Capnography With Integrated Pulmonary Index for Preventing Hypoxemia During Pediatric Urologic Surgery Under Sedation: A Randomized Controlled Trial

Introduction

Pediatric urinary procedures have become more common, with procedural sedation and analgesia (PSA) often used to relieve anxiety and pain [1–3]. PSA involves the use of fewer anesthetics and the avoidance of neuromuscular blocking agents, and is generally considered safer than general anesthesia by tracheal intubation. However, pediatric patients have narrower airways and lower respiratory reserve capacity, placing them at higher risk of respiratory depression and airway obstruction during PSA procedures. The resulting intraoperative hypoxemia poses a significant perioperative safety threat [3,4]. Related risk factors include inadequate monitoring [3]. Traditional monitoring methods (eg, visual observation, non-invasive blood pressure measurement, electrocardiography, and pulse oximetry) often fail to promptly detect changes in blood oxygen saturation following hypoventilation or apnea episodes. Particularly under routine supplemental oxygen therapy, declines in pulse oximetry (SpO2) exhibit significant delays, rendering them unreliable for early warning of ventilatory dysfunction [5].

The European Society of Anaesthesiology [2] and the American Society of Anesthesiologists [6] recommend that, in addition to traditional monitoring, capnography should be used for continuous evaluation of ventilation during PSA to allow for early recognition of apnea. CO2 monitoring compensates for delays in oximetry by estimating arterial blood CO2 partial pressure in a non-invasive manner and by detecting hypoventilation and apnea before clinical signs or changes in blood oxygen saturation [7–9]. Evidence exists to support its value in adult PSA and pediatric PSA (eg, reducing the incidence of hypoxemia) [8]. In addition, it is commonly used to monitor patients undergoing general anesthesia with intubation, cardiopulmonary resuscitation [10], sleep apnea hypopnea syndrome [11], and postoperative nausea and vomiting [9]. However, the evidence remains more limited in pediatric surgery (especially urologic surgery), and evidence of its effectiveness in pediatric surgery, particularly in urological procedures involving this specific population, remains limited. Furthermore, the optimal application model – whether used alone or in combination with other indicators – has yet to be clearly defined.

Beyond end-tidal carbon dioxide monitoring, the IPI serves as a single visual score (range, 1–10) derived from 4 parameters – end-tidal carbon dioxide (EtCO2), respiratory rate (RR), pulse oximetry saturation (SpO2), and heart rate (HR) – through a proprietary algorithm. Theoretically, this approach offers earlier warning of respiratory abnormalities. Its algorithmic logic mimics clinical decision-making, aiming to identify states where individual parameters remain within normal ranges but combined multi-parameter patterns indicate early respiratory deterioration [12]. Existing studies have confirmed the application value of IPI monitoring in specific populations. For instance, during adult cataract surgery sedation, IPI effectively reflects post-sedation respiratory parameter changes [13], and in pediatric endoscopy sedation IPI has been shown to alert for all apnea events with higher sensitivity than pulse oximetry alone [14]. These studies indicate that IPI can detect potential respiratory risks more promptly. While suggesting IPI’s potential for early warning, high-quality prospective evidence remains lacking regarding its value in pediatric PSA, particularly its efficacy and specific mechanisms when combined with end-tidal carbon dioxide monitoring to prevent hypoxemia during pediatric urological surgery.

Therefore, this randomized controlled trial aimed to definitively evaluate whether adding end-tidal carbon dioxide monitoring and IPI monitoring to standard monitoring can effectively reduce the incidence of intraoperative hypoxemia in pediatric urological surgery patients undergoing PSA, and to explore its impact on clinical intervention strategies.

Material and Methods

TRIAL DESIGN:

This prospective, single-blinded, randomized controlled trial was conducted from June 2023 to March 2024 at the Seventh Medical Center of Chinese PLA General Hospital. The study was approved by the Ethics Committee of the Seventh Medical Center of Chinese PLA General Hospital (2023-71) and was registered with the Chinese Clinical Trial Registry (ChiCTR 2300073943).

