Cromolyn sodium for the prevention of chronic lung disease in preterm infants

Ng GY, Ohlsson A

Background - Methods - Results - Characteristics of Included Studies - References - Data Tables and Graphs

Dates

Date edited: 23/08/2006
Date of last substantive update: 07/11/2000
Date of last minor update: 30/06/2006
Date next stage expected 30/05/2008
Protocol first published: Issue 2, 2001
Review first published: Issue 2, 2001

Contact reviewer

Dr Geraldine Ng
Neonatal Registrar
Neonatal Intensive Care
St. George's Hospital
Neonatal Registrar Office, 1st Floor Lanesborough Wing
Blackshaw Road
Tooting
London UK
SW17 0QT
Telephone 1: 0208 672 1255
Facsimile: 0208 725 1933
E-mail: DrGYTNg@aol.com

Contribution of reviewers

G Y Ng:
Literature search and identification of trials for inclusion
Evaluation of methodologic quality of included trials
Abstraction of data
Verifying and entering data into RevMan
Writing text of review

A Ohlsson:
Literature search and identification of trials for inclusion
Evaluation of methodologic quality of included trials
Abstraction of data
Verifying and entering data into RevMan
Revision of the final review

Internal sources of support

Chelsea and Westminster Hospital, UK
Mount Sinai Hospital, Toronto, Ontario, CANADA

External sources of support

None

What's new

This review updates the existing review "Cromolyn sodium for the prevention of chronic lung disease in preterm infants)" initially published in The Cochrane Library, Issue 2, 2001.

In an updated search to April 2006 no new eligible studies were identified.

The conclusion remains unchanged: there is no evidence from randomised trials that cromolyn sodium is effective in preventing chronic lung disease in preterm infants.

Dates

Date review re-formatted: / /
Date new studies sought but none found: 30/06/2006
Date new studies found but not yet included/excluded: / /
Date new studies found and included/excluded: / /
Date reviewers' conclusions section amended: / /
Date comment/criticism added: / /
Date response to comment/criticisms added: / /

Text of review

Synopsis

Cromolyn sodium administered in the first few days of life has not been shown to prevent chronic lung disease in preterm infants

Preterm babies (babies born before 37 weeks gestation) often need to be given oxygen for lung problems for many weeks because of chronic lung disease. This is due, in part, to inflammation (swelling) within the lungs. Theoretically, cromolyn sodium is a drug that might help prevent this inflammation. It is relatively safe and adverse effects are rare. It can be given by nebuliser or aerosol inhaler in the first few days, in an attempt to prevent chronic lung disease. However, the review of trials found no strong evidence that cromolyn sodium can prevent or reduce chronic lung disease and further research does not seem to be justified.

Abstract

Background

Chronic lung disease (CLD) frequently occurs in preterm infants (< 37 weeks gestational age) and has a multifactorial etiology including inflammation. Cromolyn sodium is a mast cell stabiliser that inhibits neutrophil activation and neutrophil chemotaxis. Therefore, it is possible that cromolyn sodium might have a role in the prevention of CLD.

Objectives

To assess the effect of prophylactic administration of cromolyn sodium on the incidence of CLD, mortality or the combined outcome of mortality or CLD at 28 days of life, in preterm infants.

Search strategy

The search strategy of the Cochrane Neonatal Review Group was used to identify studies. Searches were made of MEDLINE, EMBASE, CINAHL up to and including April 2006, Cochrane Central Register of Controlled Trials (CENTRAL, The Cochrane Library, Issue 1, 2006), personal files and reference lists of identified trials.

Selection criteria

Randomised or quasi-randomised controlled clinical trials involving preterm infants. Initiation of cromolyn sodium administration during the first two weeks of life. The intervention had to include administration of cromolyn sodium by nebulizer or metered dose inhaler with or without spacer device, versus placebo or no intervention. Eligible studies had to include at least one of the following outcomes: overall mortality, CLD at 28 days, CLD at 36 weeks postmenstrual age, or the combined outcome mortality or CLD at 28 days. Secondary outcomes included number of days on oxygen, number of days on mechanical ventilation, patent ductus arteriosus (PDA), air leaks [pulmonary interstitial emphysema (PIE), pneumothorax], any grade of intraventricular haemorrhage (IVH), necrotizing enterocolitis (NEC), sepsis and adverse effects due to cromolyn sodium.

Data collection & analysis

The standard method for the Cochrane Collaboration as described in the Cochrane Collaboration handbook was used. Both investigators extracted and assessed all data for each study. Any disagreement was resolved by discussion. Relative risk (RR) and risk difference (RD) with 95% confidence intervals (CI) are reported for dichotomous outcomes and weighted mean difference (WMD) for continuous data. Number needed to treat was not calculated since no outcome showed a statistically significant RD. A fixed effect model was used for meta-analysis. Heterogeneity was examined using the I² statistic.

