pediatrics
October 2015, VOLUME136 /ISSUE 4

Nebulized Hypertonic Saline for Acute Bronchiolitis: A Systematic Review

  1. Linjie Zhang, MD, PhDa,
  2. Raúl A. Mendoza-Sassi, MD, PhDa,
  3. Terry P. Klassen, MDb, and
  4. Claire Wainwright, MDc
  1. aFaculty of Medicine, Federal University of Rio Grande, Rio Grande, Brazil;
  2. bManitoba Institute of Child Health; Children’s Hospital Research Institute of Manitoba; Department of Pediatrics, University of Manitoba, Winnipeg, Canada; and
  3. cQueensland Children’s Medical Research Institute; Department of Respiratory and Sleep Medicine, Lady Cilento Children’s Hospital; School of Medicine, University of Queensland, Brisbane, Australia

Abstract

BACKGROUND AND OBJECTIVE: The mainstay of treatment for acute bronchiolitis remains supportive care. The objective of this study was to assess the efficacy and safety of nebulized hypertonic saline (HS) in infants with acute bronchiolitis.

METHODS: Data sources included PubMed and the Virtual Health Library of the Latin American and Caribbean Center on Health Sciences Information up to May 2015. Studies selected were randomized or quasi-randomized controlled trials comparing nebulized HS with 0.9% saline or standard treatment.

RESULTS: We included 24 trials involving 3209 patients, 1706 of whom received HS. Hospitalized patients treated with nebulized HS had a significantly shorter length of stay compared with those receiving 0.9% saline or standard care (15 trials involving 1956 patients; mean difference [MD] −0.45 days, 95% confidence interval [CI] −0.82 to −0.08). The HS group also had a significantly lower posttreatment clinical score in the first 3 days of admission (5 trials involving 404 inpatients; day 1: MD −0.99, 95% CI −1.48 to −0.50; day 2: MD −1.45, 95% CI −2.06 to −0.85; day 3: MD −1.44, 95% CI −1.78 to −1.11). Nebulized HS reduced the risk of hospitalization by 20% compared with 0.9% saline among outpatients (7 trials involving 951 patients; risk ratio 0.80, 95% CI 0.67–0.96). No significant adverse events related to HS inhalation were reported. The quality of evidence is moderate due to inconsistency in results between trials and study limitations (risk of bias).

CONCLUSIONS: Nebulized HS is a safe and potentially effective treatment of infants with acute bronchiolitis.

  • Abbreviations:
    AEs
    adverse events
    ALRIs
    acute lower respiratory infections
    BIREME
    Latin American and Caribbean Center on Health Sciences Information
    CI
    confidence interval
    CSS
    clinical severity score
    ED
    emergency department
    GRADE
    Grading of Recommendations, Assessment, Development and Evaluations
    HS
    hypertonic saline
    LOS
    length of stay
    MD
    mean difference
    NS
    normal saline
    RACS
    respiratory assessment change score
    RCTs
    randomized controlled trials
    RDAI
    respiratory distress assessment instrument
    RR
    risk ratio
    RSV
    respiratory syncytial virus
    RTI
    respiratory tract infection
  • Acute bronchiolitis in infancy, mainly caused by respiratory syncytial virus (RSV), is the most common lower respiratory infection and the leading cause of hospitalization in children younger than 2 years. In the United States, acute bronchiolitis in infancy is responsible for ∼150 000 hospitalizations each year at an estimated cost of $500 million.1,2 From 1992 to 2000, bronchiolitis accounted for ∼1 868 000 emergency department (ED) visits for children younger than 2 years.3 In the United Kingdom, hospital admissions for acute bronchiolitis increased from 21 330 in 2004 and 2005 to 33 472 in 2010 and 2011.4

    Globally, it has been estimated that, in 2005, at least 33.8 million episodes of RSV-associated acute lower respiratory infections (ALRIs) occurred in children younger than 5 years, with incidence in developing countries more than twice that of industrialized countries.5 In the same year, RSV-associated severe ALRIs were responsible for ∼3.4 million hospitalizations and 66 000 to 199 000 deaths in young children worldwide, with 99% of these deaths occurring in developing countries.