PARTICIPANTS:

Pediatric patients referred for urinary procedures requiring PSA were selected. Since IPI is not accessible for patients younger than 1 year, children aged between 3 and 12 years were included, as they are the typical group undergoing PSA. Participants needed to meet the ASA-PS classification of I or II, with I denoting healthy conditions and II denoting mild systemic diseases [15]. Exclusion criteria were ASA-PS III to VI, general anesthesia with endotracheal intubation, emergency procedures, a history of anesthetic allergies, and presence of nasal congestion-related diseases (diagnostic criteria for adenoid hypertrophy: preoperative nasal endoscopy showing adenoids blocking ≥50% of the posterior nostrils, or the presence of clinical symptoms such as sleep snoring and open-mouth breathing, confirmed by imaging). Preliminary experiments revealed that patients with nasal obstruction had unstable CO2 monitoring and IPI values during stable spontaneous breathing. Consequently, individuals with nasal obstruction, such as adenoid hypertrophy, were excluded because it was deemed difficult to trace CO2 using only mouth sampling, even though the capnometer’s brochure claimed it could sample CO2 during either nasal or mouth breathing. On the day of surgery, the patients’ legal guardians provided written informed consent, then demographic data from eligible participants were collected. Patient flow through the trial is shown in Figure 1.

RANDOMIZATION AND BLINDING:

An independent observer used opaque envelopes to randomly assign patients to groups. The randomization scheme used permuted blocks of 4, 6, and 8 [16]. Patients were blinded to their group assignment. The statistical analyst did not participate in participant recruitment or data collection.

ANESTHETIC TECHNIQUES:

Surgical procedures proceeded routinely in both groups, adhering to standard practices. A standard protocol of preoperative fasting was administered [17]. Standard monitoring, including ECG, heart rate, pulse oxygen saturation, non-invasive blood pressure, bispectral index (BIS), RR, and visual assessment of chest wall movements, was performed for all patients. Children received 2 L/min of free-flowing oxygen through a nasal catheter starting 5 min before PSA. Sedation was achieved with intravenous propofol (2–3 mg/kg) and sufentanil (0.1–0.2 μg/kg) administered immediately before the procedure. A dorsal penile nerve block was performed by surgeons after anesthesia induction using 0.2 ml/kg solution of 0.5% ropivacaine and 1% lidocaine. BIS was maintained at between 40 and 60 using continuous propofol infusion (3–5 mg/kg/h). Vasopressors were administered if blood pressure declined by 20% from baseline. No antagonists were given postoperatively.

INTERVENTIONS:

The fundamental difference between the 2 groups of patients was in the methods used to monitor and assess hypoventilation. In the intervention group, capnography and IPI monitoring (Capnostream™ 20p Bedside Monitor with Apnea-Sat Alert Algorithm, Medtronic, USA) were performed perioperatively in addition to standard monitoring. This device simultaneously provides waveform and numerical monitoring of EtCO2 and IPI monitoring. EtCO2 was continuously measured as one of the core input parameters for generating the IPI score, rather than as an independent supplementary device. IPI values range from 1 to 10: 10 represents normal; 8–9 falls within the normal range; 7 is near normal and requires attention; 5–6 suggests attention is needed and intervention might be required; 3–4 indicates intervention is necessary; and 1–2 indicates immediate intervention is required [18]. Its core early warning advantage lies in its ability to identify conditions for which a single parameter remains within the “normal” range, yet a combination of multiple parameters indicates early ventilation insufficiency or deteriorating oxygenation trends. This enables earlier intervention alerts compared to traditional threshold-based alarms. Hypoventilation was identified by manifestations such as SpO2 ≤95%, RR ≤80% of baseline, reduced or absent chest wall movements, EtCO2 variation ≥5 mmHg, a flat CO2 line lasting ≥3–10 s (depending on baseline RR, recorded as an episode of apnea), or IPI ≤6 in the intervention group [19]. As soon as any of these indicators were abnormal, interventions such as jaw thrust were immediately performed. Assessment of the control group relied solely on selected indicators from standard monitoring to identify hypoventilation, specifically through visual assessment of chest wall movement and respiratory rate displayed on the monitor (confirmed by manual counting). Intervention was triggered by SpO2 decline (<95%) or observation of significant apnea/abnormal chest wall movement, an approach inherently delayed in detecting ventilation abnormalities. Interventions were consistent between the 2 groups and included jaw thrust, supplemental oxygen, positive-pressure ventilation, oropharyngeal/nasopharyngeal airway placement, surgical pause alerts, and endotracheal intubation.