Main results

Two eligible studies were identified with small numbers of infants enrolled. Prophylaxis with cromolyn sodium did not result in a statistically significant effect on the combined outcome, death or CLD at 28 days [typical RR 1.05 (95% CI 0.73, 1.52); typical RD 0.03 (95% CI -0.20, 0.27)], CLD at 28 days [typical RR 0.93 (95% CI 0.53, 1.64; typical RD -0.03 (95% CI -0.27, 0.20)], CLD at 36 weeks postmenstrual age [RR 1.25 (95% CI 0.43, 3.63); RD 0.08 (95% CI -0.29, 0.44)], CLD in survivors at 28 days [typical RR 0.97 (95% CI 0.58, 1.63); typical RD -0.02 (95% CI -0.29, 0.26)] or CLD in survivors at 36 weeks postmenstrual age [RR 1.04 (95% CI 0.38, 2.87); RD 0.02 (95% CI -0.40, 0.43)]. Prophylaxis with cromolyn sodium did not show a statistically significant difference in overall neonatal mortality [typical RR 1.31 (95% CI 0.52, 3.29); typical RD 0.06 (95% CI -0.13, 0.26)]. There were no statistically significant differences in the incidence of air leaks, NEC, IVH, sepsis, days on mechanical ventilation or PDA. No side effects were noted.

Reviewers' conclusions

There is currently no evidence from randomized trials that cromolyn sodium has a role in the prevention of CLD. Cromolyn sodium cannot be recommended for the prevention of CLD in preterm infants. Additional clinical trials do not appear to be justified using the protocols for drug administration that are presently used, unless a more efficient type of delivery device than the jet nebulizer is employed.

Background

Chronic lung disease (CLD), defined as oxygen dependency at 28 days or at 36 weeks postmenstrual (GA) is a pulmonary disorder that frequently occurs in preterm infants (Northway 1967; Shennan 1988). It is the consequence of unresolved or abnormally repaired lung damage and has a multifactorial etiology that has been described extensively by previous authors. These factors include exposure to high oxygen concentration, volume derived trauma, barotrauma, sepsis and inflammation (Avery 1987; Paita 1991; Rojas 1995). In the past decades the survival rate of very low birth weight infants has increased and the prevalence of CLD remains high (Parker 1992). The incidence varies depending on the population studied, the diagnostic criteria used and variation between centres in clinical management (O'Brodovich 1985; Shennan 1988; Hack 1991; Avery 1987; Lee 2000). CLD may be associated with chronic respiratory difficulties, prolonged and recurrent hospitalisation, growth restriction and death (O'Brodovich 1985; Lee 2000). The administration of antenatal corticosteroids to mothers likely to give birth preterm reduces neonatal mortality and the incidence of respiratory distress syndrome (RDS), but not the incidence of CLD (Crowley 2000). Administration of prophylactic natural surfactant extract does not reduce the incidence of CLD but does reduce the combined outcome of death or CLD (Soll 2000).

Neonates who develop CLD have been demonstrated to have elevated levels of endothelin-1, interleukin (IL)-6, IL-8, and high ratios of IL-1B to IL-1 receptor antagonists in tracheobronchial lavage when compared to those with respiratory distress syndrome (RDS) or controls (Bagchi 1994; Tullus 1996; Niu 1998; Rindfleisch 1996). One study showed elevated levels of tumour necrosis factor (TNF) alpha in neonates who subsequently develop CLD (Tullus 1996), but another study did not (Bagchi 1994).

Cromolyn sodium was originally characterised as a mast cell stabiliser (Hoag 1991). However, it also inhibits neutrophil activation (Kay 1987) and neutrophil chemotaxis (Bruijnzeel 1989) and, therefore, may modulate the inflammatory process in the lung. In a small non-randomised cohort comparison study of cromolyn sodium use in infants with CLD, Viscardi 1994 found that there was a decrease in peak inspiratory pressure, a reduction in intermittent mandatory ventilation requirements and an increase in dynamic compliance after two weeks of cromolyn sodium therapy. Total airway resistance did not change significantly. Compared to matched controls, there were no differences in duration of need for supplemental oxygen, ventilation or hospitalisation. These results suggested that cromolyn therapy might improve pulmonary function in some infants with established CLD (Viscardi 1994). Initiating cromolyn sodium treatment in the first few days of life in infants at risk of CLD might inhibit the neutrophil influx and the release of inflammatory mediators in the lung and thereby decrease the lung injury leading to CLD.

Cromolyn sodium may be delivered by nebuliser or pressurised aerosol with or without a spacer device. With either method, only 0.22 - 1.3% of the dose will reach the lung (Grigg 1992; Fok 1996). Aerosolized products tend to be deposited in the central lung region rather than in the periphery (Fok 1996). Humidification of the gas reduces lower respiratory tract deposition of aerosolized products (Diot 1995). Addition of a spacer device between the nebuliser and the endotracheal tube (Harvey 1995) and synchronising nebulisation with inspiratory airflow (Diot 1995) increase deposition. There is considerable intersubject variability in lung deposition (Fok 1996). All these factors will modify therapeutic effects.