    Despite its high incidence and morbidity, there are few effective therapies for acute bronchiolitis in infancy, and the mainstay of treatment remains supportive care.6,7 Given the theoretical effects of hypertonic saline (HS) in reducing airway edema, unblocking mucus plugging, and improving mucociliary clearance, HS administered via nebulizer has been proposed as a potentially effective therapy for acute bronchiolitis in infants.8 The first randomized trial, published in 2002, showed a significant effect of nebulized 3% saline solution in improving symptom scores among 65 outpatients with acute bronchiolitis, as compared with 0.9% normal saline (NS).9 Over the past decades, a growing number of randomized trials have been undertaken to assess the effects and safety of nebulized HS in infants with acute bronchiolitis.1019 The Cochrane review published in 2013 including 11 randomized trials shows that nebulized 3% saline may significantly reduce the length of stay (LOS) in hospitalized infants with acute bronchiolitis and improve the clinical severity score (CSS) in both outpatient and inpatient populations.20 Since then, new trials with conflicting results have been published, and an updated synthesis of the literature is needed.21 We decided to conduct a new systematic review of currently available randomized trials to assess the efficacy and safety of nebulized HS in infants with acute bronchiolitis and to explore possible reasons for inconsistent results across trials. We hypothesize that nebulized HS may be less effective than previously claimed for acute bronchiolitis and effect size of HS may mainly depend on diagnostic accuracy of bronchiolitis and the treatment regimen.

    Methods

    We followed the recommendations of the Preferred Reporting Items for Systematic Reviews and Meta-Analyses statement for writing this systematic review and meta-analysis.22 The full review protocol is available in the supplementary material. We used different data sources, search strategy, and statistical techniques than that used in the 2013 Cochrane review.20

    Data Sources and Search Strategy

    We searched PubMed and the Virtual Health Library of the Latin American and Caribbean Center on Health Sciences Information (BIREME), which contains Medline, CENTRAL, LILACS, IBECS, and >20 other databases (www.bireme.br). All databases were searched from inception until May 2015. The search strategy on PubMed was as follows: (bronchiolitis OR “acute wheezing” OR “respiratory syncytial virus” OR RSV OR “parainfluenza virus”) AND (“hypertonic saline” OR “saline solution” OR 3% saline OR 5% saline OR saline). We used the limits of study type: clinical trial, randomized controlled trial (RCT). The search strategy on the Virtual Health Library of BIREME was as follows: bronchiolitis AND “hypertonic saline.” There was no restriction on language of publication. We also conducted a search of the ClinicalTrials.gov trials registry to identify completed but unpublished trials. We checked reference lists of all primary studies and review articles for additional relevant trials.

    Study Selection

    To be included in this review, studies had to meet all of the following criteria: (1) study design: RCTs or quasi-RCTs; (2) participants: infants up to 24 months of age with diagnosis of acute bronchiolitis; we classified participants into “inpatients” who were admitted to the hospital and “outpatients” who attended at an ambulatory care unit or ED; (3) interventions and comparisons: nebulized HS (≥3%) alone or mixed with bronchodilator, compared with nebulized NS alone or mixed with same bronchodilator, or standard treatment; (4) outcome measures: primary outcomes included LOS in hospital for inpatients defined as time to actual discharge or time taken to be ready for discharge, and admission rate for outpatients, and secondary outcomes included CSSs, rate of readmission to hospital or ED, oxygen saturation, respiratory rate, heart rate, time for the resolution of symptoms/signs, duration of oxygen supplementation, results of pulmonary function tests, radiologic findings, and adverse events (AEs). We excluded studies that included patients who had had recurrent wheezing or were intubated and ventilated, and studies that assessed pulmonary function alone.

    Two review authors (RM and LZ) independently assessed the titles and abstracts of all citations identified by the searches. We obtained the full articles when they appeared to meet the inclusion criteria or there were insufficient data in the title and abstract to make a clear decision for their inclusion. The definitive inclusion of trials was made after reviewing the full-text articles. We resolved any disagreements between the 2 review authors about study inclusion by discussion and consensus.

