an endotracheal tube in place after surgery. Portable ventilators and monitors were used during transportation. No antagonistic agents or arousal stimuli were administered to facilitate smooth recovery and extubation. The ventilation mode was changed to synchronized intermittent mandatory ventilation (SIMV) with oxygen concentrations of 40%–60% and a positive end-expiratory pressure of 5 mmHg (Savina ventilator, Dräger).
The patients were randomly assigned to the two groups upon entering the PACU, at a 1:1 ratio. A simple randomization method was employed, utilizing a computer-generated random number table created by a researcher who was not involved in this study. The randomization results were sealed in opaque envelopes. When a patient met the inclusion criteria and was ready for the trial, the assigned study personnel opened an envelope to determine the group allocation and proceeded with the respective intervention.
Normocapnia group: In the normocapnia group, ventilation parameters were adjusted to maintain the ETCO2 levels between 35 and 40 mmHg until spontaneous breathing returned.
Mild hypercapnia group: In the mild hypercapnia group, ventilation parameters were adjusted (tidal volume reduced to not less than 6 ml/kg and, if necessary, respiratory rate was decreased) to achieve and maintain ETCO2 levels between 50 and 55 mmHg until spontaneous breathing returned.
The administration of propofol and remifentanil was discontinued once the target ETCO2 value was achieved. When patients spontaneously regained consciousness, they voluntarily opened their eyes, demonstrated a strong hand grip upon command, exhibited smooth respiration, displayed adequate chest wall movement, and the tracheal tube was removed. Subsequently, oxygen was administered via nasal cannula at a flow rate of 5 L/min.
Patients were withdrawn from the study in the event of serious surgical complications, such as excessive bleeding, sepsis, or anaphylactic shock.
Data collection
The collected demographic data included sex, age, height, weight, body temperature, preoperative hemoglobin levels, ASA classification, and underlying medical conditions. We also recorded the surgical time (from insertion of the ureteroscope or cystoscope into the urethra until its removal), anesthesia time (from the initiation of anesthesia induction to the discontinuation of propofol infusion), the total doses of propofol and remifentanil administered, and the use of vasoactive agents. MAP, HR, ETCO2, and BIS values were recorded at each time point (T1: the time of patient entry into the operating room; T2:10 min after the operation began; T3: the end of ETCO2 adjustment in the PACU; T4: the time of voluntary eye opening; T5: immediately after extubation; T6:20 min after extubation), as well as body temperature when entering the operating room and PACU. Arterial blood gas analysis was performed after the completion of ETCO2 adjustment and 20 min after extubation. We recorded the time to three events following the discontinuation of propofol infusion: tracheal extubation (primary outcome), spontaneous breathing onset, and voluntary eye-opening (secondary outcomes). Spontaneous breathing-onset time was defined as the time from the discontinuation of propofol infusion to the appearance of an autonomous respiratory waveform in the SIMV mode or a change in the ETCO2 waveform. The MCA blood flow velocity was measured at three time points: the time of the patient’s entry into the operating room, the end of ETCO2 adjustment in the PACU, and 20 min after extubation. Additionally, we documented any anesthesia-related complications that occurred in the PACU, such as postoperative nausea, vomiting, chills, agitation/delirium, and hypoxemia (SpO2 < 90%).
Statistical analysis
Based on a retrospective analysis conducted at our center, the average extubation time for urological surgery patients with normal blood carbon dioxide levels was 38.29 ± 17.87 minutes. We assumed that by increasing patient ETCO2 to 55 mmHg, we could reduce extubation time by 20%, resulting in an estimated mean extubation time of 30.63 minutes with a standard deviation of 14.29 minutes. Accordingly, the sample size was determined using G* Power software with a significance level set at 5% (two-sided), power set at 80%. After accounting for a potential dropout rate of 10%, 82 patients were assigned to each group.
Patients were blinded to group assignment. Researchers 1 and 2 were not involved in the decision-making process. Researcher 1 adjusted ETCO2 levels to achieve the target value according to group assignment based on a random number. Researcher 2 collected TCD data and recorded indicators such as breath-holding, eye opening, and extubation, without knowledge of group allocation. Data analysis was performed by dedicated analysts who were blinded to the group assignments.
This study employed an intention-to-treat analysis approach to ensure that all randomized participants were included in the analysis according to their original group assignments, regardless of protocol deviation or loss to follow-up.