Cromolyn is thought to be a relatively safe drug. However, there are occasional reports in the literature of adverse effects. These include dermatitis (Sheffer 1975; Settipane 1979), urticaria (Sheffer 1975; Menon 1977), eosinophilic pulmonary infiltrates (Lobel 1972; Sheffer 1975), pulmonary allergic granulomatosis (Burgher 1974; Sheffer 1975) and anaphylaxis (Ahmad 1983; Brown 1981).

Objectives

To assess the effect of prophylactic administration of cromolyn sodium on the incidence of CLD at 28 days or 36 weeks postmenstrual age, mortality, or the combined outcome of mortality or CLD at 28 days in preterm infants.

Criteria for considering studies for this review

Types of studies

Randomised or quasi-randomised controlled clinical trials.

Types of participants

Preterm infants (< 37 weeks GA).

Types of interventions

The intervention had to include administration of cromolyn sodium by nebulizer or metered dose inhaler with or without spacer device, versus placebo or no intervention. Treatment with cromolyn sodium had to be initiated during the first two weeks of life.

Types of outcome measures

Primary outcomes were mortality within the study period, combined outcome of CLD or mortality at 28 days, CLD at 28 days and 36 weeks corrected GA. Secondary outcomes were number of days on oxygen, number of days on ventilator, patent ductus arteriosus (PDA), pulmonary interstitial emphysema (PIE), pneumothorax, any grade of intraventricular haemorrhage, necrotizing enterocolitis (NEC), sepsis and adverse effects of cromolyn sodium.

Search strategy for identification of studies

The search strategy used to identify studies was performed according to the guidelines of the Cochrane Neonatal Review Group. Searches were made of MEDLINE 1966 to April 2006, EMBASE 1980 to April 2006, CINAHL 1982 to April 2006, Cochrane Central Register of Controlled Trials (CENTRAL, The Cochrane Library, Issue 1, 2006), personal files and reference lists of identified trials. The search strategies were adjusted to reflect the different controlled vocabularies of the databases.

In MEDLINE the following MeSH terms were used: bronchopulmonary dysplasia, chronic disease, cromolyn sodium. The following textwords were used: chronic lung disease, cromoglycate. Limits: newborn infant < birth to 1 month >, clinical trial, controlled clinical trial, meta analysis, multicenter study, randomised controlled trial.

In EMBASE the following terms were used: cromoglycate disodium, lung dysplasia, chronic lung disease, evidence based medicine, clinical trial, multicenter study. The following textwords were used: cromoglycate, cromoglicate, cromolyn, bronchopulmonary dysplasia, clinical trial, rct, rcts, random, meta-analysis, metaanalysis or multicent:, newborn, neonate, neonatal. Limits: infant < to one year >.

In CINAHL the following terms were used: bronchopulmonary dysplasia, chronic disease, lung diseases, cromolyn sodium. The following textwords were used: cromoglycate, cromoglicate. Limits: newborn infant < birth to 1 month >.

Methods of the review

Each review author (GN, AO) selected studies and extracted and assessed data separately. The data were then compared and differences were resolved by discussion. Information on the trial participants included birth weight, GA at birth, need for mechanical ventilation, and sex. Information on clinical outcomes included CLD at 28 days, CLD at 36 weeks corrected GA, combined outcome of CLD or mortality at 28 days, mortality at 28 days, IVH, NEC, air leaks, sepsis and adverse effects due to cromolyn sodium. Information on hospital stay and days in oxygen or on mechanical ventilation was also sought.

The methodological quality of each trial was independently reviewed by two review authors (GN, AO) for the following questions:
1. Blinding of randomization
2. Blinding of intervention
3. Complete follow up
4. Blinding of outcome.
There were three responses to these questions: yes, no or cannot tell. For each study, information was sought regarding stratification by birth weight or gestational age and whether the trial was single or multi-centred.

The standard methods of the Neonatal Review Group were used to synthesize data using relative risk (RR), risk difference (RD) and number needed to treat (NNT) if there was a statistically significant reduction in RD and number needed to harm (NNH) if there was a statistically significant increase in the RD. Meta-analyses of data from the included trials were performed using RevMan 4.2, using a fixed effect model. Heterogeneity was examined using the I² statistic.

Description of studies

Two eligible studies, both single centre, were identified. No eligible studies were excluded from analysis.

