Comparison of Two Weaning Methods from Heated Humidified High-Flow Nasal Cannula Therapy in Pediatric Intensive Care Unit

Abstract

Background and Objective:

Although high-flow nasal cannula (HFNC) is widely used in children, there is no consensus on the methods for starting, maintenance, and weaning. The aim of this study was to compare weaning methods in children.

Methods:

The study included all patients in pediatric intensive care unit (PICU) who were started on HFNC treatment. The respiratory assessment score was used in the decisions for starting, continuing, and weaning from HFNC. The patients who responded and for whom weaning was planned were randomized by month into 2 groups as directly weaned from HFNC and weaned by reducing the flow. Success rates, treatment, and length of stay (LOS) in weaning methods were compared.

Results:

Of the 145 patients initially included in the study, 32 (22%) were excluded, and analysis was made of 113 patients. Successful weaning from HFNC was obtained in 76.9% of the patients, in 82.1% of flow weaning, and 73.6% of direct weaning, with no statistically significant difference determined between the groups (P = 0.286). The median duration of HFNC and the median LOS in PICU were determined to be statistically significantly shorter in direct weaning than in flow weaning [36 h interquartile range (IQR) 24–48 h] versus 60 h (IQR 60–72 h), P < 0.001 and 6 days (4–14 days) versus 9.5 days (5.25–20.75 days, P = 0.043, respectively).

Conclusion:

In patients who responded to HFNC in PICU, the responses to direct weaning and flow reduction were seen to be similar. In patients directly weaned off, both the HFNC duration and LOS in PICU were significantly shorter.

Introduction

High-flow nasal cannula (HFNC) is a noninvasive respiratory support treatment method providing continuously heated and humidified oxygen. In this method, the flow should be set at or above the patient's inspiratory flow rate. Although the definition of high-flow is debatable, considering the flow limitation in adults (50–60 L/min), a reasonable flow rate for patients is thought to be 1–2 L/kg/min up to 10 kg, followed by an increase of 0.5 L/kg/min.¹ With the high-flow effect washing the nasopharyngeal dead space, more effective oxygenation and gas exchange can be provided by increasing the amount of oxygen and carbon dioxide within the alveoli.

The frequency of HFNC use in children continues to increase. The fact that it is easy to set up and follow-up, requires less sedation, does not interfere with oral feeding, and can be used before and during transfer in the intensive care unit (ICU) increases its importance in Pediatric ICU (PICU). There are studies in the literature related to safe and effective use, primarily in bronchiolitis, and in asthma, pneumonia, sleep apnea syndrome, during patient transport, and for postextubation respiratory support indications.^1,2 A scoring method in HFNC treatment was investigated in a study of 584 patients, and the results showed that patients who underwent weaning with scoring had better results both in terms of success rates and length of stay (LOS).³

A survey study by Miller et al. showed no consensus on the definition of HFNC, how to set initial flow, or how to make adjustments.⁴ There were also differences in the definition of HFNC, which was defined by 49% of respondents as any heated gas delivered by nasal cannula, and by, 21% as heated gas delivered via nasal cannula at a flow rate equal to or greater than the patient's inspiratory demand.⁴

In a survey of 176 pediatric intensive care physicians from 36 centers by Suzanne et al., 50% of the participants reported that they directly terminated HFNC therapy and switched to room air or low-flow oxygen therapy.⁵ They suggested that this approach would speed up the weaning process, however, it might result in a lower success rate. In the same study, a written weaning protocol was available in only 10% of pediatric centers.

No randomized controlled trial could be found in literature that has investigated the efficacy of HFNC weaning methods in children. By providing the weaning protocol and method, we think that the duration of intensive care and hospital stay and related complications can be reduced, and HFNC devices can be used more efficiently.

In the Betters et al. study, Respiratory Assessment Score (RAS) was used in the HFNC treatment decision. In addition to the decision of HFNC treatment, RAS was served as an assessment if patients required positive pressure or only oxygen support.⁶ In our study, we determined the HFNC requirement with RAS. The aim of this study was to use the RAS to optimize the starting, maintenance, and weaning decisions for HFNC and to compare the efficacy of the HFNC weaning methods of directly weaning off and flow reduction weaning.

Methods

This prospective study was planned to include patients aged 1 month–17 years who were admitted to the 11-bed PICU between September 31, 2019 and April 1, 2021 and were started on HFNC treatment. The study protocol is shown in Table 1, including the exclusion criteria, the contraindications for HFNC, initial settings, definition of nonresponse, and criteria for weaning and weaning nonresponse.

