A Pilot and Feasibility Study of a Four‐Stage Neurological Staging Framework in Pediatric Tracheal Extubation During Anesthesia Emergence Authors: Megumi Okuyama, Katsuhiko Ishibashi, Shiori Nakao, Tomoharu Ukiya, Kohei Takahashi, Taiki Kojima Published: 2026 ABSTRACT ABSTRACTBackgroundPostoperative tracheal extubation is a particularly high‐risk period in pediatric anesthesia. Because emergence from anesthesia is a complex and complicated process, determining the optimal timing for safe tracheal extubation remains challenging.AimsTo evaluate the feasibility and descriptive characteristics of a four‐stage neuroanatomical framework for pediatric anesthesia emergence while providing exploratory incidence estimates of respiratory adverse events across these stages.MethodsChildren younger than 7 years undergoing general anesthesia were enrolled. Emergence stratified into four stages, defined by cranial nerve indicators: Stage 1 (Medulla: regular breathing, movement, excluding cough or swallowing); Stage 2 (Pons: tearing or grimacing); Stage 3 (Midbrain: conjugate gaze or eye‐opening); and Stage 4 (Cerebrum: purposeful movement). Primary outcomes included feasibility metrics—such as protocol completion, recruitment flow, and data completeness—and successful assignment of emergence stages. Exploratory descriptive statistics summarized the incidence of respiratory adverse events. Risk ratios were calculated to compare the occurrence of respiratory adverse events in intermediate (Stages 2 and 3) and cortical (Stage 4) stages.ResultsOf the 174 patients screened, 89 completed the protocol. Overall, 153 patients provided consent (91.6% consent rate). The overall data capture rate was 58.2% (89/153), while the worksheet processing rate—adjusted for protocol‐ineligible cases, such as those with supraglottic airway use—was 73.2% (112/153). Among the 89 patients analyzed, there were no missing data for key variables, and stage assignment was achieved in all cases. Respiratory adverse events were observed in 16.9% of children; descriptive analysis showed an event rate of 25.0% during Stage 2, 25.0% during Stage 3, and 11.6% during Stage 4.ConclusionsWhile stage assignment was possible among captured cases, the predefined data capture targets were not met. This pilot study successfully identified key workflow and feasibility barriers, which establish a foundation for refining protocols in the upcoming multicenter ExPEDIA registry.Trial RegistrationClinical Trial Registration: jRCT 1030250099 Introduction Postoperative tracheal extubation remains a high‐risk period in pediatric anesthesia [1, 2]. Despite its clinical importance, robust evidence‐based criteria for determining the optimal timing of tracheal extubation remain limited, and clinical practice often relies on the individual experience and judgment of clinicians. Emergence from anesthesia is a complex process that follows a neuroanatomical sequence, progressing from the medulla to the pons, midbrain, and, eventually, the cerebral cortex [3, 4, 5]. Brainstem structures mediate primitive airway reflexes, while higher cortical centers modulate and inhibit exaggerated responses [6, 7]. This layered recovery process has made it particularly challenging to establish reliable criteria based solely on select physiological indicators or monitoring metrics. This hierarchical recovery, mapped to specific clinical signs in our neurological four‐stage classification (Figure 1), suggests that during intermediate stages—when brainstem reflexes have returned, but cortical control remains incomplete—patients may pass through a “vulnerable window” that is prone to respiratory adverse events (RAEs), such as laryngospasm. Therefore, we recognize the need to explore the underlying physiological mechanisms and conceptual frameworks that define a “safe state” for extubation. We propose that this physiological interpretive model is applicable to all pediatric patients, regardless of the clinical setting, providing a foundational basis for extubation strategies and offering significant educational value for clinicians. Neuroanatomical framework and the four‐stage emergence classification. The schema summarizes the brainstem regions (medulla, pons, and midbrain) and the cerebrum, along with their corresponding clinical signs. (Upper) Emergence is categorized into four stages based on the sequential recovery of cranial nerve activity, progressing from the medulla toward the cerebral cortex: Stage 1 (Medulla): Regular spontaneous