In the trial by Watterberg 1993, 38 infants needing ventilatory support at birth were randomised to receiving sodium cromoglycate or saline. The preterm infants had a mean GA [standard deviation (SD)] of 25.9 wks +/- 0.9 and a mean (SD) birth weight of 845 g +/- 81 in the treatment group and a mean GA (SD) of 26 wks +/- 1.5 and a mean (SD) birth weight of 851 g +/- 150 in the control group. Patients were stratified into those below 1000 g and those from 1000 to 2000 g. Watterberg 1993 states that allocation to treatment groups was by sequential numbers, but it is unclear exactly what this involved and, therefore, whether blinding occurred is unclear. The patients were comparable in severity of illness on entry into the study (defined as FiO2 and respiratory acuity score). Infants enrolled into the study less than 12 hrs after intubation for RDS. Sodium cromoglycate or placebo (normal saline) was given as 20 mg nebulized every six hours for the duration of intubation. The length of time that the infants were intubated and, therefore, received treatment is lacking. The study was discontinued when 10 babies of birth weight under 1000 g and 28 babies from 1000 to 2000 g had been enrolled. Although we had not specified subgroup analyses by birth weight, data from this study could be obtained only from babies > 1000 g for some outcomes.

In the trial by Viscardi 1997, 26 infants requiring ventilation on Day 1 for RDS with a greater than 75% predicted probability of oxygen dependence at 28 days were randomised to receive either nebulized sodium cromoglycate given as 20 mg every six hours or placebo (normal saline). The 12 hour CLD predictive score was similar in the two groups. There was no difference between groups in FiO2, alveolar-arterial oxygen gradient, or ventilatory settings on the day of study entry. The infants received the drug from day 3 to day 28 of life. One infant was withdrawn at day 7 at parental request but is included in the analyses of outcomes. Seventy-seven percent of treated infants and 100% of infants in the placebo group received exogenous surfactant. The primary outcome was the concentration of cytokines in lung lavage in these infants.

No data were collected regarding the amount and site of drug deposition in the lung in either trial. The numbers of patients receiving postnatal dexamethasone in each group was not documented in either trial.

Methodological quality of included studies

Both studies were of small size, but of good quality overall. A sample size calculation was included in the study by Watterberg 1993, but not in the trial by Viscardi 1997. Viscardi 1997 states that the randomisation process and the allocation to groups was blinded. In the study by Watterberg 1993 blinding of randomisation was unclear. Patients were enrolled by sequential numbers in the study by Watterberg 1993, but it is unclear as to what this actually means. In the study by Viscardi 1997 randomisation was accomplished by a computer generated table of random numbers. The intervention was blinded in both studies, although in the study by Viscardi 1997 the pharmacist was aware of treatment assignment. Both Viscardi 1997 and Watterberg 1993 analysed the data on an intention to treat basis and there was complete follow up of all patients. In the study by Viscardi 1997 the clinical outcome was CLD at 28 days of life. Chest X rays were reviewed by a single radiologist who was blinded to the treatment regimens and the outcomes.

Results

Comparison 01: SODIUM CROMOGLYCATE VS PLACEBO
Outcome 01.01: Mortality or CLD at 28/30 days
Both Viscardi 1997 and Watterberg 1993 assessed this outcome. Neither found a significant effect. The meta-analysis of the results of both trials found no evidence of effect: typical RR 1.05 (95% CI 0.73, 1.52); typical RD 0.03 (95% CI -0.20, 0.27). These results are based on 2 trials, 64 babies and 41 events.

Outcome 01.02: Mortality to 28 days
Both Viscardi 1997 and Watterberg 1993 assessed this outcome. Neither found a significant effect. The meta-analysis of the results of both trials found no evidence of effect: typical RR 1.31 (95% CI 0.52, 3.29); typical RD 0.06 (95% CI -0.13, 0.26). These results are based on 2 trials, 64 babies and 14 events.

Outcome 01.03: CLD at 28 days
Both Viscardi 1997 and Watterberg 1993 assessed this outcome. Neither found a significant effect. The meta-analysis of the results of both trials found no evidence of effect: typical RR 0.93 (0.53, 1.64); typical RD -0.03 (95% CI -0.27, 0.20). These results are based on 2 trials, 64 babies and 27 events.

Outcome 01.04: CLD at 36 weeks corrected gestational age
The study by Watterberg 1993 did not look at this outcome. The study by Viscardi 1997 found no evidence of effect: RR 1.25 (0.43, 3.63), RD 0.08 (-0.29, 0.44). These results are based on one trial, 26 babies and 9 events.

Outcome 01.05: CLD among survivors at 28 days
Both Viscardi 1997 and Watterberg 1993 assessed this outcome. Neither found a significant effect. The meta-analysis of the results of both trials found no evidence of effect: typical RR 0.97 (95% CI 0.58, 1.63, typical RD -0.02 (95% CI -0.29, 0.26). These results are based on 2 trials, 50 surviving babies and 27 events.

Outcome 01.06: CLD among survivors at 36 weeks corrected gestational age
The study by Watterberg 1993 did not look at this outcome. The paper by Viscardi 1997 found no evidence of effect: RR 1.04 (95% CI 0.38, 2.87); RD 0.02 (95% CI -0.40, 0.43). These results are based on one trial, 22 surviving babies and 9 events.

