Abstract
Background:
Children undergoing moderate to deep sedation for diagnostic and therapeutic procedures are susceptible to hypoxemia because of their anatomical and physiologic features. However, optimal oxygen administration methods are unclear. This study aimed to evaluate the efficacy of oxygen supplementation during sedation using either low-flow or high-flow nasal cannula.
Methods:
This prospective, multicenter randomized controlled trial included children (younger than 18 yr) undergoing moderate to deep sedation. The participants were randomly assigned to three groups as follows: (1) control (no oxygen), (2) low-flow (2 to 6 l/min oxygen via nasal cannula), and (3) high-flow (oxygen administration via high-flow nasal cannula with a flow rate of 2 l/kg and 50% fraction of inspired oxygen). The primary outcome was hypoxemia incidence (saturation of peripheral oxygen, oxygen saturation measured by pulse oximetry 95% or less for more than 5 s). Secondary outcomes included oxygen saturation measured by pulse oximetry less than 90%, rescue interventions, and sedation-related complications. Between-group differences were compared using a logistic regression model.
Results:
A total of 253 participants were randomized, with 250 completing the study. Hypoxemia occurred in 27.6% of participants in the control group, 7.2% in the low-flow group, and 1.2% in the high-flow group (P < 0.001). The odds of hypoxemia in the low-flow and high-flow groups were lower than that in the control group (odds ratio [OR], 0.184; 95% CI, 0.067 to 0.503; P = 0.001 for low-flow; OR, 0.026; 95% CI, 0.003 to 0.207; P < 0.001 for high-flow). However, hypoxemia incidence of the high-flow group was not statistically lower than the low-flow group (OR, 0.143; 95% CI, 0.017 to 1.245; P = 0.078). Rescue interventions were conducted more frequently in the control group (52.9%) than in the low-flow (10.8%) and high-flow (3.6%) groups (P < 0.001). Sedation-related complications such as desaturation and apnea were lower in the low-flow and high-flow groups than in the control group (P < 0.001).
Conclusions:
Routine oxygen supplementation prevents hypoxemia during pediatric moderate and deep sedation. Low-flow oxygen can be a reasonable choice as it effectively reduces hypoxemia while being more cost-effective and widely accessible than high-flow oxygen.
In this randomized controlled trial of children undergoing sedation, both low-flow and high-flow oxygen reduced hypoxemia compared to no supplemental oxygen. The study was not powered to determine whether high-flow is superior to low-flow or whether they are equivalent.
Editor’s Perspective
What We Already Know about This Topic
Children often receive sedation for diagnostic and therapeutic procedures. The benefits of high-flow nasal cannula oxygenation during procedural sedation have been reported in high-risk adult patients, but data in children are limited.
What This Article Tells Us That Is New
In this randomized controlled trial of children undergoing sedation, both low-flow and high-flow oxygen reduced hypoxemia compared to no supplemental oxygen. The study was not powered to determine whether high-flow is superior to low-flow or whether they are equivalent.
Procedural sedation is the mainstay in pain and anxiety management in children undergoing diagnostic and therapeutic procedures. Pediatric patients are prone to upper airway obstruction, other respiratory events, and hypoxemia. Additionally, the frequent need for a deeper level of sedation due to greater anxiety or reduced cooperation makes them more susceptible to respiratory depression.1 Therefore, continuous oxygenation monitoring and supplemental oxygen administration may help reduce the risk of these complications.2 However, the optimal method of oxygen delivery remains unclear.
In clinical practice, oxygen supplementation is commonly provided via a conventional nasal cannula (low-flow oxygen, low-flow) or a high-flow nasal cannula (HFNC). HFNC delivers heated, humidified air at a high flow rate, maintains positive end-expiratory pressure, and consistently delivers oxygen with the ability to provide a fixed high fraction of inspired oxygen (Fio2). Therefore, the use of HFNC has increased not only in critically ill patients with respiratory failure but also areas of anesthesia, including tubeless airway surgery, preoxygenation, difficult airway management, and prevention of postoperative atelectasis.3–6
The benefits of HFNC during procedural sedation for high-risk adult patients have been reported.7–9 Insufficient studies have compared HFNC and standard oxygenation supply during procedural sedation in children.1,10–12 Among the available studies, the results vary in terms of efficacy and difference in hypoxemia incidence between two oxygen supply methods. HFNC most effectively prevents hypoxemia during sedation; nonetheless, it may not be necessary for all patients undergoing sedation. HFNC equipment may remain unavailable in all sedation settings. Additionally, clinicians should consider its cost-effectiveness, possible complications such as nasal mucosal injury, hypercapnia or barotrauma, and patient comfort.13
Therefore, in this study, we aimed to explore whether routine oxygen supply during moderate to deep sedation is necessary for preventing hypoxemia and hypoventilation in children without pulmonary disease. Additionally, we aimed to determine if HFNC is the best method for oxygen delivery. We evaluated the efficacy of oxygen administration in children undergoing moderate to deep sedation and compared hypoxemia incidence between oxygen delivery methods using a conventional nasal cannula and HFNC.