breathing, movement excluding cough, or swallowing. Stage 2 (Pons): Tearing or facial grimacing. Stage 3 (Midbrain): Conjugate gaze or eye‐opening. Stage 4 (Cerebrum): Purposeful movement. (Lower) Relationship between airway protective reflexes and the central nervous system (CNS) control. Stages 2 and 3 represent a hypothesized “vulnerable window” in which airway reflexes (e.g., laryngeal closure and coughing) are partially recovered and often exaggerated, while cortical CNS control remains limited. Based on this conceptual model, this period is hypothesized to pose a higher potential risk for tracheal extubation than the cortical stage (Stage 4), a trend that requires further definitive validation in larger registry studies. A previous single‐center observational study reported that the presence of specific clinical signs—such as facial grimacing, eye opening, purposeful movement, and conjugate gaze—is associated with successful awake tracheal extubation in children emerging from volatile anesthesia [3]. However, these findings are context‐specific and their generalizability to diverse settings, such as deep extubation or total intravenous anesthesia (TIVA), remains uncertain. To develop a more objective and anatomically grounded assessment, we integrated these clinical signs into a hierarchical neurological four‐stage classification based on the neurological framework proposed by Reshef et al. [4] This classification system maps specific cranial nerve signs to the neuroanatomical sequence of emergence, progressing from the medulla (Stage 1) to the cerebral cortex (Stage 4). This study was conducted as a preliminary pilot investigation for evaluating the procedural feasibility of this staging framework. Our primary objectives were to: (1) assess the feasibility of the protocol based on the predefined success criteria, including data completeness, resource demand, and operational challenges; (2) identify potential areas for optimization to enhance recruitment efficiency and data reliability for the subsequent multicenter Extubation in Pediatric Anesthesia (ExPEDIA) registry; and (3) provide foundational descriptive incidence data of RAEs across the different emergence stages. We hypothesized that the specific combination and neuroanatomical sequence of these signs—rather than a simple numerical count—would more accurately reflect airway stability and identify the transition from “vulnerable” to “stable” emergence. Ultimately, this pilot study serves as a critical step toward the goal of the ExPEDIA study, seeking to establish a potential universal framework that enhances safety for all pediatric patients and supports clinicians dedicated to pediatric anesthesia care. Materials and Methods Study Design and Methods This prospective, single‐center, observational cohort study was conducted at Chiba University Hospital from July 2024 to April 2025. After obtaining institutional review board approval and written informed consent, we enrolled children younger than 7 years undergoing general anesthesia with tracheal extubation. We limited our study population to patients under 7 years of age, consistent with the previous study by Templeton et al. [3]. As a pilot study, we focused on this specific age group to limit the sample and maintain cohort homogeneity. Patients with tracheostomies, those requiring postoperative mechanical ventilation, or those whose guardians refused to provide informed consent were excluded. Clinical Variables and Data Collection Data on patient characteristics, surgical parameters, anesthesia and airway management, signs of emergence from anesthesia, and RAEs were gathered by attending anesthesiologists through standardized research reporting forms. Anesthesiologists performed tracheal extubation using their preferred strategy and recorded clinical signs indicative of emergence. Clinical recovery signs observed before extubation were also documented. These signs—predetermined based on previous studies [3, 7]—included spontaneous regular breathing, tidal volume of at least 5 mL/kg, movement other than coughing, swallowing, tearing, facial grimacing, gaze, eye opening, and purposeful movement. Additionally, clinical variables were identified based on the literature [1, 2, 7] and consensus among the research team: (1) airway hypersensitivity, defined as active or recent (within 14 days) upper respiratory infection symptoms, asthma, or exposure to household tobacco smoke; (2) difficult airway features, including predictors of a