Outcome 01.07: IVH
Both Viscardi 1997 and Watterberg 1993 assessed this outcome. Neither found a significant effect. The meta-analysis of the results of both trials found no evidence of effect: typical RR 0.89 (95% CI 0.63, 1.27); typical RD -0.06 (95% CI -0.26, 0.13). These results are based on two trials, 64 babies and 36 events.

Outcome 01.08: NEC
Both Viscardi 1997 and Watterberg 1993 assessed this outcome. Neither found a significant effect. The meta-analysis of the results of both trials found no evidence of effect: typical RR 1.25 (95% CI 0.37, 4.24); typical RD 0.03 (95% CI -0.14, 0.20). These results are based on two trials, 64 babies and 9 events.

Outcome 01.09: PDA
Both Viscardi 1997 and Watterberg 1993 assessed this outcome. Neither found a significant effect. The meta-analysis of the results of both trials found no evidence of effect: typical RR 0.88 (95% CI 0.65, 1.18); typical RD -0.09 (95% CI -0.31, 0.12). These results are based on two trials, 64 babies and 47 events.

Outcome 01.10: Air leak (PIE, pneumothorax)
Both Viscardi 1997 and Watterberg 1993 assessed this outcome. Neither found a significant effect. The meta-analysis of the results of both trials found no evidence of effect: typical RR 0.67 (95% CI 0.32, 1.39); typical RD -0.13 (95% CI -0.34, 0.09). These results are based on two trials, 64 babies and 20 events.

Outcome 01.11: Days on mechanical ventilation
This is only commented on in the study by Watterberg 1993, who did not report days on mechanical ventilation for babies under 1000 g; therefore, this outcome was analyzed only for babies over 1000 g. There was no evidence of effect: mean difference 1.00 days (95% CI -6.41, 8.41). This result is based on one trial and 28 babies.

Outcome 01.12: Sepsis
Both Viscardi 1997 and Watterberg 1993 assessed this outcome. Neither found a significant effect. The meta-analysis of the results of both trials found no evidence of effect: typical RR 0.82 (95% CI 0.41, 1.63); typical RD -0.06 (95% CI -0.28, 0.15). These results are based on two trials, 64 babies and 20 events.

Other effects: cytokines in lung lavage
Viscardi 1997 showed that although the pre study (day 3) cytokine concentrations of cromolyn treated and control groups were similar, IL-8 concentration in cromolyn sodium lavage decreased 2.45 fold from day 3 to 7, whereas IL-8 concentrations increased 1.7 fold in control lavage during the same interval. Interleukin 8 (IL-8) concentrations were 4.9 fold lower on day 7 in lavage from cromolyn sodium treated infants (median 2245 pg/ml) compared to levels in controls [median 11009 pg/ml (p = 0.051)]. TNF alpha was 3.6-fold lower on day 7 in lavage from the cromolyn sodium treated infants (median 19.95 pg/ml) compared with lung lavage from controls [median 70.9 pg/ml (p=0.04)]. There was a trend toward lower lavage levels of interleukin 1B (IL-1B) on day 7 in cromolyn treated infants compared to control lavage levels [control median 208.2 pg/ml; cromolyn sodium median 46.75 pg/ml (p = 0.13)] but interleukin 1 receptor antagonist (IL-1ra) concentrations were unaffected by cromolyn sodium treatment (day 7 cromolyn sodium median 21.5 ng/ml, day 7 control median 23 ng/ml). The IL-1B/IL-1ra ratio was lower in day 7 lavage from cromolyn sodium treated infants compared to the ratio in control infant lavage [cromolyn sodium median 0.002, control median 0.005 (p = 0.09)]. It is difficult to interpret these results as there were no clinical benefits noted.

Adverse effects
Sodium cromoglycate seemed to be a well tolerated drug without adverse side effects and there were no reported adverse responses to nebulization.

Discussion

Although the study by Viscardi 1997 showed a reduction in cytokines in lung lavage from infants treated with cromolyn sodium, this did not translate into clinical benefits. Neither of the two eligible studies found a reduction in CLD or mortality. Cromolyn sodium had no statistically significant effect on the outcomes of IVH, NEC, PDA, air leak, days on mechanical ventilation or sepsis. No adverse effects were noted.

Sodium cromoglycate was given for different duration in the two studies. In the Watterberg 1993 study, sodium cromoglycate was given for the duration of intubation. In the Viscardi 1997 study, it was given for 25 days. The age of entry to the studies also differed. In the Watterberg 1993 study, the age of entry was 12 hrs after intubation and in the Viscardi 1997 study it was on day 3 of life.

Both studies involved very small numbers of infants. The sample sizes are so small that no reliable estimate of treatment effect is provided. The wide CIs show that clinically important effect sizes on major outcomes have not been excluded. Neither of these trials assessed drug deposition and it is possible that the drug wasn't received in effective doses at the correct sites. However, the available evidence does not justify further trials using the protocols for cromolyn sodium administration used to date.

Reviewers' conclusions

Implications for practice

Cromolyn sodium cannot be recommended for the prevention of CLD in preterm infants.