Materials and Methods
Ethical Statement
This prospective, multicenter, randomized controlled trial was approved by the institutional review board of the Seoul National University Hospital, Seoul, Korea (H2103-095-1205; date of approval, April 12, 2021) and Seoul St. Mary’s Hospital (KC21ONDI0543; date of approval, September 9, 2021). It is registered at https://clinicaltrials.gov (No. NCT04852432; principal investigator, K.J.T.; date of registration, April 15, 2021). The study was conducted per the ethical standards set by the 1964 Declaration of Helsinki and its later amendments. Before each procedural sedation, an anesthesiologist met with the child’s parents, explained the study protocol, and obtained written informed consent.
Inclusion Criteria
The inclusion criteria were as follows: (1) children younger than 18 yr; (2) with American Society of Anesthesiologists (Schaumburg, Illinois) Physical Status class I, II, and III; and (3) required moderate to deep sedation for examination or endoscopic procedure between April 2021 and April 2024.
Exclusion Criteria
The exclusion criterion was as follows: children with respiratory disease or respiratory distress, pneumothorax, increased intracranial pressure, recent history of nasal bleeding, airway surgery, complete nasal obstruction, skull base fracture, and pulmonary hypertension.
Group Allocation
The children were randomly assigned to three groups, namely, the control, low-flow, and high-flow groups. In the control group, nasal prongs were applied to all participants. However, oxygen was not administered. Children in the low-flow group received supplemental oxygen using nasal prongs, and the oxygen flow rate was set to 2 l/min or less for infants and 6 l/min or less for children.14 Oxygen was delivered by connecting nasal prongs to the auxiliary oxygen port of the ventilator, and the flow rate was adjusted to maintain Fio2 at 0.5, as monitored by the ventilator. Children in the high-flow group received warmed and humidified air through an Optiflow device (Fisher and Paykel Healthcare, New Zealand) at a flow rate of 2 l/kg·min. The initial Fio2 was set at 0.5 (fig. 1).
Fig. 1.
Study protocol. Fio2, fraction of inspired oxygen.
Randomization
The mixed block randomization method was utilized, with block sizes of three or six and an allocation ratio of 1:1:1. Randomization was generated using SAS software (SAS Institute Inc., USA). The stratification factors included the institution (Seoul National University Hospital, Seoul St. Mary’s Hospital) and age (0 to 12 months [infant], 13 to 36 months [toddler], 37 months to 5 yr [preschool], 6 to 12 yr [grade school], and 13 to 18 yr [teen]). Web-based random allocation was conducted using a randomization table prepared by the Seoul National University Hospital Medical Research Collaborating Center. The management and operation were overseen by the Seoul National University Medical Research Collaborating Center (Medical Research Collaborating Center). Randomization information was accessed and managed independently from the clinical investigator. An anesthetic nurse, not associated with the study, conducted a random allocation sequence by preparing coded and sealed opaque envelopes for allocation concealment. The children, guardians, and investigators were blinded to group allocation, whereas the attending anesthesiologists providing sedation were not blinded.
Sedation and Monitoring
In this study, the safety and effectiveness of sedation for pediatric surgeries was ensured through comprehensive monitoring. Vital parameters, including pulse oximetry for oxygen saturation, capnography for end-tidal carbon dioxide (ETco2) levels, transcutaneous carbon dioxide, blood pressure, and heart rate were continuously assessed.
For premedication, 1 mcg/kg dexmedetomidine (maximum 50 mcg) greater than 10 min or 0.05 to 0.1 mg/kg midazolam (maximum 5 mg) were administered. During maintenance, propofol was administered as a bolus of 1 mg/kg, followed by continuous infusion at 75 to 100 mcg/kg/min. Additional medications, such as ketamine, were administered based on clinical judgment. Capnography monitoring was conducted using nasal prongs for both the control and low-flow groups. For capnography monitoring for the high-flow group, a 20-gauge catheter was attached to the end of the carbon dioxide sampling line and positioned in front of the child’s nose or mouth. Monitoring was initiated after generating a graph, despite a small amplitude.
Sedation effectiveness was evaluated using the Pediatric Sedation State Scale (PSSS)15 to ensure optimal sedation levels and participant safety. The PSSS ranges from 0 to 5, with 5 indicating that the patient is fully awake and alert, and 0 representing deep sedation associated with abnormal physiologic parameters or anesthesia where the child is unarousable. Sedation was conducted with the target of achieving a PSSS between 1 and 2.
Outcome Variables
Our primary objective was to evaluate the efficacy of oxygen therapy and the impact of HFNC on oxygenation and ventilation impairment in children undergoing moderate to deep sedation. The primary outcome included the incidence of hypoxemia based on different oxygen delivery methods during sedation. Hypoxemia was defined as oxygen saturation measured by pulse oximetry (Spo2) 95% or less during the procedure for more than 5 s.16
The secondary outcomes included Spo2 90% or less during sedation; maximum, minimum, and average carbon dioxide partial pressure measured by transcutaneous monitoring; minimum Spo2 during sedation; the need for rescue interventions to improve oxygenation and respiration (such as oxygen administration [for the control group], increasing oxygen flow rate [for the low-flow group], increasing Fio2 [for the high-flow group], jaw thrust, head extension, insertion of oral or nasal airway, positive pressure ventilation using a facial mask, insertion of a supraglottic airway device, intubation, and others), and sedation-related complications. We considered desaturation (Spo2 less than 95%) regardless of duration as one of the sedation-related complications, which was different from our primary and secondary outcomes.