difficult airway, difficult mask ventilation during anesthesia induction, or more than three attempts at tracheal intubation; and (3) birth abnormalities, specifically preterm birth or low birth weight. Additional recorded variables included age, premedication, operating time, anesthesia maintenance, end‐tidal CO2, and end‐tidal sevoflurane concentration immediately before tracheal extubation. Definition of Emergence Signs and Classification of Neurological Staging Framework To evaluate emergence, the attending anesthesiologists documented whether eight predefined clinical signs appeared before extubation: Regular spontaneous breathing: Stable respirations without assistanceMovement excluding cough: Visible body motion (limbs, head, or face), excluding coughing and other reflexes, regardless of intentSwallowing: Laryngeal movement, deglutition, or pharyngeal gag reflex.Tearing: Tear secretion, pooling at the medial canthus, or active overflow of tearsFacial grimacing: Facial muscle contraction (e.g., brow furrowing or frowning) in response to internal or external stimuli.Conjugate gaze: Conjugate or dysconjugate eye deviation from midline.Eye‐opening: Eyelid separation occurring independently, without external stimuli or verbal commands.Purposeful movement: Intentional movements like reaching, shaking the head, or attempting to sit up. Regular spontaneous breathing: Stable respirations without assistance Regular spontaneous breathing: Stable respirations without assistance Movement excluding cough: Visible body motion (limbs, head, or face), excluding coughing and other reflexes, regardless of intent Movement excluding cough: Visible body motion (limbs, head, or face), excluding coughing and other reflexes, regardless of intent Swallowing: Laryngeal movement, deglutition, or pharyngeal gag reflex. Swallowing: Laryngeal movement, deglutition, or pharyngeal gag reflex. Tearing: Tear secretion, pooling at the medial canthus, or active overflow of tears Tearing: Tear secretion, pooling at the medial canthus, or active overflow of tears Facial grimacing: Facial muscle contraction (e.g., brow furrowing or frowning) in response to internal or external stimuli. Facial grimacing: Facial muscle contraction (e.g., brow furrowing or frowning) in response to internal or external stimuli. Conjugate gaze: Conjugate or dysconjugate eye deviation from midline. Conjugate gaze: Conjugate or dysconjugate eye deviation from midline. Eye‐opening: Eyelid separation occurring independently, without external stimuli or verbal commands. Eye‐opening: Eyelid separation occurring independently, without external stimuli or verbal commands. Purposeful movement: Intentional movements like reaching, shaking the head, or attempting to sit up. Purposeful movement: Intentional movements like reaching, shaking the head, or attempting to sit up. Emergence was categorized into four neuroanatomical stages based on observed signs (Figure 1): Stage 1 (Medulla): Regular spontaneous breathing, movement (excluding cough), or swallowing.Stage 2 (Pons): Tearing or facial grimacing.Stage 3 (Midbrain): Conjugate gaze or eye‐opening.Stage 4 (Cerebrum): Purposeful movement. Stage 1 (Medulla): Regular spontaneous breathing, movement (excluding cough), or swallowing. Stage 1 (Medulla): Regular spontaneous breathing, movement (excluding cough), or swallowing. Stage 2 (Pons): Tearing or facial grimacing. Stage 2 (Pons): Tearing or facial grimacing. Stage 3 (Midbrain): Conjugate gaze or eye‐opening. Stage 3 (Midbrain): Conjugate gaze or eye‐opening. Stage 4 (Cerebrum): Purposeful movement. Stage 4 (Cerebrum): Purposeful movement. If multiple signs were observed, the patient was assigned to the highest stage reached. Feasibility Outcomes and Success Criteria The primary outcome of this study was to assess the feasibility of the outlined protocol. In accordance with the CONSORT extension for pilot and feasibility trials, we identified the following feasibility outcomes: (1) rates of recruitment, approach, and consent; (2) data capture and completion rate; (3) missingness of key variables; (4) capacity to assign stages; and (5) protocol burden. To evaluate the feasibility of this approach, we established success criteria: (1) recruitment, approach, and consent rate ≥ 90%; (2) data capture and completion rate ≥ 95%; (3) missingness of key variables ≤ 5%. Reasons for non‐participation, such as ineligibility or missed eligible cases due to clinical workload, were documented. The progression of participants