Implications for research

Although results are available only from a small number of infants randomised, the available evidence does not appear to justify further trials using the protocols for drug administration reviewed here, unless a more efficient type of delivery device than the jet nebulizer is employed.

Acknowledgements

We would like to acknowledge the help of Elizabeth Uleryk, Director of the Hospital Library at the Hospital for Sick Children for help with search strategies.

Potential conflict of interest

None

Characteristics of included studies

Study	Methods	Participants	Interventions	Outcomes	Notes	Allocation concealment
Viscardi 1997	Randomised prospective double-blind placebo controlled trial. 1. Blinding of randomisation- Yes 2. Blinding of intervention - Yes 3. Complete follow up - Yes 4. Blinding of outcome - Yes, CXRs were reviewed by a single observer blinded to the treatment regimes	Number of patients entered into the study - 26 Mean BW (SD) 687 +/- 46g in the treatment group and 702 +/- 35g in the control group Mean GA (SD) 24.6 +/- 0.4wk in the treatment group and 25.1 +/- 0.4wk in the control group. Age of enrollment into study - Day 3 Other characteristics - high probability of oxygen dependence at 28 days predicted by a CLD score at 12 hrs of age Exclusion criteria - documented sepsis, congenital cardiopulmonary anomalies	Nebulised 20mg cromolyn sodium (n = 13) or 2mls normal saline placebo (n = 13) every 6 hrs from day 3 until day 28	Primary outcome was CLD at 28 days of life and at 36 wks corrected gestational age. Changes in inflammatory cell populations and the cytokines in lung lavage in response to cromolyn therapy. Secondary outcomes were number of ventilator days, duration of oxygen supplementation, PDA, air leak, sepsis, NEC, IVH	100% of the control group and 77% of treated infants received exogenous surfactant. No adverse effects which could be attributed to the administration of cromolyn sodium were noted.	A
Watterberg 1993	Randomised prospective double-blind placebo controlled trial. 1. Blinding of randomisation - can't tell. Patients were stratified into those below 1000g and those from 1000 to 2000g and enrolled by "sequential numbers". 2. Blinding of intervention - Yes 3. Complete follow up - Yes 4. Blinding of outcome - can't tell	Number of patients entered into the study - 38 Mean BW (SD): in the < 1000g arm in the treatment group 845 +/- 81g and in the placebo group 851 +/- 150g; in the > 1000g arm 1353 +/- 184g in the treatment group and 1331 +/- 227g in the control group. Mean GA (SD): in the < 1000g arm 25.9 +/- 0.9wk in the treatment group and 26.2 +/- 1.5wk in the control group; in the > 1000g arm group 29.9 +/- 1.8wk in the treatment group and 30.8 +/- 2wk in the control group. Age of enrollment into study - less than 12 hours after intubation	Aerosolised 20mg cromolyn sodium (n = 19) or 2mls normal saline placebo (n = 19) every 6 hrs whilst intubated	Primary outcome was survival without oxygen dependence (oxygen saturations > 90% in FiO2 < 25% at 30 days of life). Secondary outcomes were the number of days dependent on mechanical ventilation during the first 30 days of life, the intensity of mechanical ventilation, and pulmonary complications	No adverse effects which could be attributed to the administration of cromolyn sodium were noted. Days on mechanical ventilation were not reported on for babies < 1000g. Therefore this outcome was analysed in the review only for babies > 1000g. We have written to Watterberg et al regarding the allocation concealment and this information will be available for a future update.	B

Abbreviations:
CLD = chronic lung disease
g = grams
GA = gestational age
BW = birthweight
IVH = intraventricular haemorrhage
NEC = necrotising enterocolitis
PDA = patent ductus arteriosus
wk = weeks

References to studies

References to included studies

Viscardi 1997 {published data only}

Viscardi RM, Hasday JD, Gumpper KF, Taciak V, Campbell AB, Palmer TW. Cromolyn sodium prophylaxis inhibits pulmonary proinflammatory cytokines in infants at high risk for bronchopulmonary dysplasia. American Journal of Respiratory and Critical Care Medicine 1997;156:1523-9.

Watterberg 1993 {published data only}

Watterberg KL, Murphy S. Failure of cromolyn sodium to reduce the incidence of bronchopulmonary dysplasia: a pilot study. Pediatrics 1993;91:803-6.

* indicates the primary reference for the study

Other references

Additional references

Ahmad 1983

Ahmad S. Cromolyn sodium and anaphylaxis. Annals of Internal Medicine 1983;99:882-3.

Avery 1987

Avery ME, Tooley WH, Keller JB, Hurd S, Bryan MH, Cotton RB, Epstein MF, Fitzhardinge PM, Hansen CB, Hansen TN, Hodson WA, James LS, Kitterman JA, Nielsen HC, Poirier TA, Truog WE, Wung JT. Is chronic lung disease in low birth weight infants preventable? A survey of eight centers. Pediatrics 1987;79:26-30.