    Data Extraction and Management

    One review author (LZ) extracted study details from the included trials by using a standardized data extraction form. These were checked by another review author (RM). We resolved any disagreements by discussion and consensus. We extracted the following data: (1) study characteristics: year of publication, and country and setting of study; (2) methods: study design, methods of random sequence generation, allocation concealment and blinding, and description of withdrawal; (3) participants: sample size, age, gender, and inclusion and exclusion criteria; (4) interventions and controls: concentration and volume of saline, type of nebulizer, interval of administration, treatment duration, and cointerventions; (5) outcomes: primary and secondary outcomes as described previously. For continuous outcomes, we extracted sample size, mean (median) and precision of measurements (SD, SE, 95% confidence interval [CI], or interquartile range) of each treatment arm. For dichotomous outcomes, we extracted number of events and total number of participants of each treatment arm. We contacted the principal investigators of 5 trials10,12,18,23,24 for methodological details and additional trial data, of whom 310,12,18 provided the requested data. We used Engauge digitizing software (digitizer.sourceforge.net) to extract the 25th and 75th percentiles of LOS in hospital from the figure of 1 paper.24 For 2 trials,24,25 we estimated mean and SD from median and interquartile range of LOS in hospital by using the method described by Wan et al.26 When the trial recruited multiple groups, we combined them into HS and NS groups.14,15,17,24,27

    Assessment of Risk of Bias

    Two reviewers (RM and LZ) independently assessed the risk of bias in included trials by examining the 6 key domains according to the recommendations of the Cochrane Collaboration.28 We graded each potential source of bias as yes, no, or unclear, relating to whether the potential for bias was low, high, or unknown. We resolved any disagreements between the 2 review authors by discussion and consensus.

    Data Synthesis and Statistical Analysis

    We performed meta-analysis for quantitative data synthesis whenever there were available data from the primary studies. For continuous outcomes, we used weighted mean difference (MD) between treatment groups and 95% CI as the metrics of effect size. Dichotomous data were synthesized by using risk ratios (RR) and 95% CIs as the effect measures. We used the random-effects model for meta-analyses.

    We assessed heterogeneity in results between studies by using the Cochrane Q test (P < .1 considered significant) and the I2 statistic. The I2 statistic ranges from 0% to 100% and measures the degree of inconsistency across studies, with values of 25%, 50%, and 75% corresponding to low, moderate, and high heterogeneity, respectively.29

    We conducted a priori subgroup analysis based on the treatment regimen. We also conducted post hoc subgroup analyses according to diagnosis criteria for bronchiolitis (presence of wheeze as essential diagnostic criteria and availability of virological testing) and risk of bias in the trials. We performed post hoc sensitivity analyses excluding open trials, trials in which mean and SD were estimated from median and interquartile range, trials with high risk of attrition bias (withdrawal rate >20% or data obtained from a part of study sample), and trials that did not use 0.9% saline as the control. All meta-analyses were performed by using Stata version 11.0 (Stata Corp, College Station, TX).

    Results

    Literature Search and Study Selection

    The search strategy identified 97 unique records from PubMed and 125 records from BIREME. After screening the titles and abstracts, we retrieved 26 potentially relevant full-text articles for further evaluation. Five articles were excluded for reasons shown in Fig 1. We obtained the data from clinical trials registry (ClinicalTrials.gov) to assess the eligibility of 3 completed but unpublished trials and all met the inclusion criteria. No additional trials were found by checking the reference lists of primary studies and review articles. Thus, a total of 24 trials4,919,2325,27,3037 involving 3209 patients were included in the review. All but 2 trials14,35 contributed data to the meta-analyses.

    FIGURE 1

    Flow diagram of study selection.