Sample Size Calculation
First, we hypothesized that the incidence of hypoxemia would differ between the control and low-flow groups and the control and high-flow groups. Second, we hypothesized that this rate would differ between the low-flow and high-flow groups. Both hypotheses were assessed in two stages. If the hypoxemia incidence differed between the control and low-flow or high-flow groups in the first stage, the second hypothesis was assessed. Both hypotheses were assessed sequentially to avoid an increase in Type I error; however, the power of the test in the second stage depended on its power in the first stage.
For hypotheses 1 and 2, significance levels (α) of 0.025 and 0.05, respectively, were selected, with a Type II error rate (β) of 0.10 to maintain a power of 90%. Assuming equal distribution among the three groups (control, low-flow, and high-flow), the required sample size was based on the anticipated incidence of hypoxemia in each group. Previously, HFNC reduced the risk of hypoxemia from 8.6% to 0%17 and 20% to 2%18 during moderate to deep sedation, compared with oxygenation using a facial mask or nasal prong. Without oxygen, Spo2 less than 95% occurred in 47% of adults during endoscopy.19 Therefore, we assumed that Spo2 less than 95% would occur in 1%, 15%, and 40% of participants in the high-flow, low-flow, and control groups, respectively. Considering a 10% dropout rate and assumed hypoxemia incidence in each group, the sample size was calculated as 258 (86 participants per group). PASS 15 software (NCSS Statistical Software, USA) was used for sample size calculation.
Statistical Analysis
All statistical analyses were conducted using SAS (version 9.4 or later, SAS Institute, USA) with a two-sided significance level of 5%. Demographic and baseline characteristics were summarized for each group, with continuous variables presented as mean ± SD and categorical variables presented as frequencies and percentages.
For categorical variables, the incidence and frequency of occurrences were presented within each group. Between-group differences were compared using a logistic regression model, adjusted for stratification factors (institution, age) and confounding variables (underlying diseases and the type of procedures). Post hoc analyses were conducted if the test results among the three groups demonstrated 5% significance. Initially, the significance between the control and low-flow groups and between the control and high-flow groups was assessed at 2.5%. Then, for significant comparisons, the incidence of hypoxemia between the low-flow and high-flow groups was assessed at 5%. For group effects significant at 5%, the adjusted odds ratio (OR) was presented with 95% CIs and P values.
Descriptive statistics are presented for the continuous variables. Group comparisons were analyzed using linear regression, adjusted for the stratification factors (institution, age) and confounding variables (underlying diseases and the type of procedures). For 5% significant differences among the three groups, the least square means difference was presented along with 95% CIs and P values.
Demographic information and sedation characteristics were obtained for all participants. The outcome analysis was primarily based on a full analysis set, with the supplementary analysis based on the per-protocol (PP) set. The full analysis set includes all randomized subjects who met the inclusion and exclusion criteria, and the PP set consisted of subjects from the full analysis set who completed the study without any major violations, with completed measurement of efficacy outcomes.
Results
A total of 258 participants enrolled in this trial; of them, 5 did not meet the inclusion criteria and were excluded. Consequently, 253 participants were randomized (control group, 87; low-flow group, 83; and high-flow group, 83). Of them, three (one in the control group, two in the high-flow group) discontinued the trial prematurely. Therefore, 250 participants (control group, 86; low-flow group, 83; and high-flow group, 81) were included in the PP analysis (fig. 2). Table 1 summarizes their demographic data. There were no significant differences in age, sex, height, weight, and body mass index between groups. In addition, no group difference was found in the distribution of the patient’s underlying disease. Table 2 summarizes the procedures and sedative medications. The mean sedation time or types of procedures did not differ among the three groups. Propofol was the most commonly used sedative in all groups (94.3%, 91.6%, and 91.6% of the control, low-flow, and high-flow groups, respectively), followed by midazolam, dexmedetomidine, and ketamine.
Fig. 2.
Consolidated standards of reporting trials diagram. FAS, full allocation set; HF, high flow; HFNC, high flow nasal cannula; LF, low flow; PP, per-protocol.
Table 1.