Table 1.

High-Flow Nasal Cannula Initial Settings, Criteria of Nonresponse, Exclusion Criteria, Contraindications, Criteria of Nonresponse, Follow-Up, and Weaning Protocol

Initial settings	Flow:
	Body weight ≤10 kg; 2 L/kg/min
	Body weight >10 kg; 0.5–1 L/kg maximum 50 LPM
	FiO2
	If SpO2 >92%, FiO2 = 40%
	If SpO2 ≤92%, start with FiO2 = 100% and decrease FiO2 by 5%–10% every 5–10 min (until it is reduced to 40%) every 5–10 min, with SpO2 >92%
	Temperature
	36°C
Exclusion criteria	HFNC therapy initiated before PICU admission or patients transferred from the PICU while HFNC therapy is ongoing Patients who were HFNC device dependent at home before admission and those discharged as new device dependents. HFNC therapy started prophylactically for postextubation (without scoring) Patients receiving HFNC therapy via tracheostomy cannula Subsequent admissions of patients with multiple PICU admissions and HFNC therapy
Contraindications	Nasal anatomical defects Presence of recent nasal-maxillofacial trauma and surgery Presence of air leak syndromes Hemodynamic instability Neurological instability (GCS below 8)
Nonresponder HFNC	Respiratory arrest Desaturation Worsening of retraction Hypercapnia Circulation disorder Unexplained tachycardia Unconsciousness Patient noncompliance If FiO2 cannot be reduced at the end of the first hour in patients with 100% FiO2
Weaning decision	Respiratory score ≤5 and FiO2 ≤ 40% evaluated at 12-h intervals from the 24th hour of treatment
Unsuccessful weaning	Respiratory score ≥6 within 48 h after weaning from HFNC and switching to HFNC or other respiratory support methods

FiO₂, fraction of inspired oxygen; GCS, Glasgow coma scale; HFNC, high-flow nasal cannula; PICU, pediatric intensive care unit.

HFNC treatment was applied with a Fisher Paykel Airvo2^® device. The nasal cannulas used were appropriate to the age groups, not exceeding 50% of the nasal cavity. In our unit, there is a ratio of 0.35 devices per bed. In the decisions for starting, maintaining, and weaning HFNC, a modified version of the respiratory score defined by Betters et al. was used (Table 2).⁶ The study protocol was prepared by the researchers. Accordingly, the initiation of treatment, titration, monitorization, and record-keeping according to this protocol was made by the attending physician in the PICU.

Table 2.

Respiratory Assessment Score

	0	1	2
Respiratory rate (bpm)
1–3 months	<55	55–70	>70
3–6 months	<45	45–65	>65
6–12 months	<40	40–60	>60
1–3 years	<30	30–45	>45
3–6 years	<25	25–35	>35
6–12 years	<22	22–33	>33
>12 years	<20	20–27	>27
Nasal flaring	None	Minimum	Marked
Intercostal retraction	None	Minimum	Marked
Subcostal-xiphoid retraction	None	Minimum	Marked
Chest movement	None	Forced breathing	Forced and irregular breathing
Expiratory grunt	None	Stethoscope audible	Directly audible

bpm, breaths per minute; FiO₂, fraction of inspired oxygen.

All patients included in the study were given oxygen therapy with a nonrebreathing mask before HFNC. According to the study protocol, patients examined with a score of 0–12, and with RAS ≥6 started HFNC treatment. On a nonroutine basis, according to the clinician's assessment, patients with agitation were started on ketamine at a dose of 0.5 mg/kg/min.

Throughout HFNC treatment, each patient was monitored with continuous electrocardiography, pulse rate, respiratory rate, and peripheral oxygen saturation. In patients meeting the criteria of nonresponse at any time during the treatment HFNC was terminated and treatment was changed to advanced respiratory support treatments (noninvasive or invasive mechanical ventilation).

The RAS of the patients were evaluated and recorded by the attending physician at the start of treatment, after 60 min, and in the 12th hour. For patients with a score of ≤5, HFNC weaning was planned, and these patients were randomly separated into 2 groups. In odd months, HFNC wean off was applied directly, and in even months, wean off was applied by reducing flow. According to this method, patients began to be allocated as reduced flow weaning in October 2019, direct weaning in November 2019, and were allocated in this way for the following 18 months.