The Statistical Package for Social Sciences (SPSS Statistics for Windows, version 26.0) was used for statistical analyses. We described continuous variables using mean ± standard deviation (SD) or median (Q1, Q3). Categorical data, such as sex, are presented as numbers (percentages). The Kolmogorov–Smirnov test was used to assess the normality of continuous variables. To compare normally distributed continuous variables between groups, Student's t-test was used, and the outcomes were adjusted using covariance analysis when necessary. Conversely, when comparing differences in non-normally distributed continuous variables between groups, the Mann–Whitney U test was used. The chi-square test or Fisher's exact probability method was used to compare categorical data between groups. Repeated-measures analysis of variance was used to assess between-group differences in multiple continuous variables. Sidak’s method was used to correct multiple comparisons. In addition, we employed multiple linear regression analysis to explore the effects of the perioperative variables on the primary outcome. A negative β value was taken to indicate a negative predictive effect on extubation time. The threshold for statistical significance was set at P < 0.05.
Results
The recruitment process for the trial participants is summarized in Fig. 1. Participants were recruited from September to December 2022, resulting in 164 patients being screened and randomly assigned to the normocapnia and mild hypercapnia groups, with 82 patients in each group completing the study. After completion of ETCO2 adjustment in the PACU, five patients in the normocapnia group and one patient in the mild hypercapnia group received additional low-dose propofol sedation due to coughing.
The baseline demographic characteristics and perioperative randomized events in both groups are presented in Table 1. The two groups did not differ significantly in terms of surgical time, anesthesia time, or total infusion doses of propofol and remifentanil. Furthermore, no anesthesia-related complications were observed in either group.
We collected HR, MAP, and BIS values at different time points. The trends in the changes in HR, MAP, and BIS values in the mild hypercapnia group were similar to those in the normocapnia group (data not shown). However, at T3, T4, and T5, HR was significantly higher in the mild hypercapnia than in the normocapnia group. The MAP and BIS values did not differ significantly between the two groups at any time point.
Arterial blood gas analysis (Table 2) demonstrated that, at T3, the median PaCO2 was significantly higher in the mild hypercapnia than in the normocapnia group (P < 0.001). However, at T6, PaCO2 values no longer differed significantly between the groups (P = 0.2).
The results of the between-group comparisons of the primary and secondary outcome measures are presented in Table 3. The mild hypercapnia group exhibited a significantly shorter extubation time (13.9 ± 5.9 min) than that of the normocapnia group (16.3 ± 7.6 min, P = 0.024). While the reduction of 2.42 minutes did not meet the predefined threshold for clinical significance, it represented a 14.85% reduction relative to the normocapnia group's extubation time of 16.3 minutes. Additionally, a similar reduction was observed in spontaneous breathing-onset time (P = 0.021) and voluntary eye-opening time (P = 0.008). Considering the variation in age distribution, we used covariance analysis and treated age as a covariate to account for outcome measures; in these analyses, both primary and secondary outcomes continued to exhibit statistically significant differences.
The selection of variables for multivariate linear regression analysis was primarily based on factors that had been confirmed in previous studies, variables that were found to be significant in univariate linear regression analysis, and variables considered to have a significant influence in clinical practice. According to the results (Table 4), adjusted ETCO2 levels and absence of a history of hypertension had a negative predictive effect on extubation time (P = 0.004 and 0.009, respectively). Other perioperative variables did not significantly affect the extubation time.
Variations in MCA blood flow velocity monitored by TCD at different time points are shown in Fig. 2. At T3, the mild hypercapnia group exhibited a significantly higher average MCA blood flow velocity than did the normocapnia group (P < 0.001). However, the difference between the groups had disappeared by 20 minutes after extubation (T6) (P = 0.114).