Bagchi 1994

Bagchi A, Viscardi RM, Taciak V, Ensor JE, McCrea KA, Hasday JD. Increased activity of interleukin-6 but not tumor necrosis factor alpha in lung lavage of premature infants is associated with the development of bronchopulmonary dysplasia. Pediatric Research 1994;36:244-51.

Brown 1981

Brown LA, Kaplan RA, Benjamin PA, Hoffman LS, Shearer WT. Immunoglobulin E-mediated anaphylaxis with inhaled cromolyn sodium. Journal of Allergy and Clinical Immunology 1981;68:416-20.

Bruijnzeel 1989

Bruijnzeel PLB, Warringa RAJ, Kok PTM. Inhibition of platelet-activating factor and zymosan-activated serum-induced chemotaxis of human neutrophils by nedocromil sodium, BN 52021 and sodium cromoglycate. British Journal of Pharmacology 1989;97:1251-7.

Burgher 1974

Burgher LW, Kass I, Schenken JR. Pulmonary allergic granulomatosis: a possible drug reaction in a patient receiving cromolyn sodium. Chest 1974;66:84-6.

Crowley 2000

Crowley P. Prophylactic corticosteroids for preterm birth (Cochrane Review). In: Cochrane Library, Issue 4, 2000. Oxford: Update Software.

Diot 1995

Diot P, Morra L, Smaldone GC. Albuterol delivery in a model of mechanical ventilation. Comparison of metered-dose inhaler and nebulizer efficiency. American Journal of Respiratory and Critical Care Medicine 1995;152:1391-4.

Fok 1996

Fok TF, Monkman S, Dolovich M, Gray S, Coates G, Paes B, Rashid F, Newhouse M, Kirpalani H. Efficiency of aerosol medication delivery from a metered dose inhaler versus jet nebulizer in infants with bronchopulmonary dysplasia. Pediatric Pulmonology 1996;21:301-9.

Grigg 1992

Grigg J, Arnon S, Jones T, Clarke A, Silverman M. Delivery of therapeutic aerosols to intubated babies. Archives of Disease in Childhood 1992;67:25-9.

Hack 1991

Hack M, Horbar JD, Malloy MH, Tyson JE, Wright E, Wright L. Very low birth weight outcomes of the National Institute of Child Health and Human Development Neonatal Network. Pediatrics 1991;87:587-96.

Harvey 1995

Harvey CJ, O'Doherty MJ, Page CJ, Thomas SHL, Nunan TO, Treacher DF. Effect of a spacer on pulmonary aerosol deposition from a jet nebuliser during mechanical ventilation. Thorax 1995;50:50-3.

Hoag 1991

Hoag JE, McFadden ER. Long term effect of cromolyn sodium on non-specific bronchial hyperresponsiveness: a review. Annals of Allergy 1991;66:53-63.

Kay 1987

Kay AB, Walsh GM, Moqbel R, MacDonald AJ, Nagakura T, Carroll MP, Richerson HB. Disodium cromoglycate inhibits activation of human inflammatory cells in vitro. Journal of Allergy and Clinical Immunology 1987;80:1-8.

Lee 2000

Lee SK, McMillan DD, Ohlsson A, Pendray M, Synnes A, Whyte R, Chien L-Y, Sale J. Variations in practice and outcomes in the Canadian NICU Network 1996-1997. Pediatrics 2000;106:1070-9.

Lobel 1972

Lobel H, Machtey I, Eldror MY. Pulmonary infiltrates with eosinophilia in an asthmatic patient treated with disodium cromoglycate. Lancet 1972;2:1032.

Menon 1977

Menon MPS, Das AK. Asthma and urticaria during disodium cromoglycate treatment. Scandinavian Journal of Respiratory Diseases 1977;58:145-50.

Niu 1998

Niu JO, Munshi UK, Siddiq MM, Parton LA. Early increase in endothelin-1 in tracheal aspirates of preterm infants: correlation with bronchopulmonary dysplasia. Journal of Pediatrics 1998;132:965-9.

Northway 1967

Northway WH , Rosan RC, Porter DW. Pulmonary disease following respirator therapy of hyaline-membrane disease. New England Journal of Medicine 1967;276:357-68.

O'Brodovich 1985

O'Brodovich HM, Mellins RB. Bronchopulmonary dysplasia - unresolved neonatal acute lung injury. American Review of Respiratory Disease 1985;132:694-709.

Paita 1991

Paita M, Gabbert D, Weinstein MR, Peters ME. Multivariate assessment of traditional risk factors for chronic lung disease in very low birth weight neonates. Journal of Pediatrics 1991;119:285-92.

Parker 1992

Parker RA, Lindstrom DP, Cotton RB. Improved survival accounts for most but not all of the increase in bronchopulmonary dysplasia. Pediatrics 1992;90:663-8.