Patients in the direct weaning group were weaned off directly when the fraction of inspired oxygen (FiO₂) ≤40%, RAS was ≤5 and to increase the success of weaning, nonrebreather oxygen mask was placed after HFNC. In flow weaning group when RAS was ≤5, the flow setting was reduced by 25% of the initial flow setting, once every 12 h. Accordingly, HFNC treatment was terminated at 3 steps over a 36 h period. In patients with lower initial flow setting due to lower body weight, HFNC treatment was terminated if the new flow setting after the reduction step was <4 L/min. Following HFNC, oxygen continued to be administered to all the patients with a nonrebreather mask.

Weaning was accepted as unsuccessful with a RAS of ≥6 within 36 h of weaning in direct weaning group, and within 36 h after the flow reduction process and weaning in flow weaning group. A form was prepared for each patient to record the HFNC indications, the RAS calculated every 12 h during the treatment, the RAS given in the weaning decision, the length of treatment (LOT) of HFNC, PICU LOS, response to HFNC, the respiratory support method applied in cases with nonresponse, the need for sedation, and HFNC-related complications.

Approval for the study was granted by the Clinical Research Ethics Committee of Ondokuzmayis University (decision no: 2020000254-2). Parental informed consent was obtained.

Statistical method

Data obtained in the study were analyzed statistically using SPSS for Windows vn.22.0 software. Descriptive statistics are stated as mean ± standard deviation or median and interquartile range (IQR) for continuous variables, and as number (n) and percentage (%) for categorical variables. In the comparisons between the groups, the chi-square test was applied to categorical data, the Independent Sample t-test to continuous variables with normal distribution, and the Mann–Whitney U-test to continuous data not showing normal distribution. A value of P < 0.05 was accepted as statistically significant. A pilot study was conducted, including 36 cases at the initiation of the study. The sample size was estimated using the software GPower vn.3.1.9.4 and, was calculated to be 77 (effect size = 0.321, α = 5%, power = 80%).

Results

In the 18-month period of this study, a total of 429 patients were admitted to the PICU, of which 162 (37.7%) started HFNC treatment. Of these, 17 patients did not meet the study inclusion criteria, and of the remaining 145 patients, 32 (22%) were excluded as they were nonresponsive to the HFNC treatment, so final evaluation was made of 113 patients (Fig. 1). For patients with recurrent admissions to PICU, only the first admission was included. In patients with repeated HFNC treatments in PICU, only the first was included in the study.

FIG. 1.

Flow chart of patients treated with HFNC. HFNC, high-flow nasal cannula; PICU, pediatric intensive care unit.

In 32 (22%) patients, nonresponse to HFNC was determined. The characteristics of the HFNC responsive and nonresponsive patients are shown in Table 3. The initial RAS was determined to be statistically significantly higher in HFNC nonresponders compared to the HFNC responders [median (IQR), 10 (8–10) and 9 (9–11); P < 0.001]. The Pediatric Risk of Mortality III (PRISM III) scores [median (IQR) 3 (2–4) versus 2 (1–3); P = 0.002] and mortality rates (34.4% versus 2.7%; P < 0.001) were significantly higher in the HFNC nonresponder group than in the responder group (Table 3).

Table 3.

Demographic and Clinical Characteristics of Patients and Comparison of Responders and Nonresponders

	Total (n = 145)	Responders (n = 113)	Nonresponders (n = 32)	P
Age (months) mean ± SD	40.2 ± 4	39.9 ± 4.55	41.1 ± 9.2	0.689
Gender (male %)	93 (64%)	74 (65.4%)	19 (59.4%)	0.537
Body weight (kg), mean ± SD	14.11 ± 1.23	14.44 ± 1.43	12.93 ± 2.37	0.662
Initial RAS, median (IQR)	9 (8–10)	9 (9–11)	10 (8–10)	0.001
Initial FiO2, mean ± SD (min–max)	57.46 ± 8.13 (35–82)	56.38 ± 7.69 (35–74)	61.28 ± 8.62 (45–82)	0.002
Initial flow (LPM), median (IQR)	18 (12–25)	20 (10–29)	16.50 (14.25–25)	0.903
HFNC LOT (hours), median (IQR)	36 (24–60)	48 (36–65.5)	8.5 (3.25–21.75)	<0.001
PICU LOS (days), median (IQR)	10 (5–21)	7 (4–16.5)	19 (11–29.75)	<0.001
Hospital LOS (days), median (IQR)	14 (6–24.5)	9 (5–21)	21.5 (17–32)	<0.001
PRISM III, median (IQR)	2 (1–3)	2 (1–3)	3 (2–4)	0.002
Sedation, n (%)	43 (29.6)	36 (31.8)	7 (21.8)	0.381
Mortality, n (%)	14 (9.7)	3 (2.7)	11 (34.4)	<0.001
Reason of HFNC initiation
Pneumonia, n (%)	44 (30.4)	34 (30)	10 (31.2)
Postextubation, n (%)	43 (29.6)	33 (29.2)	10 (31.2)
Bronchiolitis/reactive airway disease, n (%)	25 (17.2)	25 (22.1)	0 (0)
Neuromuscular, n (%)	17 (11.9)	11 (9.7)	6 (18.7)
Upper airway stenosis, n (%)	11 (7.5)	7 (6.2)	4 (12.5)
Others, n (%)	5 (3.4)	3 (2.6)	2 (6.2)