OUTCOMES MEASURES:

The primary outcome was oxygen desaturation, defined as pulse oximetry <95% for >5 s intraoperatively, as mild hypoxemia has been reported to precede adverse events [20].

Secondary outcomes included severe desaturation (pulse oximetry <90% for >5 s), airway management, and perioperative complications. Airway management encompassed jaw thrust, positive-pressure ventilation, oropharyngeal or nasopharyngeal airway placement, and endotracheal intubation. Predefined criteria for perioperative complications included intraoperative hypotension (systolic blood pressure <90 mmHg and/or >20% decrease from baseline), intraoperative arrhythmia, postoperative hypoxia (oxygen saturation <90%, from the end of the procedure until 24 h after the procedure), and postoperative nausea and vomiting (PONV, assessed by the Simplified PONV Impact Scale, from the end of the procedure until 24 h after surgery) [21].

STATISTICAL ANALYSIS:

Regarding the primary endpoint (perioperative oxygen desaturation rate), we assumed an incidence of 43.8% in the control group and 18.8% in the intervention group, based on our preliminary experiment with 32 participants, which corresponds to a 25% absolute difference. With a two-tailed α of 0.05 and a β of 0.20, 102 patients were needed. Considering a possible dropout rate of 10%, the required number of study cases was set at 112 [22].

The primary null hypothesis, which posited that equal proportions of patients in the intervention and control groups would experience oxygen desaturation, was tested using the chi-squared test. For dichotomous variables (eg, occurrence or absence of hypoxemia, presence or absence of intraoperative arrhythmia), between-group comparisons were performed using the chi-square test, and the results were presented as frequencies (%). Continuous variables (eg, age, body weight, number of interventions) are expressed as mean±standard deviation. The Shapiro-Wilk test was used to test the normal distribution of the data, the t test was used to compare the differences between the 2 groups for normal distribution, and the Wilcoxon (Mann-Whitney) sample test was used to compare the groups for data that were not normally distributed. The significance level for all statistical tests was set at α=0.05 and P<0.05 was considered to indicate a statistically significant difference. Data analysis was done using IBM SPSS Statistics 25.0 software.

Results

PARTICIPANTS:

Between June 30, 2023, and March 1, 2024, 142 children scheduled for urinary surgery with PSA were enrolled. Two cases were excluded due to nasal obstruction, resulting in 140 patients randomized. Seven patients were removed after randomization (3 from the control group and 4 from the intervention group): 5 did not tolerate the cannula, and 2 underwent anesthesia alteration due to non-anesthetic factors. Additional participants were randomized until the sample size was met. Ultimately, 133 patients were included in the analysis: 67 in the intervention group and 66 in the control group. Age, sex, height, weight, ASA-PS classification, procedure type, sedation duration, procedure duration, propofol dosage, sufentanil dosage, and emergence time were similar between the 2 groups (Table 1).

PRIMARY AND SECONDARY OUTCOMES:

Oxygen desaturation was lower in the intervention group (34.33% vs 56.06%, P=0.015, OR=0.410, 95% CI: 0.203 to 0.825). Regarding secondary outcomes, patients in the intervention group experienced less severe oxygen desaturation than those in the control group (19.40% vs 36.40%, P=0.034, OR=0.421, 95% CI: 0.192 to 0.925) (Table 2).

Compared to the control group, patients in the intervention group were more likely to require jaw thrust (1.84±1.31 vs 1.17±1.12 times, 95% CI: −1.087 to −0.251], P=0.002) and less likely to need positive-pressure ventilation (1.64±1.31 vs 2.39±1.48 times, 95% CI: 0.273 to 1.231], P=0.002). No differences were observed in the need for oropharyngeal or nasopharyngeal airway placement or endotracheal intubation (Table 2). This finding directly supports the primary hypothesis of this study that combined monitoring can effectively reduce the risk of hypoxemia.

PERIOPERATIVE COMPLICATIONS:

The incidence of intraoperative hypotension (19.70% vs 17.90%, P=0.827, OR=0.890, 95% CI [0.372, 2.124]), intraoperative arrhythmia, postoperative hypoxia, and PONV score did not differ between the groups (Table 3). These negative results indicate that combined monitoring improves respiratory outcomes without increasing the risk of other perioperative complications.