from screening through final analysis was monitored to guide recruitment strategies for the future multicenter ExPEDIA registry. The Outcomes of Interest The outcomes of interest included the incidence of any RAEs from the moment of tracheal extubation until the patient's discharge from the operating room. To maintain consistency with established pediatric literature [1, 2], the composite outcome of interest for RAEs was defined as the occurrence of one or more of the following: sustained desaturation (SpO2 < 93% for ≥ 30 s), persistent cough (lasting ≥ 10 s), upper airway obstruction (stridor or paradoxical chest movement requiring airway maneuvers), laryngospasm (complete glottic closure requiring positive pressure ventilation or administration of muscle relaxants), and bronchospasm (expiratory wheezing accompanied by increased airway pressure). “Any desaturation” (SpO2 < 95%) was evaluated as a separate variable. Statistical Analysis The study aimed to recruit 300 patients in 12 months but was terminated early in April 2025 due to a smaller‐than‐expected population and sufficient interim data to refine the protocol for the upcoming multicenter registry. For summary statistics, continuous variables are reported as means and standard deviations (SDs), or medians and interquartile ranges (IQRs), and categorical variables as counts and percentages. Given the pilot nature of this study and the limited number of observed RAEs, our analysis focused on descriptive and exploratory measures to avoid over‐interpretation. For each of the four emergence stages, we calculated the number of patients, the incidence of RAEs (risk %), and their corresponding 95% confidence intervals (CIs). The risk and 95% CIs were calculated using the exact (Clopper–Pearson) method. To investigate our physiological hypothesis regarding the “vulnerable window,” we calculated relative risks (RRs) with 95% CIs to assess the association between individual clinical signs and RAEs. Due to the exploratory nature of these analyses and the potential for type I error inflation due to multiple comparisons, p‐values were not emphasized; instead, all estimates were interpreted as foundational data for future study planning. Relative risks and relative differences (RDs) with 95% CIs were computed using the Wald method. For clinical signs with zero events in one group (e.g., tearing), the Haldane–Anscombe correction was applied by adding 0.5 to each cell count, enabling calculation of RRs and 95% CIs. All statistical analyses were conducted with SAS version 9.4 (SAS Institute Inc., Cary, NC, USA). Results Participant Flow and Feasibility Outcomes A participant flow diagram is summarized in Figure 2. Among 174 patients screened for eligibility, 167 satisfied the inclusion criteria. Of these, 13 eligible individuals were not approached for consent due to clinical workload and administrative issues, including delays in communication with the research team during peak hours, while one patient declined participation. Thus, the guardians of 153 patients consented to the study (approach/consent rate: 91.6% [153/167]). Data worksheets were completed for 89 of the 153 consenting participants, yielding a data capture rate of 58.2% (89/153) within routine clinical operations. Participant flow diagram. The recruitment and exclusion processes for study participants are illustrated. Data capture failures (n = 64) arose primarily from worksheet omission during periods of high clinical activity by attending anesthesiologists (n = 42), selection of a supraglottic airway device (SGA) during anesthesia planning (n = 21), and alteration of postoperative airway management requiring sedation and ventilation by the surgical team (n = 1). Excluding these 22 protocol‐ineligible cases (SGA and postoperative ventilation), the data capture rate for eligible cases managed with endotracheal intubation was 67.9% (89/131). Considering pre‐operative exclusions as successful protocol tracking, the worksheet completion rate reached 73.2% (112/153). Ultimately, 89 patients were included in the final analysis. All 89 patients were successfully classified into emergency stages, with no missing data for the clinical parameters documented in the analyzed worksheets (Table S1). Incidence of Any Respiratory Adverse Events Patient characteristics, backgrounds, and descriptive statistics for relevant confounding factors are summarized in Table 1. Respiratory adverse events occurred in 15/89 children (16.9%; 95% CI, 10.5–26.0), predominantly