FiO₂, fraction of inspired oxygen; HFNC, high-flow nasal cannula; IQR, interquartile range; LOS, length of stay; LOT, length of treatment; LPM, liters per minute; PICU, pediatric intensive care unit; PRISM III, Pediatric Risk of Mortality III; RAS, respiratory assessment score; SD, standard deviation.

The indications for starting HFNC treatment in the 145 patients included in the study were pneumonia in 44 (30.4%) patients, as rescue treatment after extubation in 43 (29.6%), and bronchiolitis in 25 (17.2%). The median RAS on starting HFNC was 9 (range, 6–12), and the median RAS when the decision was made for weaning was 4 (range, 1–5) (Table 3). Sedation was required in 43 (29.6%) patients, including 36 (31.8%) responders and 7 (21.8%) nonresponders P = 0.381, and ketamine was used for sedation in all these patients. PICU LOS was median 7 days (IQR 4–16.5 days) and HFNC LOT was median 48 h (IQR 36–65.5 h). Of the 43 patients treated with HFNC after extubation, 10 were nonresponders and 33 were responders, and the median invasive mechanical ventilation (IMV) durations were 13.5 days (IQR 5.75–18.25) and 6 days (IQR 3–11), respectively (P = 0.381).

The HFNC responder patients for whom the weaning decision was taken were randomly separated into 2 groups, as direct weaning of 57 patients and flow weaning of 56 patients. The characteristics of both groups are shown in Table 4. There was no difference between the groups in respect of age, gender, and body weight. The median RAS values for starting HFNC and for weaning were similar in the 2 groups [starting median 8 (IQR 7–10) versus 9 (IQR 8–10); P = 0.456, weaning median 3 (IQR 3–4) versus 4 (IQR 3–4); P = 0.161].

Table 4.

Comparison of Direct Weaning and Flow Weaning

	Direct weaning (n = 57)	Flow weaning (n = 56)	P
Age (months), mean ± SD, min–max	43.3 ± 6.8 (2–168)	36.4 ± 6.07 (2–190)	0.557
Gender (male %)	36 (63.1%)	38 (67.8%)	0.693
Body weight (kg), mean ± SD, min–max	16.8 ± 2.41 (4–81)	12.03 ± 1.49 (3–65)	0.188
Initial RAS, median (IQR)	8 (7–10)	9 (8–10)	0.456
Weaning RAS, median (IQR)	3 (3–4)	4 (3–4)	0.161
Initial FiO2, mean ± SD (min–max)	55.29 ± 8.44 (35–74)	57.50 ± 6.74 (37–71)	0.129
Initial flow (LPM), median (IQR)	20 (12–30)	17.5 (10–27.25)	0.195
HFNC LOT (hours), median (IQR)	36 (24–48)	60 (60–72)	<0.001
PICU LOS (days), median (IQR)	6 (4–14)	9.5 (5.25–20.75)	0.043
Hospital LOS (days), median (IQR)	9 (5–16)	15 (6–25.5)	0.146
PRISM III, median (IQR)	2 (3–4)	2 (3–4)	0.872
Sedation, n (%)	17 (30)	19 (34)	0.689
Mortality, n (%)	1 (1.8)	2 (3.6)	0.618
Weaning success, n (%)	42 (73.6)	45 (82.1)	0.286
Reason of HFNC initiation, n (%)
Pneumonia	18 (31.5)	16 (28.5)
Postextubation	16 (28)	17 (30.3)
Bronchiolitis/reactive airway disease	13 (22.8)	12 (21.4)
Neuromuscular	5 (8.8)	6 (10.7)
Upper airway stenosis	2 (3.5)	5 (8.9)
Others	3 (5.2)	0 (0)

FiO₂, fraction of inspired oxygen; HFNC, high-flow nasal cannula; IQR, interquartile range; LOS, length of stay; LOT, length of treatment; LPM, liters per minute; PICU, pediatric intensive care unit; PRISM III, Pediatric Risk of Mortality III; RAS, respiratory assessment score; SD, standard deviation.