    Study Characteristics and Risk of Bias

    Table 1 summarizes the characteristics of the 24 included trials. All studies were parallel-group RCTs except 1 that was a quasi-RCT.17 The criteria for diagnosis of bronchiolitis were clearly defined by 19 trials. Eighteen trials1219,23,24,27,3137 defined bronchiolitis as the first episode of wheezing associated with viral respiratory infection in children <2 years of age. In 1 trial,4 bronchiolitis was defined as an apparent viral respiratory tract infection associated with airways obstruction manifest by hyperinflation, tachypnea, and subcostal recession with widespread crepitations on auscultation. Virological investigation was available in 13 trials4,914,16,18,19,24,30,32 and the positive rate for RSV varied from 56% to 88%. The concentration of HS was defined at 3% in all but 5 trials, in which 5%14,27,35 (n = 165), 6%24 (n = 83), and 7%32 saline (n = 52) was used. Treatment regimen of nebulized HS varied across studies, especially outpatient trials (Table 1).

    TABLE 1

    Characteristics of Included Trials

    All trials were double-blind except 3 open trials,4,34,37 in which performance bias and detection bias might occur (Supplemental Table 5). All trials but 117 were stated as randomized; however, 11 trials912,15,16,18,25,30,35,37 did not describe the methods for random sequence generation and/or allocation concealment. Attribution bias might occur in 3 trials25,32,37 because of high and unbalanced withdrawal rate after randomization.

    Efficacy of Nebulized HS in Inpatients

    LOS in Hospital

    Among 14 inpatient trials, 13912,16,18,19,2325,33,34,37 used LOS as the primary outcome and 127used LOS as the secondary outcome. One ED trial30 involving 408 patients provided the data of LOS among 145 hospitalized patients. We included the data of these 145 inpatients in the meta-analysis. The pooled results of 15 trials with a total of 1956 inpatients showed a statistically significant shorter mean LOS among infants treated with HS compared with those treated with 0.9% saline or standard care (MD of −0.45 days, 95% CI −0.82 to −0.08, P = .01) (Fig 2). There was significant heterogeneity in results between studies (I2 statistic = 82%). The data were suitable for conducting 5 subgroup analyses (Table 2). Nine trials4,1012,16,18,19,24,30 in which virological investigation was available showed significant effects of HS on reducing LOS, whereas 6 trials23,25,27,33,34,37 in which such testing was not available did not show significant benefits (P = .02 for subgroup comparison). The effect size of HS on LOS appeared to be greater in trials1012,16,18,25,30,37 with unclear or high risk of selection bias, compared with trials4,19,2325,27,33 with low risk of selection bias. However, the difference between subgroups was not statistically significant.

    FIGURE 2

    Effects of nebulized HS on reduction of LOS among inpatients.

    TABLE 2

    Subgroup Analyses on LOS (Inpatients) and Admission Rate (Outpatients)

    Four sensitivity analyses, excluding 2 trials24,25 with estimated mean and SD of LOS, 3 trials25,33,37 with high risk of attrition bias, 2 open trials,4,34 and 1 trial4 that did not use 0.9% saline as the control, did not significantly affect the results of the meta-analysis.

    Improvement in CSSs

    Eleven inpatient trials used bronchiolitis severity scores as outcome measure. Two trials12,34 used Respiratory Distress Assessment Instrument (RDAI)38 scores based on wheezing and retractions, but 112 did not report the results and the other34 reported RDAI scores only on day 1 of admission. One trial33 used a clinical score based on respiratory rate, wheezing, retractions, and oxygen saturation. This trial did not find a significant difference between HS and NS groups in clinical scores through day 1 to day 4 of admission. All the remaining 8 trials used Wang’s clinical scores,39 grading respiratory rate, wheezing, retractions, and general condition from 0 to 3. However, only 5 trials10,11,16,18,19 with a total of 404 patients provided suitable data for the meta-analysis, showing a significant effect of HS in improving clinical scores on day 1 (MD of −0.99, 95% CI −1.48 to −0.50, P < .0001, I2 statistic = 67%), day 2 (MD of −1.45, 95% CI −2.06 to −0.85, P < .0001, I2 statistic = 79%), and day 3 of admission (MD of −1.44, 95% CI −1.78 to −1.11, P < .0001, I2 statistic = 53%).