comprising upper airway obstruction (n = 10), sustained desaturation (n = 4), and persistent cough (n = 3). No instances of laryngospasm or bronchospasm were identified. Episodes of desaturation (SpO2 < 95%) occurred in 21 children (23.6%; 95% CI: 16.0–33.4). TABLE 1Patient characteristics and perioperative factors.CharacteristicTotalAny RAENon RAE (−)Patients (n)891574Patient backgroundAge (months, median [IQR])26.0 (12.0–46.0)25.0 (14.0–60.0)26.5 (12.0–42.0)Sex (male), n (%)48 (53.9)7 (46.7)41 (55.4)Weight (kg, median [IQR])11.0 (8.1–15.1)11.4 (9.2–17.4)10.3 (7.8–15.0)Birth abnormalities a , n (%)19 (21.3)3 (20.0)16 (21.6)Airway and preoperative riskAirway sensitivities b , n (%)28 (31.5)8 (53.3)20 (27.0)DAM features c , n (%)8 (9.0)2 (13.3)6 (8.1)Premedication, n (%)42 (47.2)9 (60.0)33 (44.6)Surgical and anesthetic dataOperation time (min, median [IQR])53.0 (35.0–99.0)138.0 (24.0–243.0)51.5 (35.0–92.0)Anesthesia maintenance (Sevo), n (%)57 (64.0)8 (53.3)49 (66.2)Et‐Sevoflurane d before TE (%, median [IQR])0.1 (0.0–0.2)0.0 (0.0–0.2)0.1 (0.0–0.2)EtCO2 before TE (mmHg, median [IQR])51.0 (46.0–56.0)51.0 (45.0–62.0)51.5 (47.0–55.0) Note: Age is presented as median (IQR). The actual age range was 0–81 months.Abbreviations: EtCO2: End‐tidal carbon dioxide tension measured immediately before tracheal extubation; IQR, interquartile range; RAE, respiratory adverse events; TE, tracheal extubation. a Birth Abnormalities: Includes preterm birth (< 37 weeks of gestation) or low birth weight (< 2500 g). b Airway Sensitivities: Defined as the presence of active or recent (within 14 days) upper respiratory infection symptoms, asthma, or exposure to environmental tobacco smoke. c DAM features: Difficult Airway Management features, including clinical predictors of a difficult airway, difficult mask ventilation, or more than three intubation attempts. d Et‐Sevoflurane: End‐tidal sevoflurane concentration measured immediately before tracheal extubation. Patient characteristics and perioperative factors. Note: Age is presented as median (IQR). The actual age range was 0–81 months. Abbreviations: EtCO2: End‐tidal carbon dioxide tension measured immediately before tracheal extubation; IQR, interquartile range; RAE, respiratory adverse events; TE, tracheal extubation. Birth Abnormalities: Includes preterm birth (< 37 weeks of gestation) or low birth weight (< 2500 g). Airway Sensitivities: Defined as the presence of active or recent (within 14 days) upper respiratory infection symptoms, asthma, or exposure to environmental tobacco smoke. DAM features: Difficult Airway Management features, including clinical predictors of a difficult airway, difficult mask ventilation, or more than three intubation attempts. Et‐Sevoflurane: End‐tidal sevoflurane concentration measured immediately before tracheal extubation. Risk Estimates Across Emergence Stages The distribution of RAEs across the four neurological stages is summarized in Table 2. In this pilot study, descriptive analysis showed that the incidence of RAEs was 25.0% (95% CI: 7.3–52.4) during Stage 2, 25.0% (95% CI: 5.5–57.2) during Stage 3, and 11.6% (95% CI: 3.9–25.1) during Stage 4. However, due to small subgroup sizes, limited event counts, wide overlapping confidence intervals, and a lack of adjustment for potential confounders, these descriptive trends do not infer true stage‐specific risk differences. TABLE 2Risk ratios for respiratory adverse events across emergence stages.Emergence stagePatients (N)Incidence of RAEs % (n/N)95% CIRR95% CI for RRStage 1 (medulla)1816.7% (3/18)3.6–41.41.430.38–5.37Stage 2 (pons)1625.0% (4/16)7.3–52.42.150.66–7.02Stage 3 (midbrain)1225.0% (3/12)5.5–57.22.150.60–7.74Stage 4 (cerebrum)4311.6% (5/43)3.9–25.11.00 (Ref.)— Note: Incidence of RAEs was defined as the proportion of patients who experienced at least one RAE within each stage. The RR and its 95% Cis were calculated using the exact (Clopper–Pearson) method. Relative risks were calculated using Stage 4 (Cerebrum) as the reference group (RR = 1.00) because it represents the state of highest neurological recovery.Abbreviations: CI, confidence interval; RAEs, respiratory adverse events; RR, relative risk. Risk ratios for respiratory adverse events across emergence stages. Note: Incidence of RAEs was defined as the proportion of patients who experienced at least one RAE within each stage. The RR and its 95% Cis were calculated using the exact (Clopper–Pearson) method. Relative risks were calculated using Stage 4 (Cerebrum) as the reference group (RR = 1.00) because it represents the state of highest neurological recovery. Abbreviations: CI, confidence interval; RAEs, respiratory adverse events; RR, relative risk. The incidence of RAEs in relation to individual clinical signs is descriptively summarized in Table 3. Grimacing (RR: 1.62; 95% CI: 0.56–4.68) and swallowing (RR: 1.17; 95% CI: 0.46–2.96) demonstrated numerically higher point estimates, while tearing (RD: −20.3%; 95% CI: −29.4 to −11.1) and purposeful movement (RR: 0.53; 95% CI: 0.20–1.44) had lower estimates. Excluding tearing, most 95% CIs encompassed the null value. TABLE 3Incidence of respiratory adverse events based on the presence or absence of clinical signs.StageClinical signsIncidence of RAEs (sign +, n/N)Incidence of RAEs (sign −, n/N)RR (95% CI)RD (95% CI)1Movement16.4% (11/67)18.2% (4/22)0.90 (0.32–2.55)−1.8% (−20.2–16.6)Swallowing18.4% (7/38)15.7% (8/51)1.17 (0.46–2.96)2.7% (−13.1–18.6)2Tearing0.0% (0/15)20.3% (15/74)0.00 (0.00–0.85) a −20.3% (−29.4 to −11.1)Grimacing19.6% (11/56)12.1% (4/33)1.62 (0.56–4.68)7.5% (−7.7–22.8)3Gaze12.5% (1/8)17.3% (14/81)0.72 (0.11–4.81)−4.8% (−29.1–19.6)Eye opening13.8% (4/29)18.3% (11/60)0.75 (0.26–2.16)−4.5% (−20.5–11.4)4Purposeful movement11.6% (5/43)21.7% (10/46)0.53 (0.20–1.44)−10.1% (−25.4–5.2) Note: Relative risk and relative difference with 95% CIs were calculated using the Wald method.Abbreviations: CI, confidence interval; RAEs, respiratory adverse events; RD, relative difference; RR, relative risk. a The RR for “tearing” was calculated using the Haldane–Anscombe correction due to zero events in the Sign+ group. Incidence of respiratory adverse events based on the presence or absence of clinical signs. Note: Relative risk and relative difference with 95% CIs were calculated using the Wald method. Abbreviations: CI, confidence interval; RAEs, respiratory adverse events; RD, relative difference; RR, relative risk. The RR for “tearing” was calculated using the Haldane–Anscombe correction due to zero events in the Sign+ group. The distribution of observed neurological signs across the four emergence stages is summarized in Table 4. Swallowing (n = 8, 44.4%) and non‐coughing movement (n = 3, 16.7%) were predominant indicators in Stage 1. Movement and grimacing increased in frequency from Stage 2 onward, with eye opening (n = 12, 100%) prevalent in Stage 3. Although gaze was theoretically categorized as the highest clinical sign in Stage 3, it was observed exclusively among patients classified as Stage 4 (n = 8, 19.0%) and coincided with purposeful movement (n = 43, 100.0%). According to the study protocol, when multiple signs were present, patients were assigned to the highest neuroanatomical stage reached, resulting in these cases being assigned to Stage 4. TABLE 4Distribution and frequency of neurological signs across the four arousal stages.Stage 1 (n = 18), n (%)Stage 2 (n = 16), n (%)Stage 3 (n = 12), n (%)Stage 4 (n = 43), n (%)Movement3 (16.7)12 (75.0)11 (91.7)41 (95.3)Swallowing8 (44.4)6 (37.5)3 (25.0)21 (48.9)Tearing0 (0)3 (18.8)2 (16.7)10 (23.3)Grimacing0 (0)15 (93.8)8 (66.7)33 (76.7)Gaze0 (0)0 (0)0 (0)8 (18.6)Eye opening0 (0)0 (0)12 (100.0)17 (39.5)Purposeful0 (0)0 (0)0 (0)43 (100.0) Distribution and frequency of neurological signs across the four arousal stages. Discussion In this pilot and feasibility study, we assessed a four‐stage neuroanatomical framework for pediatric anesthesia emergence to determine its procedural feasibility, while identifying critical clinical workflow barriers and necessary protocol modifications to maximize data capture rates and minimize missing values. The emergence‐stage classification was feasible and effectively implemented across all participants included in the final analysis, yielding 100% data completeness for essential variables on the analyzed worksheets. The overall data capture rate was 58.2% (89/153), reflecting real‐world challenges in data collection. Notably, 21 patients managed with a supraglottic airway rather than an endotracheal tube, along with one patient whose postoperative airway protocol was altered, were considered per‐protocol exclusions rather than data capture failures. Adjusting for these 22 cases, the worksheet processing rate increased to 73.2% (112/153). Despite this adjusted improvement, 42 of 154 consenting patients (27.3%) were excluded solely due to data‐collection failures, a higher‐than‐expected attrition rate concentrated in the initial study phase and during staff