Baseline Characteristics and Procedure Information of Study Population
| Variables | Control Group (n = 87) | LF Group (n = 83) | HF Group (n = 83) |
|---|---|---|---|
| Age | 5.3 ± 5.4 (range, 0–17) | 4.9 ± 5.3 (range, 0–16) | 5.2 ± 5.3 (range, 0–17) |
| ≤ 12 months | 30 (34.5) | 30 (36.1) | 29 (34.9) |
| 13–36 months | 9 (10.3) | 9 (10.8) | 9 (10.8) |
| 37 months–5 yr | 6 (6.9) | 5 (6.0) | 6 (7.2) |
| 6–12 yr | 26 (29.9) | 26 (31.3) | 25 (30.1) |
| 13–18 yr | 16 (18.4) | 13 (15.7) | 14 (16.9) |
| Sex (M/F) | 51/36 (58.6/41.4) | 45/38 (54.2/45.5) | 47/36 (56.6/43.4) |
| Height (cm) | 105.1 ± 37.8 | 102.9 ± 38.4 | 104.7 ± 37.5 |
| Weight (kg) | 23.2 ± 18.7 | 21.5 ± 18.1 | 22.5 ± 17.2 |
| BMI (kg/m2) | 17.8 ± 3.1 | 17.3 ± 3.2 | 17.7 ± 2.6 |
| ASA Physical Status classification | 2 (2–2) | 2 (2–2) | 2 (2–2) |
| Median PSSS during sedation | 2 (2–2) | 2 (2–2) | 2 (2–2) |
| Underlying disease | |||
| Neurologic | 20 (23.0) | 15 (18.1) | 12 (14.5) |
| Hemato-oncologic | 15 (17.2) | 13 (15.7) | 13 (15.7) |
| Endocrine | 10 (11.5) | 10 (12.0) | 10 (12.0) |
| Gastrointestinal | 10 (11.5) | 10 (12.0) | 10 (12.0) |
| Renal/urologic | 7 (8.0) | 8 (9.6) | 8 (9.6) |
| Spinal cord/orthopedic | 22 (25.3) | 21 (25.3) | 21 (25.3) |
| Others | 3 (3.4) | 6 (7.2) | 9 (10.8) |
Data are presented as mean ± SD, median (interquartile ranges), and n (%).
ASA, American Society of Anesthesiologists; F, female; HF, high flow; LF, low flow; M, male; PSSS, Pediatric Sedation State Scale.
Table 2.
Sedation Characteristics of Study Population
| Variables | Control Group (n = 87) | LF Group (n = 83) | HF Group (n = 83) | P Value |
|---|---|---|---|---|
| Duration of sedation (min) | 40.6 ± 17.8 | 40.2 ± 16.2 | 43.2 ± 24.1 | 0.879 |
| Procedures | 0.923 | |||
| Urodynamic study | 42 (48.8) | 48 (57.8) | 43 (51.8) | 0.453 |
| EGD, colonoscopy | 27 (31.4) | 23 (27.7) | 25 (30.1) | 0.888 |
| MRI | 8 (9.2) | 5 (6.0) | 7 (8.4) | 0.728 |
| Ureteral stent removal | 7 (8.0) | 5 (6.0) | 5 (6.0) | 0.830 |
| Ear procedures | 1 (1.1) | 2 (2.4) | 1 (1.2) | 0.761 |
| Other procedures | 2 (2.3) | 0 | 2 (2.4) | 0.370 |
| Medications used during sedation | 0.840 | |||
| Propofol (IV) | 82 (94.3) | 76 (91.6) | 76 (91.6) | 0.743 |
| Midazolam (IV) | 54 (62.1) | 57 (68.7) | 57 (68.7) | 0.572 |
| Dexmedetomidine (IV) | 22 (25.3) | 14 (16.9) | 18 (21.7) | 0.406 |
| Ketamine (IV) | 4 (4.6) | 6 (7.2) | 7 (8.4) | 0.592 |
Data are presented as mean ± SD and n (%).
EGD, esophagogastroduodenoscopy; HF, high flow; IV, intravenous; LF, low flow; MRI, magnetic resonance imaging.
Hypoxemia occurred in 24 (27.6%), 6 (7.2%), and 1 (1.2%) children in the control, low-flow, and high-flow groups, respectively (P < 0.0001). After adjusting for the stratification factors (institution and age), compared with the control group, the estimated OR of hypoxemia for the low-flow and high-flow groups was 0.184 (95% CI, 0.067 to 0.503; P = 0.001) and 0.026 (95% CI, 0.003 to 0.207; P < 0.001), respectively. Compared with the low-flow group, the OR for hypoxemia in the high-flow group was 0.143 (95% CI, 0.017 to 1.245; P = 0.078). Similar results were observed in the PP analysis.
Spo2 less than 90% for more than 5 s occurred in four (4.6%), two (2.4%), and no children in the control, low-flow, and high-flow groups, respectively. The mean ETco2 during sedation was lower in the high-flow group than in other groups (all P < 0.001). However, the mean transcutaneous carbon dioxide levels did not differ among the groups. Rescue interventions were more frequent in the control group compared with the low-flow and high-flow groups (52.9% vs. 10.8% vs. 3.6%, respectively; P < 0.001; table 3).
Table 3.