Rindfleisch 1996

Rindfleisch MS, Hasday JD, Taciak V, Broderick K, Viscardi RM. Potential role of interleukin-1 in the development of bronchopulmonary dysplasia. Journal of Interferon and Cytokine Research 1996;16:365-73.

Rojas 1995

Rojas MA, Gonzales A, Bancalari E, Claure N, Poole C, Siva-Nieto G. Changing trends in the epidemiology and pathogenesis of neonatal chronic lung disease. Journal of Pediatrics 1995;126:605-10.

Settipane 1979

Settipane GA, Klein DE, Boyd GK, Sturam JH, Freye HB, Weltman JK. Adverse effects to cromolyn. JAMA 1979;241:811-3.

Sheffer 1975

Sheffer AL, Rocklin RE, Goetzi EJ. Immunologic components of hypersensitivity reactions to cromolyn sodium. New England Journal of Medicine 1975;295:1220-4.

Shennan 1988

Shennan AT, Dunn MS, Ohlsson A, Lennox K, Hoskins EM. Abnormal pulmonary outcomes in premature infants: prediction from oxygen requirement in the neonatal period. Pediatrics 1988;82:527-32.

Soll 2000

Soll RF. Prophylactic natural surfactant for preventing morbidity and mortality in preterm infants (Cochrane Review). In: Cochrane Library, Issue 4, 2000. Oxford: Update Software.

Tullus 1996

Tullus K, Noack GW, Burman LG, Nilsson R, Wretlind B, Brauner A. Elevated cytokine levels in tracheobronchial aspirate fluids from ventilator treated neonates with bronchopulmonary dysplasia. European Journal of Pediatrics 1996;155:112-6.

Viscardi 1994

Viscardi RM, Adeniyi-Jones SC. Retrospective study of the effectiveness of cromolyn sodium in bronchopulmonary dysplasia. Neonatal Intensive Care 1994;7:18-20.

Other published versions of this review

Ng 2001

Ng GY, Ohlsson A. Cromolyn sodium for the prevention of chronic lung disease in preterm infants. In: The Cochrane Database of Systematic Reviews, Issue 2, 2001.

Ng 2002

Ng GY, Ohlsson A. Cromolyn sodium for the prevention of chronic lung disease in preterm infants. In: The Cochrane Database of Systematic Reviews, Issue 4, 2002.

Comparisons and data

Comparison or outcome	Studies	Participants	Statistical method	Effect size
01 Sodium cromoglycate vs placebo
01 Mortality or CLD at 28/30 days	2	64	RR (fixed), 95% CI	1.05 [0.73, 1.52]
02 Mortality to 28 days	2	64	RR (fixed), 95% CI	1.31 [0.52, 3.29]
03 CLD at 28 days	2	64	RR (fixed), 95% CI	0.93 [0.53, 1.64]
04 CLD at 36 weeks GA	1	26	RR (fixed), 95% CI	1.25 [0.43, 3.63]
05 CLD among survivors to 28 days	2	50	RR (fixed), 95% CI	0.97 [0.58, 1.63]
06 CLD among survivors to 36 weeks GA	1	22	RR (fixed), 95% CI	1.04 [0.38, 2.87]
07 IVH	2	64	RR (fixed), 95% CI	0.89 [0.63, 1.27]
08 NEC	2	64	RR (fixed), 95% CI	1.25 [0.37, 4.24]
09 PDA	2	64	RR (fixed), 95% CI	0.88 [0.65, 1.18]
10 Air leak (PIE, pneumothorax)	2	64	RR (fixed), 95% CI	0.67 [0.32, 1.39]
11 Days on mechanical ventilation	1	28	WMD (fixed), 95% CI	1.00 [-6.41, 8.41]
12 Sepsis	2	64	RR (fixed), 95% CI	0.82 [0.41, 1.63]

01 Sodium cromoglycate vs placebo

01.01 Mortality or CLD at 28/30 days

01.02 Mortality to 28 days

01.03 CLD at 28 days

01.04 CLD at 36 weeks GA

01.05 CLD among survivors to 28 days

01.06 CLD among survivors to 36 weeks GA

01.07 IVH

01.08 NEC

01.09 PDA

01.10 Air leak (PIE, pneumothorax)

01.11 Days on mechanical ventilation

01.12 Sepsis

Contact details for co-reviewers

Dr Arne Ohlsson
Director Evidence Based Neonatal Care and Outcomes Research
Department of Paediatrics
Mount Sinai Hospital
600 University Avenue
Toronto
Ontario CANADA
M5G 1X5
Telephone 1: +1 416 586 8379
Telephone 2: +1 416 341 0444
Facsimile: +1 416 586 8745
E-mail: aohlsson@mtsinai.on.ca

This review is published as a Cochrane review in The Cochrane Library, Issue 4, 2006 (see http://www.thecochranelibrary.com for information). Cochrane reviews are regularly updated as new evidence emerges and in response to feedback. The Cochrane Library should be consulted for the most recent version of the review.