    Other Efficacy Outcomes

    Three trials24,33,37 used duration of in-hospital oxygen supplementation as efficacy outcome. Other efficacy outcomes used by at least 1 trial included duration of tube feeding, time for the resolution of respiratory symptoms and signs, radiograph scores, measurement of respiratory rate, heart rate and oxygen saturation, readmission within 28 days from randomization, and infant and parental quality-of-life questionnaire. Two trials18,19 reported a shorter duration of respiratory symptoms and signs (cough, wheezing, and crackles) in patients treated with HS compared with those receiving NS. None of the trials showed significant effects of HS on other previously mentioned outcomes.

    Efficacy of Nebulized HS in Outpatients

    Admission Rate

    Seven outpatient trials with a total of 951 patients assessed the efficacy of nebulized 3% saline on reducing the risk of hospitalization. The pooled RR was 0.80 (95% CI 0.67–0.96, P = .01) (Fig 3). There was no significant heterogeneity in results between studies (I2 statistic = 2%). The data were available for conducting 4 subgroup analyses (Table 2). The effect size of HS on the risk of hospitalization was significantly greater in trials9,13,30,32 in which virological investigation was available and in trials9,17,30 in which multiple doses (≥3) of saline solutions were administered, compared with trials15,17,31 in which virological testing was not available and trials13,15,31,32 by using only 1 to 2 doses of saline solutions, respectively. Four trials9,15,17,30 with unclear or high risk of selection bias showed significant effects of HS on reducing the risk of hospitalization, whereas 3 trials13,31,32 with low risk of selection bias did not show significant benefits of HS; however, the difference between subgroups was not statistically significant.

    FIGURE 3

    Effects of nebulized HS on reducing the risk of hospitalization among outpatients.

    Improvement in CSSs

    All 10 outpatient trials used bronchiolitis severity scores as the outcome measure. Variation in scoring methods and time points of assessment makes it inappropriate to conduct meta-analyses. Thus, we narratively summarized the main results of 9 trials in terms of effects of HS on improving clinical scores (Table 3). These trials did not show significant effects of nebulized HS in improving clinical scores, except 3 of the trials. One9 showed significant benefits of 3% saline compared with NS on each of 3 treatment days, the second14 showed consistent trend favoring 5% saline compared with 3% and 0.9% saline solutions from 8 to 72 hours after randomization, and the third34 showed the superiority of both 5% and 3% saline solutions over NS on each of 3 treatment days, but no significant difference was found between 5% and 3% saline groups.

    TABLE 3

    Narrative Summary of the Main Findings of 10 Outpatient Trials in Terms of Effects of HS on Improving Clinical Scores

    Rate of Readmission to Hospital or ED

    Five outpatient trials reported the rate of readmission to hospital and/or the ED 24 hours to 1 week after discharge. The meta-analysis did not show significant effects of HS in reducing the risk of readmission to hospital (4 trials1315,31 with 428 patients, RR of 1.45, 95% CI 0.67–3.14, P = .34, I2 statistic = 1%) and to ED (5 trials1315,31,36 with 523 patients, RR of 0.78, 95% CI 0.46–1.32, P = .36, I2 statistic = 29%).

    Other Efficacy Outcomes

    Oxygen saturation was used as an efficacy outcome by 4 trials.13,15,17,31 Other efficacy outcomes used by at least 1 trial included duration of oxygen supplementation, measurement of respiratory rate and heart rate, radiograph scores, and parental perception of improvement. None of the trials showed beneficial effects of HS on previously mentioned outcomes.

    Safety of Nebulized HS

    Of 24 trials included in this review, 21 reported safety data among 2897 participants, 1557 of whom received HS (3% saline: n = 1257; 5% saline: n = 165; 6% saline: n = 83; 7% saline: n = 52). Fourteen trials911,14,15,18,23,25,27,31,32,34,36,37 did not find any significant AEs among a total of 1548 participants, of whom 828 received nebulized HS (mixture with bronchodilators: n = 673, 81.3%; HS alone: n = 155, 18.7%). In the remaining 7 trials4,12,13,19,24,30,35 involving 1324 participants of whom 729 received nebulized HS (mixture with bronchodilators: n = 190, 26%; HS alone: n = 539, 74%), at least 1 AE was reported. Variation in reporting and in outcomes precluded the possibility of conducting meta-analysis of safety data. We narratively summarized the safety data of 7 trials (Table 4). Various AEs were reported in both HS and control groups. In most of cases, AEs were mild and resolved spontaneously. Only 1 inpatient trial4 involving 142 patients receiving 3% saline alone without bronchodilator reported 1 serious AE (bradycardia and desaturation) possibly related to HS inhalation but resolved the following day.