Comparison of Outcome Variables in Study Cohort
| Variables | Control Group (n = 87) | LF Group (n = 83) | HF Group (n = 83) | P Value |
|---|---|---|---|---|
| Hypoxemia (< 95% for >5 s) | 24 (27.6) | 6 (7.2) | 1 (1.2) | < 0.001* |
| Odds ratio (95% CI) | 1.000 | 0.184 (0.067 to 0.503) | — | 0.001† |
| 1.000 | — | 0.026 (0.003 to 0.207) | < 0.001† | |
| — | 1.000 | 0.143 (0.017 to 1.245) | 0.078† | |
| Hypoxemia (<95% for > 5 s), PP | 24 (27.9) | 6 (7.2) | 1 (1.2) | < 0.001* |
| Odds ratio (95% CI) | 1.000 | 0.182 (0.067 to 0.499) | — | < 0.001† |
| 1.000 | — | 0.027 (0.004 to 0.216) | < 0.001† | |
| — | 1.000 | 0.151 (0.017 to 1.307) | 0.086† | |
| Hypoxemia (<90% for > 5 s) | 4 (4.6) | 2 (2.4) | 0 (0.0) | 0.171‡ |
| Odds ratio (95% CI) | 1.000 | 0.571 (0.096 to 2.674) | — | 0.479† |
| 1.000 | — | 0.110 (0.001 to 1.067) | 0.058† | |
| Minimum Spo2 (%) | 94.1 ± 6.2 | 98.9 ± 2.6 | 99.3 ± 1.2 | < 0.001§ |
| Adjusted mean difference (95% CI) | Ref. | 4.802 (3.589 to 6.015) | — | < 0.001∥ |
| Ref. | — | 5.143 (3.931 to 6.355) | < 0.001∥ | |
| — | Ref. | 0.341 (–0.885 to 1.568) | 0.584∥ | |
| Mean ETco2 (mmHg) | 35.2 ± 6.41 | 34.2 ± 6.03 | 21.57 ± 12.31 | < 0.001§ |
| Adjusted mean difference (95% CI) | Ref. | –0.771 (–3.032 to 1.490) | — | 0.502∥ |
| Ref. | –13.395 (–15.755 to –11.035) | < 0.001∥ | ||
| Ref. | –12.624 (–15.002 to –10.246) | < 0.001∥ | ||
| Mean tCO2 | 39.9 ± 4.21 | 39.63 ± 4.39 | 39.34 ± 4.55 | 0.528§ |
| Patients with rescue intervention | 46 (52.9) | 9 (10.8) | 3 (3.6) | < 0.001* |
| Odds ratio (95% CI) | 1.000 | 0.101 (0.044 to 0.232) | — | < 0.001† |
| 1.000 | — | 0.030 (0.009 to 0.104) | < 0.001† | |
| — | 1.000 | 0.297 (0.077 to 1.147) | 0.078† | |
| Oxygen supplementation/increasing oxygen flow rate or Fio2 | 45 (51.7) | 4 (4.8) | 1 (1.2) | < 0.001* |
| Jaw thrust/head extension | 12 (13.8) | 7 (8.4) | 1 (1.2) | 0.009* |
| oral/nasal airway | 0 | 1 (1.2) | 0 | 0.358* |
| Assisted ventilation | 1 (1.1) | 0 | 0 | 0.384* |
| Reducing drug dose | 0 | 0 | 1 (1.2) | 0.358* |
Data are presented as mean ± SD and n (%).
Chi-square test; †Logistic regression analysis; ‡Fisher exact test; §Kruskal-Wallis test; ∥Linear regression analysis.
ETco2, end-tidal carbon dioxide; HF, high flow; LF, low flow; PP, per-protocol; tCO2, transcutaneous CO2.
Table 4 summarizes the incidence of sedation-related complications. Sedation-related complications occurred in 46 (52.9%), 9 (10.8%), and 3 (3.6%) children in the control, low-flow, and high-flow groups, respectively (P < 0.001). All complications were respiratory events, with desaturation being the most common complication, followed by apnea. Compared with the control group, the OR for complications was 0.108 and 0.033 for the low-flow and high-flow groups, respectively, indicating a reduced risk of complications. All sedation-related complications were resolved without sequelae, and no other severe complications occurred.
Table 4.
Occurrence of Sedation Related Complications
| Variables | Control Group (n = 87) | LF Group (n = 83) | HF Group (n = 83) | P Value |
|---|---|---|---|---|
| Patients with complications | 46 (52.9) | 9 (10.8) | 3 (3.6) | < 0.001* |
| Odds ratio (95% CI) | 1.000 | 0.108 (0.048–0.244) | < 0.001 † | |
| 1.000 | 0.033 (0.010–0.114) | < 0.001 † | ||
| 1.000 | 0.908 (0.080–1.183) | 0.086 † | ||
| Desaturation (< 95%) | 35 (40.2) | 4 (4.8) | 1 (1.2) | < 0.001* |
| Desaturation (< 90%) | 10 (11.5) | 3 (3.6) | 0 | 0.002* |
| Apnea | 6 (6.9) | 3 (3.6) | 2 (2.4) | 0.33* |
Data are presented as n (%).
Chi-square test; †Logistic regression analysis.
Discussion
This prospective, multicenter, randomized controlled trial describes the efficacy of oxygen supplementation and the clinical feasibility of HFNC in children undergoing moderate to deep sedation. The incidence of hypoxemia was significantly lower in both oxygen supplementation groups (low-flow and high-flow) compared to the control group, confirming the effectiveness of supplemental oxygen in preventing desaturation during pediatric sedation. However, while HFNC showed a numerically lower incidence of hypoxemia than low-flow oxygen, this difference did not reach statistical significance. The lower-than-expected event rate limits the precision of the estimate of a difference between these groups, and therefore, no definitive conclusion can be drawn as to whether one is superior to the other or that they are equivalent. Oxygen supplementation during moderate and deep sedation can prevent sedation-related respiratory events, such as desaturation and apnea.