turnover. These omissions may not have occurred at random, raising a risk of selection bias, as cases that were more difficult or clinically complex around extubation may have been less likely to be captured. These missing data delineate the operational limits of the current protocol and inform data collection strategies for the upcoming multicenter registry. This study has certain limitations that warrant consideration. First, as a single‐center pilot study with a modest sample size, our data capture rate did not meet the predefined target, and the associated risk of selection bias represents a major barrier to real‐time manual data capture in this setting. Second, some items in our staging framework required direct, synchronous observation, making it challenging for clinicians to recall and record them accurately after conclusion of the case. Furthermore, as an investigator‐developed tool, the inter‐rater reliability and formal validation of this staging framework have not yet been evaluated, which introduces a potential risk of misclassification and interpretation bias. For instance, while conjugate gaze was theoretically categorized as a Stage 3 sign in our initial framework, it was observed exclusively in Stage 4 and was documented in only 9.0% of total cases. This discrepancy likely reflects a practical reporting bias rather than a true physiological absence. Consequently, it was challenging to determine whether the sign was genuinely missing or simply unverified. Third, the current analysis lacked multivariable adjustment for key confounders—such as difficult airway features, airway sensitivities, and anesthetic techniques—that might have influenced risk across stages. Finally, the composite respiratory adverse event (RAE) outcome encompasses a wide range of clinical severity and distinct pathophysiologies; hence, arousal stage influences may vary significantly by specific event type and severity. To address these challenges in the upcoming multicenter ExPEDIA registry, protocol adjustments have been implemented based on these feasibility findings. To minimize selection bias and improve generalizability, the data collection tool will prioritize essential signs requiring real‐time observation and will be supplemented by electronic anesthesia records and physiological data logs where possible. An “unconfirmed” (not assessed) category will be added to address the gaze‐reporting issue noted above. The registry will also leverage the data validation and site‐network infrastructure of the JPEDIA registry, which has achieved a data capture rate exceeding 95% [8], including specific research leaders responsible for data integrity, a REDCap‐based system that restricts submission of incomplete records, and a pre‐study run‐in period to build staff proficiency before enrollment. This multifaceted strategy is anticipated to comprehensively eliminate missing data and enhance overall data accuracy. Although no cases of laryngospasm were observed in this pilot cohort, a 2.0% incidence rate was adopted based on previous literature,1 yielding an estimated requirement of approximately 8500 cases for the ExPEDIA registry. In conclusion, while stage assignment was possible among captured cases, the predefined data capture targets were not met due to clinical workload constraints. The primary value of this pilot study was the identification of key feasibility and workflow barriers, which establish a foundation for refining protocols and data capture strategies in the upcoming multicenter ExPEDIA registry. Author Contributions K.I., S.N., T.U., K.T., T.K., and M.O. were involved in the study design, critical revision, and final approval of the manuscript. M.O. drafted the manuscript. S.N., K.I., K.T., and M.O. collected data, and S.N. and K.I. verified the data. K.T. and M.O. provided statistical expertise. Funding This work was supported by Grants‐in‐Aid for Scientific Research (Kakenhi) (25K12202, April 1, 2025). Disclosure This study is an observational study in children without intervention or harmful stress. Ethics Statement This study will be conducted in accordance with the Declaration of Helsinki. The prospective cohort study was approved by the Institutional Review Board at Chiba University Hospital (HK202406‐04, 1 July 2024). Consent We have obtained written informed consent from all patients and their families. Conflicts of Interest The authors declare no conflicts of interest. Supporting information Table S1: Feasibility outcomes and success criteria.