The frequency of the use of sedation was similar in both groups (30% versus 34%, P = 0.689). Median HFNC LOT was statistically significantly shorter in direct weaning at 36 h (IQR 24–48) compared to 60 h (IQR 60–72) in flow weaning (P < 0.001). The PICU LOS was significantly shorter in direct weaning at median 6 days (IQR 4–14) compared to 9.5 days (IQR 5.25–20.75) in flow weaning (P = 0.043). No significant difference was found between the direct weaning and flow weaning groups in terms of length of hospital stay [median (IQR), 9 days (5–16) and 15 days (6–25.5), respectively; P = 0.146].

Weaning from HFNC was successful in 87 (76.9%) patients, at a rate of 82.1% (n: 45) in flow weaning and 73.6% (n: 42) in direct weaning. The difference between the 2 groups in respect of weaning success, PRISM III scores and mortality rates were not statistically significant (P = 0.286, P = 0.872 and P = 0.618, respectively) (Table 4). Pre-extubation IMV durations of the patients who were given HFNC treatment for postextubation in direct weaning and flow weaning were compared, and no significant difference was found, [median (IQR), 6 days (3–10.5) versus 7 days (2.5–16.5) respectively; P = 0.971].

Of the 15 patients with unsuccessful weaning in direct weaning, 14 were applied with HFNC treatment again, and for 1 patient with a diagnosis of pneumonia, the decision made by the attending clinician was for bilevel positive airway pressure treatment. Of the 11 patients with unsuccessful weaning in flow weaning, 10 started HFNC treatment again. In the other 1 patient who started HFNC with a diagnosis of spinal muscular atrophy, treatment was directly changed to IMV after 16 h because of hypoventilation. Complications developed in 2 patients in flow weaning, as nasal bleeding in 1 and nasal dermatitis in the other 1.

Discussion

In this study, comparisons were made of 2 different HFNC weaning methods in patients who were started on HFNC in PICU for different indications and obtained response. Patients directly weaned off HFNC and those weaned by reducing the flow were compared, and no statistically significant difference was observed in terms of weaning success. HFNC has become a widely used respiratory support application in modern PICUs for pediatric patients with respiratory problems. The frequency of HFNC use in all the patients followed up in PICU in the current study was 37.7%.

It has been reported in literature that the most common indication for HFNC use in children is infants with bronchiolitis.^7–10 In a study by Kawaguchi et al.,¹¹ the most common indications for use were reported to be bronchiolitis (97%) and neuromuscular weakness (50%). In a study by Schibler et al., HFNC was applied to 56% of patients aged <2 years in PICU, most often with a diagnosis of bronchiolitis, and following HFNC, the requirement for intubation fell from 37% to 7%.¹² In children, HFNC has been mostly studied in patients with bronchiolitis, initiation and weaning recommendations for its use in other indications are insufficient.^5,12

In the current study, HFNC was started most often for pneumonia, followed by as postextubation rescue treatment. Acute bronchiolitis was determined to be the third most frequent indication for use. All the acute bronchiolitis cases in the current study were in the first year of the study before the COVID-19 pandemic. During the pandemic, HFNC was not applied in PICU because of a lower incidence of acute bronchiolitis resulting from the societal restrictions and public health measures, such as mask wearing. All the acute bronchiolitis cases responded to HFNC treatment and did not require any other respiratory support method.

In a study by Peterson et al., the use of an HFNC starting and weaning protocol applied under the control of respiratory physiotherapists was shown to reduce the HFNC treatment duration and related complications and to shorten the LOS in hospital and PICU.¹³ There is no consensus as to whether weaning from HFNC should be made directly or by reducing the flow.