    TABLE 4

    Narrative Summary of AEs of Treatment Reported by 7 Trials

    Discussion

    This new systematic review and meta-analysis shows a modest but statistically significant benefit of nebulized 3% saline in reducing LOS in infants hospitalized for acute bronchiolitis. The review also shows that nebulized HS could reduce the risk of hospitalization by 20% compared with normal saline among outpatients with bronchiolitis.

    The results of this new review confirmed our hypothesis that nebulized HS may be less effective than previously claimed for infants with acute bronchiolitis. The effect size of nebulized HS on reducing LOS in hospitalized patients shown by the present review is only approximately one-third of that shown by the 2013 Cochrane review,20 which included 6 inpatient trials involving 500 patients (MD −1.15 days, 95% CI −1.49 to −0.82 days). It is interesting to note that all 8 trials4,2325,27,30,33,34 published in 2013 and thereafter, including 2 European multicenter studies4,24 with relatively large sample size, did not find significant effects of nebulized HS on LOS among inpatients with bronchiolitis. For outpatients, this new review showed a 20% reduction on the risk of hospitalization associated with nebulized HS in contrast with a 37% non–statistically significant reduction shown by the 2013 Cochrane review,20 which included 4 outpatient trials involving 380 participants (RR 0.63, 95% CI 0.37–1.07).

    We conducted subgroup analyses to explore potential effect modifiers and sources of heterogeneity in the results across studies. We found that trials in which virological investigation was available showed a significantly greater effect size of nebulized HS than trials without such testing in both inpatients and outpatients, measured by reduction of LOS and risk of hospitalization. These data suggest that diagnostic accuracy of bronchiolitis may affect the treatment outcomes with HS. The number and frequency of saline inhalations may also appear to influence the effect size of HS. Trials undertaken in an outpatient setting in which multiple doses (≥3) of saline solutions were administrated showed a significantly greater reduction on the risk of hospitalization compared with trials that used 1 to 2 doses of saline solutions. However, for inpatients, no significant difference was observed in reduction of LOS between trials that used more frequent saline inhalations (3 initial doses given every 1–2 hours, followed by every 4–6 hours) and those in which saline solutions were given every 6 to 8 hours. Another factor that could possibly influence the effect size of HS was risk of selection bias. Trials with unclear or high risk of selection bias showed significant effects of HS on reducing LOS and risk of hospitalization, whereas trials with low risk of selection bias did not show significant benefits of HS on these outcomes. This does cast some doubt on the overall effect estimates of HS; however, the difference between subgroups was not statistically significant. A tight seal between the mask and the infant’s face is crucial for an effective drug delivery with nebulizer.40 The performance of the nebulizer may also affect drug delivery.41 Thus, variability in drug delivery could be considered one of the potential sources of heterogeneity across studies; however, lack of data from primary studies did not allow us to include this important factor for subgroup analyses.

    Clinical score is generally considered a relatively objective measure to assess the severity of illness. Eleven inpatient trials used bronchiolitis severity scores as the efficacy outcome, but only 5 trials that used Wang’s clinical scores provided suitable data for meta-analysis. The pooled results of these 5 trials showed a significant effect of HS in improving clinical scores through day 1 to day 3 of admission. However, the inability to include another 6 inpatient trials in the meta-analysis may have affected the results of the analysis. Seven of 10 outpatient trials did not show significant effects of nebulized HS in improving clinical scores.

