Nasal high-frequency oscillatory ventilation versus nasal continuous positive airway pressure as primary respiratory support strategies for respiratory distress syndrome in preterm infants: a systematic review and meta-analysis

Nasal high-frequency oscillatory ventilation (NHFOV) is a new respiratory support strategy despite lacking of enough evidence in preterm infants with respiratory distress syndrome (RDS). The aim of the present systematic review was to explore whether NHFOV reduced the intubation rate as compared with nasal continuous positive airway pressure (NCPAP) as the primary respiratory support strategies in preterm infants with RDS. Medline, the Cochrane library, the Cochrane Controlled Trials Register, EMBASE, Chinese National Knowledge Infrastructure (CNKI), and Wanfang data Information Site were searched from inception to Jan 1, 2021(Prospero2019 CRD42019129316, date and name of registration: Apr 23,2019, The clinical effectiveness of NHFOV vs NCPAP for preterm babies with respiratory distress syndrome). Pooled data from clinically randomized controlled trials (RCTs) comparing NHFOV with NCPAP as the primary respiratory supporting strategies in preterm infants with RDS were performed using the fixed-effects models whenever no heterogeneity was shown. The primary outcome was intubation rate. Four randomized controlled trials involving 570 participants were included. Comparing with NCPAP, NHFOV resulted in less intubation (relative risk (RR) 0.44; 95% confidence interval (CI) 0.29–0.67, P = 0.0002), and heterogeneity was not found among the trials in the fixed-effects model (P = 0.78, I2 = 0%). Similar result also appeared in sensitivity analysis after excluding one study with significant difference (RR 0.44; 95% CI 0.25–0.78, P = 0.005) (P = 0.58, I2 = 0%). Conclusion: NHFOV decreased the intubation rate as compared with NCPAP as primary respiratory supporting strategies in preterm infants suffering from RDS. Future research should assess whether NHFOV can reduce the incidence of bronchopulmonary dysplasia (BPD) and intubation rate in preterm infants with BPD. Fund by Natural Science Foundation of Chongqing (cstc2020jcyj-msxmX0197), and “guan’ai” preterm Study Program of Renze Foundation of Beijing(K022). What is Known: • Nasal high-frequency oscillatory ventilation (NHFOV) has been described to be another advanced version of nasal continuous positive airway pressure (NCPAP). However, its beneficial effects among different studies as the primary modes in the early life of preterm infants with respiratory distress syndrome (RDS) were inconsistent. What is New: • Comparing with NCPAP, NHFOV decreases the risk of intubation as a primary respiratory supporting strategy in early life for preterm infants suffering from RDS. What is Known: • Nasal high-frequency oscillatory ventilation (NHFOV) has been described to be another advanced version of nasal continuous positive airway pressure (NCPAP). However, its beneficial effects among different studies as the primary modes in the early life of preterm infants with respiratory distress syndrome (RDS) were inconsistent. What is New: • Comparing with NCPAP, NHFOV decreases the risk of intubation as a primary respiratory supporting strategy in early life for preterm infants suffering from RDS.

age (11.9% vs 32.4%, 95%CI, 0.088 to 0.898, P = 0.032) and BW < 1500 g (10.4% vs 29.6%, 95%CI, 0.104 to 0.736, P = 0.010). However, as far as the primary outcome was concerned, NHFOV was not shown superior to NCPAP (9.9% vs 17.3%, 95%CI, 0.264 to 1.031, P = 0.066). It was consistent with the report of Malakian et al. [19] in which NHFOV did not reduce the need for IV during the first 72 h after birth as compared with NCPAP. Therefore, there is an urgent need to assess the beneficial effects between NHFOV and NCPAP on intubation rate and the other complications.
The purpose of the systematic review and meta-analysis was to evaluate whether NHFOV would reduce the need for intubation as compared with NCPAP in the treatment of preterm infants with RDS as the primary respiratory supporting modes.

Methods
The protocol of this systematic review and meta-analysis was registered before the study search in PROSPERO (Pros-pero2019 CRD42019129316, date and name of registration: Apr. 23, 2019, The clinical effectiveness of NHFOV vs NCPAP for preterm babies with respiratory distress syndrome). And it was performed conforming to the Methodological Expectations of Cochrane Intervention Reviews and was shown in accordance with the recommendations of the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) statement [20].