Discussion

The central finding of this study is that the addition of CO2 monitoring + IPI monitoring to standard monitoring combined with supplemental oxygen significantly reduced the incidence of intraoperative hypoxemia (34.33% vs 56.06%) and severe hypoxemia (19.40% vs 36.40%, P=0.034) in patients undergoing urinary surgery with PSA. By prompting earlier implementation of jaw thrust (1.84±1.31 vs 1.17±1.12), the need for subsequent positive-pressure ventilation (1.64±1.31 vs 2.39±1.48) was reduced. This result provides an important basis for filling the gap in the use of relevant monitoring techniques in pediatric surgery.

The CapnostreamTM 20p uses infrared spectroscopy to measure CO2 levels and RR [23]. This technology can be applied to non-intubated patients and has been shown to significantly enhance the safety of intraoperative and postoperative management in non-intubated adult surgeries [24]. Additionally, the IPI, based on capnography, offers a convenient and intuitive 0 to 10 scale, simplifying anesthesia management.

The positive results (reduced incidence of hypoxemia) observed in this study in elective pediatric urologic surgery compared with existing studies may be related to the specific study population, type of surgery, and intervention process. The study population consisted of children undergoing urological surgeries with PSA, chosen due to their lower tolerance to hypoxia [25] and the high prevalence of urological procedures among pediatric PSA cases [26].

The value of monitoring is better demonstrated by the fact that pediatric urologic surgery is usually a daytime procedure and that urologic surgery often involves vagal reflexes, such as changes in cough and RR, which may be more likely to induce ventilation abnormalities. In contrast, some studies in adults (eg, endoscopic retrograde cholangiopancreatography [27,28]) did not find a significant benefit of CO2 monitoring, perhaps because adults are more capable of compensating for hypoxia, and ventilation abnormalities progress more slowly. In emergency medicine [29] or gastrointestinal endoscopy [3], patients have more complex underlying diseases and the interventions used are more heterogeneous, which may have masked the advantages of monitoring; or their outcome definitions (eg, hypoxic threshold, duration) are different from those of the present study, leading to differences in results.

More importantly, the results of the present study support the synergistic value of IPI and CO2 mapping. IPI is not a single parameter, but a comprehensive score (1–10 points) that integrates multiple indicators such as RR, EtCO2, and pulse oximetry, and its early warning logic is closer to the overall judgment of clinical decision-making [18]. In clinical practice, early warning of IPI ≤6 can precede the signal of a change in EtCO2 (≥5 mmHg) or a decrease in RR (≤80% of baseline) alone. For example, when RR is slightly reduced but EtCO2 is not yet significantly elevated, IPI may have signaled “attention required” (5–6 points) in advance through multi-parameter integration, thus creating time for early intervention. The earlier and more frequent use of jaw thrust in the intervention group in the present study confirms the advantage of this integration warning. This benefit can be summarized in a clear causal chain: microfluidic CO2 mapping provides early recognition of hypoventilation by changes in the EtCO2 waveform (eg, flat line, plunging/precipitating), and the IPI provides a visual warning through a composite score (IPI ≤6) → earlier implementation of less invasive interventions (eg, jaw thrust) are prompted → airway obstruction or shallow slow breathing is corrected to prevent further worsening of inadequate ventilation → the incidence of oxygen desaturation (pulse oximetry <95%) is reduced → less need for more invasive interventions (eg, positive-pressure ventilation). The results of “increased mandibular thrust” versus “decreased positive-pressure ventilation” in the present study (positive-pressure ventilation: 1.64±1.31 vs 2.39±1.48) provide potential evidence for this chain.

PSA provides anxiolysis, analgesia, sedation, and motor control during the diagnosis and treatment of pain and discomfort, thereby enhancing patient tolerance and acceptance [3]. However, adverse effects associated with sedation, such as hemodynamic instability, airway obstruction, aspiration, and respiratory depression, are common. Preventing or detecting these side effects is essential to ensure the safety of pediatric patients [4]. Nevertheless, visual assessment and conventional monitoring methods often fail to detect several episodes of respiratory depression and hypotension. Hypoventilation and hypoxemia are distinct phenomena [30]. Pulse oximetry usually reflects only hypoxemia, whereas the RR and observed chest wall motion shown on the electrocardiogram may be interrupted and low tidal volumes may not be detected. Therefore, effective monitoring of ventilatory function is essential.