    Potential adverse effects of intervention with nebulized HS, such as acute bronchospasm, remain a potential concern. In this review, there were 14 trials involving 828 patients receiving nebulized HS that did not report any significant AEs. In 81.3% of these patients, saline solutions were mixed with bronchodilators. In contrast, there were 7 trials involving 729 patients treated with nebulized HS of which 74% received HS alone and reported at least 1 AE. Most AEs were mild and resolved spontaneously. These results suggest that nebulized HS is a safe treatment in infants with bronchiolitis, especially when administered in conjunction with a bronchodilator.

    This systematic review included trials conducted in both high-income and low-income countries and in different settings (inpatient, ambulatory care unit, and ED). Thus, evidence derived from the review may have a wide applicability. However, the quality of evidence could be graded only as moderate, mainly due to inconsistency in the results between studies and risk of bias in some trials, according to the Grading of Recommendations, Assessment, Development and Evaluations (GRADE) criteria.42 Moreover, all but 3 trials excluded patients requiring mechanical ventilation, intensive care, or having an oxygen saturation reading <85% on room air, so caution should be taken when extrapolating the findings of this review to infants with more severe bronchiolitis. The underlying airway pathologic changes may vary between infants with different severity of bronchiolitis, so different responses to treatments with HS may be expected in more severe cases. The results of meta-analysis for effects of HS on clinical scores among inpatients may be biased because only 5 of 11 trials measuring this outcome were included in the analysis. The number of trials and patients in outpatient settings is limited, and 1 trial30 with a relatively large sample size has contributed 43% of weight to the overall summary estimate of effects of HS on reduction of risk of hospitalization. All but 1 trial4 used NS as the comparison. The use of NS allows the trial to be double-blind; however, NS is not technically a placebo, as high-volume NS inhalation could potentially have physiologic effects by improving airway mucociliary clearance, which may have beneficial effects on acute bronchiolitis.8 Use of NS as the control may tend to minimize the effect size of HS.

    In conclusion, this new systematic review shows that nebulized HS is associated with a mean reduction of 0.45 days (∼11 hours) in LOS among infants admitted for acute bronchiolitis and a mean reduction of 20% in the risk of hospitalization among outpatients. This review also suggests that nebulized HS is a safe treatment in infants with bronchiolitis, especially when administered in conjunction with a bronchodilator. Given the high prevalence of bronchiolitis in infants and huge burden on health care systems throughout the world, benefits of nebulized HS shown by this review, even though smaller than previously estimated, may still be considered clinically relevant. Moreover, good safety profile and low cost make nebulized HS a potential attractive therapeutic modality for bronchiolitis in infants. However, further large multicenter trials are still warranted to confirm benefits of nebulized HS in both inpatients and outpatients with bronchiolitis, given the limited number of available trials, the small sample sizes of most previous trials, and conflicting results across studies. Further trials should use the most widely accepted clinical criteria and virological investigation for diagnosis of bronchiolitis. When LOS in hospital and admission rate are used as the primary efficacy outcomes, well-defined admission and discharge criteria should be used. Multiple doses of saline inhalations should be administered in outpatients; however, the optimal treatment regimen of nebulized HS for infants with bronchiolitis remains to be determined by further trials in both inpatients and outpatients.

    Footnotes

      • Accepted July 24, 2015.
    • Address correspondence to Linjie Zhang, MD, PhD, CNPq research fellow, Faculty of Medicine, Federal University of Rio Grande, Rua Visconde de Paranagua 102, Centro, Rio Grande-RS, Brazil 96200-190. E-mail: lzhang{at}furg.br; linjie.zhang{at}pq.cnpq.br
    • Dr Zhang conceptualized and designed the study, participated in trial selection, quality assessment, data collection and data analysis, and drafted the protocol and the review article; Dr Mendoza-Sassi provided input for study design, critically reviewed the protocol and the review article, and participated in trial selection, quality assessment, and data collection; Drs Klassen and Wainwright provided input for study design, and critically reviewed the protocol and review article; and all authors approved the final manuscript as submitted and agree to be accountable for all aspects of the work.

    • FINANCIAL DISCLOSURE: The authors have indicated they have no financial relationships relevant to this article to disclose.

    • FUNDING: No external funding.

    • POTENTIAL CONFLICT OF INTEREST: The authors have indicated they have no potential conflicts of interest to disclose.

    References