Types of studies, eligibility criteria, and interventions
The criteria for a trial to be included in the meta-analysis were as follows: (1) trial involving human newborn infants in the early stage of RDS; (2) trial comparing NHFOV with NCPAP as the primary respiratory support strategies; (3) it was a randomized controlled trial. Only RCTs are included because they are the best design to obtain unbiased assessment of beneficial effects through restricting the known and underlying confounders affecting the primary outcome.
The excluded criteria were (1) trials involving animal or pre-clinical studies, (2) duplicate reports, (3) studies without the primary outcome, and (4) the studies were cross-over, quasi-RCTs, or observational studies.

The primary and secondary outcome measures
The primary outcome was intubation rate within 72 h after birth, and the secondary outcomes included the incidences of bronchopulmonary dysplasia (BPD) [21], necrotizing enterocolitis(NEC) [22], air leak and in-hospital mortality.

NEC
Necrotizing enterocolitis NCPAP Nasal continuous positive airway pressure NIPPV Nasal intermittent positive pressure ventilation NHFOV Nasal high-frequency oscillatory ventilation

Introduction
Invasive ventilation (IV) is a key procedure to reduce respiratory mortality and morbidity in preterm infants with respiratory distress syndrome (RDS). However, it is associated with the increased risks of bronchopulmonary dysplasia (BPD) [1], re-hospitalization, and brain injury among the survivors [2][3][4].
Therefore, how to avoid IV is a critical procedure for reducing respiratory mortality and morbidity in early life after birth.
To date, nasal continuous positive airway pressure (NCPAP) remains one widely used way of the noninvasive ventilation modes to avoid intubation in preterm infants. However, the failure rate is high and ranges from about 20 to 60% [5][6][7][8]. Nasal intermittent positive pressure ventilation (NIPPV) is another respiratory supporting strategy. Systematic review has proven that NIPPV is superior to NCPAP to avoid re-intubation in preterm infants after extubation [7]. But, early use of NIPPV was not shown beneficial effects for decreasing the need for intubation as compared with NCPAP alone in preterm infants with RDS [5,6].
Nasal high-frequency oscillatory ventilation (NHFOV) appears to combine the advantages of continuous distending pressure and non-invasive ventilation without need for synchronization [9][10][11] and may reduce the incidence of intubation in preterm infants with RDS. A multicenter study has indicated that NHFOV is related to the reduced numbers of apnea, bradycardia, and oxygen desaturation as a remedial measure after failing other noninvasive modes in preterm infants [12]. It is consistent with the previous small sample and non-randomized controlled studies [13][14][15]. Despite no compelling evidences about the safety and efficacy of NHFOV from multicenter randomized controlled trials, NHFOV is widely used, and no obvious side effects are observed [16]. In 2017, we reported a small randomized controlled study, and the result showed that comparing with NCPAP, NHFOV was associated with lower rate of intubation as primary respiratory support strategy [17]. It was consistent with the subgroup analysis of Zhu et al. [18] of which NHFOV showed a lower rate of treatment failure than NCPAP in the strata of 26 +0 to 29 +6 weeks' gestational

Search strategies
A systematic literature search was conducted in Jan 1, 2021, using the methods of the Cochrane Collaboration for Systematic Reviews of Interventions [20]. We searched studies included in Medline, the Cochrane library, the Cochrane Controlled Trials Register, EMBASE, Chinese National Knowledge Infrastructure (CNKI), and Wanfang data Information Site. Furthermore, any ongoing or unpublished trials were also identified. We did not restrict studies according to language and searched for studies written in any language from inception to Jan 1, 2021.
The following keywords were used: "noninvasive highfrequency oscillatory ventilation" or "nasal high-frequency oscillatory ventilation" or "noninvasive high-frequency oscillation ventilation" or "nasal high-frequency oscillation ventilation" and "preterm" or "infant." Other neonatal experts were asked to search any unpublished or ongoing trials, and two studies were identified. We applied the Cochrane sensitivity-maximizing and Cochrane sensitivity and precision-maximizing strategies as our special search strategies [20].