From a clinical practice perspective, combined monitoring prompts earlier, less invasive interventions (eg, jaw thrust), which are valuable not only for reducing hypoxemia but may also reduce the potential risks of highly invasive interventions. Positive-pressure ventilation, while rapidly correcting hypoxia, can increase the risk of gastric distension and reflux aspiration [31], especially in pediatric patients, where airway and GI structures are more fragile and such risks require vigilance. The reduced need for positive-pressure ventilation in the intervention group in the present study may have indirectly reduced the potential risk of such complications (although not directly observed in the present study). In addition, although not directly assessed in this study, the potential health economic value of reducing severe hypoxic events and highly invasive interventions, which may shorten postoperative recovery time and reduce the cost of adverse event management, warrants further exploration.

The results of this study suggest that integrating end-tidal carbon dioxide monitoring with IPI into routine pediatric urological PSA practice is an effective safety enhancement strategy. For clinicians, it is reasonable to consider implementing this combined monitoring for pediatric patients undergoing PSA, particularly those with higher respiratory risk factors, when resources permit. Future studies should validate these findings in a multicenter setting, further explore cost-effectiveness, evaluate application across different types of pediatric surgeries and in children with upper-airway anomalies, and develop standardized intervention protocols based on such monitoring to achieve broader safety benefits.

A strength of this study is the prospective randomized controlled trial design used to control bias through rigorous randomization and standardized procedures. The focus on a specific scenario, “elective PSA in pediatric urology,” resulted in a homogeneous population with well-targeted results. A commercially proven device (Capnostream™ 20p) was used, and the intervention program (monitoring indicators, intervention thresholds) was clearly defined and highly reproducible. This study also had some negative findings, such as no significant difference in the incidence of perioperative complications (eg, intraoperative hypotension, postoperative nausea and vomiting) between the 2 patient groups. This suggests that while combined monitoring reduces respiratory-related risks, it does not significantly affect other types of perioperative outcomes.

However, there are limitations to this study. First, being a single-center study with a relatively homogenous patient population and healthcare process, results need to be extrapolated to other centers with caution. Second, this study was “single-blind” (patient-blind only), the investigator who performed the intervention was aware of the subgroups, and there may have been a performance bias. For example, a greater focus on ventilation signals for the intervention group resulted in a more aggressive recording of interventions such as jaw thrust, which may have affected interpretation of the results. Although bias was minimized by standardizing the intervention process (clear definition of hypoventilation and management steps), it could not be completely excluded. Third, this study excluded patients with nasal congestion (eg, adenoid hypertrophy), which is common in pediatrics, so the results may not be directly applicable to patients with upper-airway obstruction. Fourth, the results may be dependent on specific algorithms (eg, IPI calculation logic, apnea alert thresholds) of the Capnostream™ 20p, which may perform differently on other brands of devices and require further validation. Fifth, long-term outcomes (eg, respiratory events more than 24 h postoperatively) and cost-effectiveness were not assessed to fully reflect the combined value of monitoring. Finally, the false-positive rates of CO2 monitoring and IPI were not documented in this study, and excessive false-positives can reduce clinician acceptance and affect practical application.

Conclusions

CO2 and IPI monitoring can reduce intraoperative hypoxemia risk in pediatric urologic PSA. This strategy enables anesthesiologists to implement less invasive airway interventions in a timelier manner (57% increase in the frequency of mandibular thrust use) by early identification of inadequate ventilation (eg, IPI ≤6 or abnormal EtCO2), which in turn reduces the need for positive-pressure ventilation (31% decrease in the frequency of use), without increasing perioperative complications. This finding provides evidence-based support for optimizing respiratory management in pediatric PSA. We recommend that this combined monitoring be considered as a supplementary tool for managing PSA in high-risk infants in clinical practice, particularly when routine oxygen administration can mask early hypoventilation. Future research should focus on validating its benefits in broader populations and establishing evidence-based monitoring and intervention protocols.