Sedation can result in hypoventilation with increasing carbon dioxide,20 and oxygen supplementation during sedation can mitigate the effects of hypoventilation on oxygenation.12,21 In such cases, patients may experience severe hypoventilation without corresponding drops in oxygen saturation, leading to substantial respiratory compromise.22 According to adult sedation guidelines, the role of oxygenation during sedation remains debatable.23 However, children are physiologically more susceptible to upper airway obstruction during deep sedation, which can result in rapid-onset hypoxemia.24 Our findings support the need for supplemental oxygen during moderate to deep sedation in children, even in those without underlying respiratory conditions.1,2
In clinical settings, oxygen is administered via nasal cannula or facial mask during procedural sedation. However, HFNC has been introduced as a specialized device, increasing its use during sedation. It is superior to standard oxygen administration.1,11,17 In adult patients undergoing elective gastroscopy under propofol sedation, HFNC supportive oxygen therapy decreased the incidence of hypoxemia and severe hypoxemia (Spo2 less than 75%) compared with nasal cannula.17 In children undergoing dental procedures under propofol sedation, HFNC improved the lowest Spo2 and upper airway obstruction more effectively than nasal cannula.1 Similar results were observed in children undergoing flexible bronchoscopy.11
However, the Fio2 level differed among the studies; in some studies, oxygen was administered via HFNC with Fio2 of 1.0.11,17 Administering high-flow oxygen with Fio2 of 1.0 throughout sedation was not physiologic and disadvantageous for the low-flow group. Therefore, we selected Fio2 of 0.5 in both high-flow and low-flow groups by controlling the flow rate based on Fio2 monitored on a ventilator in the low-flow group. Consequently, conventional oxygen administration sufficiently reduced hypoxemia and sedation-related complications in children under moderate to deep sedation.
The incidence of hypoxemia, number of rescue interventions, and sedation-related complications did not differ between the low-flow and high-flow groups. However, in the high-flow group, only one child experienced desaturation, and significant hypoxemia (Spo2 less than 90%) did not occur. Of the children requiring rescue interventions, jaw thrust/head extension maneuvers were used in 45, 7, and 1 child in the control, low-flow, and high-flow groups, respectively, suggesting that even low-flow oxygen supplementation helped reduce the need for airway interventions. The higher intervention rate in the control group may be due to an increased risk of desaturation from shallow breathing and atelectasis, which pediatric patients are more prone to during sedation.25 Without supplemental oxygen, these effects likely led to more frequent airway maneuvers. In contrast, low-flow oxygen likely compensated for mild desaturation, reducing the need for intervention. The benefits of HFNC, including reduced dead space and positive end-expiratory pressure, are well-known and may be particularly advantageous for patients requiring oxygen support before sedation or undergoing invasive procedures such as bronchoscopy.26
Capnography measures exhaled carbon dioxide levels and serves as an early indicator of hypoventilation, offering a more immediate indication of respiratory compromise compared to other indicators such as low Spo2 level.27 Capnography is highly recommended for monitoring during sedation with high-flow nasal oxygenation. However, the high-flow flow rates with HFNC are expected to considerably dilute exhaled carbon dioxide, making accurate sampling of carbon dioxide challenging or impossible.28 Additionally, HFNC did not effectively eliminate carbon dioxide during apnea in children with high metabolic rates.29 In this study, ETco2 levels were lower in the high-flow group than in other groups, while transcutaneous carbon dioxide pressure levels remained similar, suggesting potential underestimation of carbon dioxide levels with ETco2 monitoring in patients receiving oxygen via HFNC.
Our study has several limitations. First, this study was not blinded for the attending anesthesiologists, which may have influenced clinical decision-making. Second, we did not stratify the results based on the type of procedure, which may have influenced the efficacy of different oxygen delivery methods. However, the diversity of underlying diseases and procedures in our study population enhanced the generalizability of our findings across various developmental stages and procedural contexts. Third, we did not evaluate the long-term outcomes of oxygen therapy, which limited the understanding of its sustained effects in children. We exclusively enrolled healthy children without pre-existing respiratory conditions; therefore, the role of HFNC may differ in individuals with respiratory compromise. Fourth, the actual incidence of hypoxemia in the control and low-flow groups was lower than the rates used for sample size calculation. Therefore, the study was underpowered to test for any difference between the low-flow and high-flow groups, and no conclusion can be drawn as to whether one is superior to the other or that they are equivalent. A larger study would be required to test this hypothesis. Finally, we did not explore the benefits of HFNC in more invasive procedures, such as bronchoscopy, or in patients with underlying respiratory diseases. HFNC may be particularly beneficial in patients undergoing procedures involving airway manipulation or in patients with respiratory distress or failure. Nonetheless, further research is needed to confirm this finding in a broader clinical setting.
In conclusion, routine oxygen supplementation reduces the incidence of hypoxemia during moderate to deep sedation in children. In this study, HFNC showed a numerically lower incidence of hypoxemia compared to low-flow oxygen, although the difference did not reach statistical significance. While HFNC may provide additional benefits in specific patient groups, low-flow oxygen was effective in preventing hypoxemia and required fewer resources. These findings highlight the need for further research with larger sample sizes to better define the comparative advantages of low-flow oxygen and oxygen delivery via HFNC during pediatric sedation.