Collection and assessment of the included and excluded studies
The obtained studies through the search strategies above were imported to an electronic bibliographic management program. We reviewed the titles and abstracts of the remaining articles and excluded those that were not related to our topic and those that did not meet the eligibility criteria. The full-text versions were obtained for the relevant articles that could be included in the review.
The research strategies, article-extracting, and data analysis were performed independently by three reviewers (Li Jie, Chen Long, Shi Yuan). Data analysis included study design, study interventions, number of subjects in each group, demographic characteristics, inclusion and exclusion criteria, primary and secondary outcomes, and variables used to assess study quality.

Statistical analyses
The Cochrane Risk of Bias tool [20] was applied to assess the methodological quality of the included studies. The statistical analyses were similar to the previous study [23]. Discrepancies among the three reviewers were resolved by discussion. Meta-analysis was performed using version 5.2 of Review Manager. To assess heterogeneity, 2 distribution and Higgins I 2 statistics were calculated to determine the percentage of total variation across studies resulting from heterogeneity. I 2 statistics approximating 25%, 50%, and 75% were considered low, medium, and high heterogeneity, respectively. The fixed-effects models were present, and the random-effects models were used whenever considerable heterogeneity was shown. For categorical data, the effect is expressed as the relative risk (RR), and for continuous data, the effect is expressed as the weighted mean difference (MD) (95% confidence interval (CI)). Otherwise, to further assess the effect of the individual study on the primary outcome, sensitivity or subgroup analysis would be performed after excluding the included studies with significant difference.

Description of the included studies
Generally, fourteen randomized controlled studies were identified, of which ten were excluded for the following causes: one was the protocol [24] and one was the preliminary report [25] in Chinese by Zhu et al. [17], three were compared after extubation [26][27][28], three were randomized crossover studies [13,29,30], one was compared between NHFOV and DuoPAP [31], and one was compared between NHFOV and BP-CPAP as rescuing treatment of failure of NCPAP [14]. Finally, four eligible studies and 570 participants were included in the subsequent analysis ( Fig. 1) [17][18][19]32]. Tables 1 and 2 summarize the characteristics of these studies.

Quality assessment of included studies
The assessment of publication bias of the included four RCTs is summarized in Fig. 2. All four studies were randomized controlled trials. The random sequence generation, concealment of allocation, and blinding of participants and personnel were thought to be adequate in four studies. Moderate to high risk of publication bias was shown in two studies [19,32]. Publication bias was from the blinding of outcome assessment, incomplete outcome data, and selective reporting.

The primary and secondary outcomes
The meta-analysis indicated that NHFOV resulted in less intubation rate (RR 0.44; 95% CI 0.29-0.67, P = 0.0002) as compared with NCPAP, and heterogeneity was not found among the four trials in the fixed-effects model (P = 0.78, I 2 = 0%) (Fig. 4). In sensitivity analysis, after excluding the study by Zhu et al. [17], the result also showed that NHFOV reduced the incidence of intubation (RR 0.44; 95% CI 0.25-0.78, P = 0.005) as compared with NCPAP, and heterogeneity was not found among the other three trials in the fixed-effects model (P = 0.58, I 2 = 0%). Otherwise, after excluding the study by Iranpour et al., the present study also demonstrated that NHFOV reduced intubation rate (RR 0.47; 95% CI 0.30-0.72, P = 0.0006) as compared with NCPAP, and heterogeneity was not found in the fixed-effects model (P = 0.92, I 2 = 0%).