Discussion
ETCO2 is often regarded as a reliable substitute for invasive monitoring or repeated arterial blood gas analyses to measure PaCO2 in healthy adults [15–17]. Our results demonstrated that, even after adjusting for age, statistically significant differences persisted in both primary and secondary outcomes. Moreover, multiple linear regression analysis revealed a significant negative association between the adjusted ETCO2 levels and extubation time. We speculate that the following mechanisms may underlie these findings:
First, the cerebral vasculature is highly sensitive to changes in PaCO2 [18], with elevated PaCO2 causing cerebral vasodilation, reducing cerebrovascular resistance and ultimately causing increased CBF [19,20]. Previous studies have shown that measuring the mean blood flow velocity in the MCA using TCD can be used for indirect assessment of changes in CBF [21,22]. In our study, the mean MCA blood flow velocity at the end of ETCO2 adjustment was significantly higher in the mild hypercapnia than in the normocapnia group (P < 0.001), suggesting that the mild hypercapnia group may have higher CBF.
Second, hypercapnia can activate the sympathetic nervous system, leading to increased HR and CO [2,18,23]. In this study, after adjusting for ETCO2 levels, HR in the mild hypercapnia group was significantly higher than that in the normocapnia group. This finding suggests a greater likelihood of increased CO in the mild hypercapnia than in the normocapnia group. The combined effects of increased CO and CBF may expedite elimination of intravenous anesthetics from the brain.
Previous studies have suggested that mild hypercapnia not only shortens the emergence time after inhalational anesthesia, but also inhibits laryngeal reflexes and improves cardiovascular stability [24,25]. In our study, fewer patients in the mild hypercapnia group required supplementary propofol sedation in the PACU. Therefore, we speculate that mild hypercapnia may reduce the risk of early coughing caused by stimulation from the tracheal tube and thus decrease the requirement for sedative medications, allowing more patients to regain consciousness rapidly and smoothly.
While we speculate that mild hypercapnia may shorten the emergence time from TIVA through the aforementioned pathway, further research is required to confirm these hypotheses.
From a clinical perspective, even modest reductions in the emergence time can be meaningful, particularly in settings where rapid recovery is critical, such as in high-risk patient populations or busy operating rooms. Although the observed difference in extubation time did not meet the predefined threshold for clinical significance, the results support our hypothesis that ETCO2 levels during the recovery period affect the emergence time after TIVA and establish a research foundation for future larger-scale prospective studies. No cardiovascular or cerebrovascular complications related to hypercapnia occurred during the perioperative period, suggesting that transient mild hypercapnia during the recovery period may be a safe and feasible approach to shorten the emergence time from TIVA [26].
We also found that a history of hypertension may be an adverse factor affecting extubation time. Previous study has indicated delayed awakening in hypertensive patients [27], although the specific mechanisms remain unclear. Further investigations should involve a more detailed comparison and analysis based on the classification of hypertension grades, medication categories, and blood pressure control status.
In our study, we opted for extubation in the PACU without using antagonistic drugs or other awakening techniques (such as calling the patient's name or tapping the patient's shoulder, etc.). This management approach is a routine workflow at our institution that aims to provide patients with a more comfortable recovery experience. Prior research has reported that implementation of this “non-contact” extubation strategy can lead to an enhanced awakening quality, attenuated hemodynamic response, and a reduced occurrence of airway-related complications [28–30].
This study had some limitations. First, although a statistically significant difference in extubation time was found, the reduction did not meet the predefined threshold for clinical significance. However, the observed 2.42-minute reduction, which represents a 14.85% decrease in extubation time, may be clinically relevant, particularly in scenarios where even small improvements in recovery time are valuable. Importantly, the minimal clinically important difference was not explicitly set in the sample size calculation, limiting the ability to assess the clinical significance of the observed differences fully. Further research is needed to explore the effects of different degrees of hypercapnia on the recovery process after TIVA, as well as to identify specific patient populations or clinical settings where this intervention may provide the most benefit. Secondly, the use of blood samples to determine the blood concentration of propofol and its metabolites in different regions may provide a more direct reflection of the differences in the propofol clearance rate [31]. Third, neuromuscular monitoring or reversal of muscle relaxation was not conducted before extubation. Objective neuromuscular monitoring measures should be included in future studies to enhance patient safety and research reliability.
In conclusion, mild hypercapnia during the recovery period significantly reduced the extubation time in patients undergoing TIVA. Increasing ETCO2 levels has potential value for promoting rapid emergence from intravenous anesthesia. The mild elevation of carbon dioxide levels, along with the concurrent increase in cerebral arterial blood flow velocity during the recovery period, can be reversed, and has no significant adverse effects.
PDF Links
PubReader
ePub Link
XML Download
Full text via DOI
Download Citation
Print