In a survey study by Kawaguchi et al., which included 919 PICU specialists worldwide, there was great variability in the clinical use of HFNC, and it was emphasized that the most variability was observed in weaning and support treatments.¹¹ It was reported that in 68% of the participants, first FiO₂ was reduced (although it varied, most often when FiO₂ was reduced to 40%), 11% started with flow reduction and when it reached a specific rate (most often 30%), HFNC was directly terminated, and at a low rate such as 4%, weaning was applied by reducing the flow without changing FiO_2. In a survey study related to HFNC applied to adults, 81% of clinicians reported that first FiO₂ was reduced, 6% reduced flow first and 13% stated that it was necessary to reduce both at the same time.¹⁴

As there are no widely accepted criteria for the starting, maintenance, and weaning decisions for HFNC treatment in children, it has been reported that there is great variability in practice between clinicians.¹⁵ In the evaluation of HFNC efficacy in the survey studies by Kawaguchi et al., increased respiratory rate and breathing effort were reported to be the most used parameters, whereas heart rate and a scoring system were rarely used.¹¹ This variability reported in HFNC use makes it difficult to compare the results of studies conducted in this area. In addition, the lack of starting and weaning protocols may cause unnecessary initiation of HFNC and prolonged HFNC LOT and PICU LOS.

This result was also demonstrated in a study by Wiser et al.³ This may restrict those in need being able to access supportive treatments especially in countries with insufficient resources. The RAS used in the current study in the decisions for starting HFNC and weaning was observed to both prevent unnecessary HFNC use and shorten HFNC LOT and PICU LOS, although this varied according to indications and could not be proved as the HFNC LOT before the study could be included.

In this study, the rate of HFNC nonresponse was 22%. In a study by Morris et al. investigating HFNC use in PICUs in England and Ireland, this rate was 26%, and a young age, HFNC indication, high PIM2 score and unplanned admission were determined as independent risk factors for nonresponse.¹⁶ In the current study, the PICU LOS of nonresponders was longer, PRISM III score was higher, and which was similar to the findings of Morris et al.

Due to the low number of nonresponder patients, comparisons according to HFNC indications could not be made. The initial RAS was significantly high in patients determined as HFNC nonresponders suggests that the scoring system used was effective, and it can be considered that patients with high RAS should be monitored more closely in respect of nonresponse, and conversion to alternative respiratory support treatments should not be delayed.

HFNC LOT shows variability in pediatric patients and very few studies have reported HFNC LOT. Metge et al. reported LOS of 6 days in patients with bronchiolitis applied with HFNC.¹⁷ In the HFNC responder group in the current study, median LOT was 48 h and PICU LOS was 7 days. Successful HFNC weaning was achieved in 76% of the patients and there was no significant difference between the groups in respect of success rates. As expected, both the HFNC LOT and PICU LOS were significantly shorter in Group 1.

In a study by Besnier et al., for patients receiving HFNC treatment in PICU, the decision for continuing was made by intermittent HFNC treatment twice a day and it was reported that 70% of the patients could be weaned off from HFNC on the first attempt. Charvat et al. evaluated 283 infants diagnosed with bronchiolitis in the general ward and reported that with the creation of a “HFNC holiday protocol,” HFNC LOT and LOS were shortened.¹⁸

Limitations

There were some limitations to this study, primarily that the study was single-centered and the methodology was not double-blind randomized. Using calendar months for randomization could potentially introduce seasonal bias. All the patients applied with HFNC for different indications within the defined study dates were included. The efficacy of the weaning methods could show a difference according to indications, but as the patient numbers were not sufficient in each indication group, it was not possible to compare the weaning methods according to indications. In addition, due to the age range of the patients, the study results cannot be generalized to all pediatric age groups.

Conclusion

To the best of our knowledge, this is the first study to have compared the results of directly and flow-reducing wean off from HFNC in children. The results of the study suggest that generally pediatric patients can be directly weaned off from HFNC. Throughout the world, there has been a great increase in all respiratory support methods because of the COVID-19 pandemic. Lack of devices may be a problem, especially in countries with insufficient resources, and therefore, directly weaning off could lead to more effective use of these devices. Another advantage of directly weaning off is a reduction in PICU LOS and complications associated with a shortening of HFNC LOT. Nevertheless, there is a need for further studies to investigate HFNC weaning methods in different patient groups.

Authors' Contributions

M.Ü.: Article preparation, literature search, and review of article. H.A.: Data collection and analysis of data. H.E.K.K.: Data collection and analysis of data. N.Y.: Study design, literature search, and article preparation.

Author Disclosure Statement

No competing financial interests exist.

Funding Information

No funding was received for this article.