Discussion
In the present systematic review and meta-analysis, we aimed to evaluate whether NHFOV would reduce intubation rate as compared with NCPAP as the primary respiratory supporting modes in preterm infants with RDS. As a result, we found that NHFOV decreased the intubation rate as compared with NCPAP, and no heterogeneity was found. Similar result also appeared in sensitivity analysis after excluding one study with significant difference. And it was consistent with the subgroup analysis of the multicenter study by Zhu et al. [18] The result suggests that NHFOV is a more reasonable selection to reduce the risk of intubation as compared with NCPAP in preterm neonate with RDS.
The reasonable cause to explain the differences among the studies might be the severities of RDS. Similar intubation rate may appear between the two groups in preterm infants with mild-moderate RDS, not severe RDS. Among the trials included in the present study, the diagnosis of "the included criteria" was not consistent. The diagnosis of the included criteria was "RDS" in the studies by Malakian et al. [19] and Iranpour et al. [32]. But it was "moderate-severe RDS" and "moderate RDS" in the studies by Zhu et al. [17] and Zhu et al. [18]. It was consistent with the previous reports by Mukerji et al. [12] and Wang et al. [35] in which NHFOV was successfully used in reducing intubation  In contrast, similar intubation rate was found between NHFOV and NCPAP when the infants were stable on NCPAP after extubation [29].
A pathophysiological plausibility is that NHFOV lead to more removal of CO 2 than NCPAP. Gaertner et al. [36] performed a randomized crossover trial comparing NHFOV with NCPAP, and the result showed that in preterm infants, oscillatory volumes during NHFOVs are transmitted to the lung. Compared with the regional distribution of tidal breaths, oscillations preferentially reach the right and nongravity-dependent lung. Some clinical studies also suggest the beneficial effects of NHFOV over NCPAP for CO 2 clearance [37,38]. Bottino et al. [39] performed a multicenter non-blinded prospective randomized crossover study, and the results showed that NHFOV delivered through nasal prongs was more effective than NCPAP in improving the elimination of CO 2 (47.5 ± 7.6 versus 49.9 ± 7.2 mmHg, P = 0.0007). Meantime, a different TcCO2 behavior was found according to the random sequence: in patients starting on NCPAP, TcCO 2 significantly decreased from 50.0 ± 8.0 to 46.6 ± 7.5 mmHg during NHFOV (P = 0.001). In patients starting on NHFOV, TcCO 2 slightly increased from 48.5 ± 7.8 to 49.9 ± 6.7 mmHg during NCPAP (P = 0.13). In the present study, the included study by Malakian A et al. also reported that PCO 2 level changes (mmHg) in the first half hour and 6 h after entering the study in the NCPAP and NHFOV groups were 50.6, 42.7, and 52.7, 32.8 mmHg, respectively. In the NHFOV group, the PCO 2 in neonates with RDS was significantly decreased (P = 0.001) [19].
Besides efficacy, safety is another important focus when NHFOV was used. Similar to other noninvasive ventilation, NHFOV could also result in side effects. In the four included studies, Malakian et al. [19] reported that there were no differences in traumatization of nasal skin and mucosa, air leaks, intraventricular hemorrhage, feed intolerance, and time to full feeds between the two groups. And there were no relevant reports in the studies by Iranpour et al. [32] and Zhu et al. [17]. Air-trapping and NEC were shown in the study of Czernik C et al. [37]. Ruegger et al. perform a randomized, crossover trial, and the results showed that comparing with NCPAP, NHFOV reduced the number of desaturations and bradycardia but was associated with increased oxygen requirements and higher heart rates in preterm infants [30]. Zhu et al. [18] indicated that the rate of thick secretions causing an airway obstruction was higher in the NHFOV group than in the NCPAP group (13.8 vs. 5.3%, 95% CI of risk difference 1.9-15.1, P = 0.01). It was consistent with the study by Fischer HS et al. [16]. The bench study by Ulrich et al. showed NHFOV settings with low frequencies, high amplitudes, and high I:E ratios may place infants at an increased risk of upper airway desiccation [40].
The major limitations of the present study: (1) The inclusion criteria of the meta-analysis allowed inclusion of RCTs concerned with more mature babies whose respiratory symptoms were probably due to other causes than RDS. (2) The initiation and adjustment of respiratory parameters of NHFOV were different among the four trials. (3) The assessment for secondary outcomes was not enough in the small sample size. (4) The four trials were performed in Iran and China, but no data from western developed countries. (5) The INSURE technique needed further assessment due to no detailed description on the use of caffeine and sedation [41]. They might induce potential bias, including restricted application scope. These problems could be overcome in additional studies according to the present data. Recently, we have organized multi-center, randomized controlled trials regarding comparing NHFOV and NCPAP as the respiratory support modes after extubation in preterm infants (NCT03099694), and the results could give us more reasonable explanations.
In summary, among preterm infants with RDS, NHFOV decreased the intubation rate as compared with NCPAP as the primary respiratory support strategies in the early life. Future research should assess whether NHFOV can reduce the incidence of BPD and intubation rate in preterm infants with BPD. Fig. 4 The comparison of the intubation between groups