Research Support
This research was supported by a grant from the Patient-Centered Clinical Research Coordinating Center (PACEN) funded by the Ministry of Health & Welfare, Seoul, Republic of Korea (grant No. RS-2020-KH094707).
Competing Interests
The authors declare no competing interests.
Reproducible Science
Full protocol available at: jintae73@gmail.com. Raw data available at: jintae73@gmail.com.
Abbreviations:
- ETco₂
- end-tidal carbon dioxide
- Fio₂
- fraction of inspired oxygen
- HFNC
- high-flow nasal cannula
- OR
- odds ratio
- PP
- per-protocol
- PSSS
- Pediatric Sedation State Scale
- Spo₂
- oxygen saturation measured by pulse oximetry
Published online first on April 11, 2025.
J.-H.L. and H.K. contributed equally to this article.
The article processing charge was funded by the authors.
Contributor Information
Ji-Hyun Lee, Email: muslab6@snu.ac.kr.
Hyun Jung Ko, Email: uglyko@hanmail.net.
Jung-Bin Park, Email: jb4001@snu.ac.kr.
Sang-Hwan Ji, Email: jsh1@snu.ac.kr.
Jin-Tae Kim, Email: kimjintae73@dreamwiz.com;jintae73@gmail.com;01119@snuh.org.
References
- 1.Sago T, Watanabe K, Kawabata K, Shiiba S, Maki K, Watanabe S: A nasal high-flow system prevents upper airway obstruction and hypoxia in pediatric dental patients under intravenous sedation. J Oral Maxillofac Surg 2021; 79:539–45. doi:10.1016/j.joms.2020.10.018 [DOI] [PubMed] [Google Scholar]
- 2.Cote CJ, Wilson S; American Academy of Pediatrics; American Academy of Pediatric Dentistry: Guidelines for monitoring and management of pediatric patients before, during, and after sedation for diagnostic and therapeutic procedures. Pediatrics 2019; 143:e20191000. doi:10.1542/peds.2019-1000 [PubMed] [Google Scholar]
- 3.Lee JH, Rehder KJ, Williford L, Cheifetz IM, Turner DA: Use of high flow nasal cannula in critically ill infants, children, and adults: A critical review of the literature. Intensive Care Med 2013; 39:247–57. doi:10.1007/s00134-012-2743-5 [DOI] [PubMed] [Google Scholar]
- 4.Kim EH, Ji SH, Lee JH, et al. : Use of high-flow nasal oxygen in spontaneously breathing pediatric patients undergoing tubeless airway surgery: A prospective observational study. Medicine (Baltimore) 2022; 101:e29520. doi:10.1097/MD.0000000000029520 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Patel A, Nouraei SA: Transnasal humidified rapid-insufflation ventilatory exchange (THRIVE): A physiological method of increasing apnoea time in patients with difficult airways. Anaesthesia 2015; 70:323–9. doi:10.1111/anae.12923 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Lee JH, Ji SH, Jang YE, Kim EH, Kim JT, Kim HS: Application of a high-flow nasal cannula for prevention of postextubation atelectasis in children undergoing surgery: A randomized controlled trial. Anesth Analg 2021; 133:474–82. doi:10.1213/ANE.0000000000005285 [DOI] [PubMed] [Google Scholar]
- 7.Sakazaki R, Suzuki T, Ikeda N: High-flow nasal cannula oxygen supported-transesophageal echocardiography under sedation in a respiratory compromised patient. J Cardiothorac Vasc Anesth 2019; 33:255–6. doi:10.1053/j.jvca.2018.10.009 [DOI] [PubMed] [Google Scholar]
- 8.Lee CC, Perez O, Farooqi FI, Akella T, Shaharyar S, Elizee M: Use of high-flow nasal cannula in obese patients receiving colonoscopy under intravenous propofol sedation: A case series. Respir Med Case Rep 2018; 23:118–21. doi:10.1016/j.rmcr.2018.01.009 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Lomas C, Roca O, Álvarez A, Masclans JR: Fibroscopy in patients with hypoxemic respiratory insufficiency: Utility of the high-flow nasal cannula. Respir Med CME 2009; 2:121–4. doi:10.1016/j.rmedc.2008.12.008 [Google Scholar]
- 10.Duan X, Wei N, Wei J, et al. : Effect of high-flow nasal cannula oxygen therapy on pediatric patients with congenital heart disease in procedural sedation: A prospective, randomized trial. J Cardiothorac Vasc Anesth 2021; 35:2913–9. doi:10.1053/j.jvca.2021.03.031 [DOI] [PubMed] [Google Scholar]
- 11.Sharluyan A, Osona B, Frontera G, et al. : High flow nasal cannula versus standard low flow nasal oxygen during flexible bronchoscopy in children: A randomized controlled trial. Pediatr Pulmonol 2021; 56:4001–10. doi:10.1002/ppul.25655 [DOI] [PubMed] [Google Scholar]
- 12.Klotz D, Seifert V, Baumgartner J, Teufel U, Fuchs H: High-flow nasal cannula vs standard respiratory care in pediatric procedural sedation: A randomized controlled pilot trial. Pediatr Pulmonol 2020; 55:2706–12. doi:10.1002/ppul.24975 [DOI] [PubMed] [Google Scholar]
- 13.Li J, Albuainain FA, Tan W, Scott JB, Roca O, Mauri T: The effects of flow settings during high-flow nasal cannula support for adult subjects: A systematic review. Crit Care 2023; 27:78. doi:10.1186/s13054-023-04361-5 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Milesi C, Boubal M, Jacquot A, et al. : High-flow nasal cannula: Recommendations for daily practice in pediatrics. Ann Intensive Care 2014; 4:29. doi:10.1186/s13613-014-0029-5 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Cravero JP, Askins N, Sriswasdi P, Tsze DS, Zurakowski D, Sinnott S: Validation of the Pediatric Sedation State Scale. Pediatrics 2017; 139:e20162897. doi:10.1542/peds.2016-2897 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.von Ungern-Sternberg BS, Boda K, Chambers NA, et al. : Risk assessment for respiratory complications in paediatric anaesthesia: A prospective cohort study. Lancet 2010; 376:773–83. doi:10.1016/S0140-6736(10)61193-2 [DOI] [PubMed] [Google Scholar]
- 17.Lin Y, Zhang X, Li L, et al. : High-flow nasal cannula oxygen therapy and hypoxia during gastroscopy with propofol sedation: A randomized multicenter clinical trial. Gastrointest Endosc 2019; 90:591–601. doi:10.1016/j.gie.2019.06.033 [DOI] [PubMed] [Google Scholar]
- 18.Teng WN, Ting CK, Wang YT, et al. : High-flow nasal cannula and mandibular advancement bite block decrease hypoxic events during sedative esophagogastroduodenoscopy: A randomized clinical trial. Biomed Res Int 2019; 2019:4206795. doi:10.1155/2019/4206795 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Wang CY, Ling LC, Cardosa MS, Wong AKH, Wong NW: Hypoxia during upper gastrointestinal endoscopy with and without sedation and the effect of pre-oxygenation on oxygen saturation. Anaesthesia 2000; 55:654–8. doi:10.1046/j.1365-2044.2000.01520.x [DOI] [PubMed] [Google Scholar]
- 20.McQuillen KK, Steele DW: Capnography during sedation/analgesia in the pediatric emergency department. Pediatr Emerg Care 2000; 16:401–4. doi:10.1097/00006565-200012000-00005 [DOI] [PubMed] [Google Scholar]
- 21.Miner JR, Heegaard W, Plummer D: End-tidal carbon dioxide monitoring during procedural sedation. Acad Emerg Med 2002; 9:275–80. doi:10.1111/j.1553-2712.2002.tb01318.x [DOI] [PubMed] [Google Scholar]
- 22.Abramo TJ, Wiebe RA, Scott S, Goto CS, McIntire DD: Noninvasive capnometry monitoring for respiratory status during pediatric seizures. Crit Care Med 1997; 25:1242–6. doi:10.1097/00003246-199707000-00029 [DOI] [PubMed] [Google Scholar]
- 23.Hinkelbein J, Lamperti M, Akeson J, et al. : European Society of Anaesthesiology and European Board of Anaesthesiology guidelines for procedural sedation and analgesia in adults. Eur J Anaesthesiol 2018; 35:6–24. doi:10.1097/EJA.0000000000000683 [DOI] [PubMed] [Google Scholar]
- 24.Isono S: Instability of upper airway during anesthesia and sedation: How is upper airway unstable during anesthesia and sedation? Structure-Function Relationships in Various Respiratory Systems: Connecting to the Next Generation. Edited by Yamaguchi K. Singapore, Springer Singapore, 2020, pp 67–91 [Google Scholar]
- 25.Riva T, Pascolo F, Huber M, et al. : Evaluation of atelectasis using electrical impedance tomography during procedural deep sedation for MRI in small children: A prospective observational trial. J Clin Anesth 2022; 77:110626. doi:10.1016/j.jclinane.2021.110626 [DOI] [PubMed] [Google Scholar]
- 26.Pelaia C, Bruni A, Garofalo E, et al. : Oxygenation strategies during flexible bronchoscopy: A review of the literature. Respir Res 2021; 22:253. doi:10.1186/s12931-021-01846-1 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Wollner E, Nourian MM, Booth W, et al. : Impact of capnography on patient safety in high- and low-income settings: A scoping review. Br J Anaesth 2020; 125:e88–e103. doi:10.1016/j.bja.2020.04.057 [DOI] [PubMed] [Google Scholar]
- 28.Greenland KB: A potential method for obtaining wave-form capnography during high flow nasal oxygen. Anaesth Intensive Care 2019; 47:204–6. doi:10.1177/0310057X19836430 [DOI] [PubMed] [Google Scholar]
- 29.Humphreys S, Lee-Archer P, Reyne G, Long D, Williams T, Schibler A: Transnasal humidified rapid-insufflation ventilatory exchange (THRIVE) in children: A randomized controlled trial. Br J Anaesth 2017; 118:232–8. doi:10.1093/bja/aew401 [DOI] [PubMed] [Google Scholar]