Alpha-1 proteinase inhibitor (a1PI) for preventing chronic lung disease in preterm infants

Cover sheet

Title

Reviewers

Dates

Contact reviewer

Contribution of reviewers

Dr. A Ohlsson
Literature search and identification of trials
Evaluation of methodological quality of trials
Data collection
Verification of data
Revision of the review

Internal sources of support

External sources of support

What's new

The updated review has no further information as no new studies were identified.

Dates

Text of review

Synopsis

There is not enough evidence to show the long term effect of using Alpha-1 proteinase inhibitor for chronic lung disease in premature babies.

Inflammation of the lungs is one of the causes of chronic lung disease (CLD) in babies born before 37 weeks. Babies with CLD need extra oxygen and the disease can also cause serious long-term problems. Lung damage is caused by the release of enzymes and other anti-oxidants because babies with CLD have a low level of Alpha-1 proteinase inhibitor (a1P1), a substance that stops lung tissue being destroyed. A medication version of AlP1 is sometimes given to protect their lungs. The review of the trials found that there is not enough evidence to show long term beneficial effects of a1P1. More research is needed.

Abstract

Background

Inflammation of the pulmonary parenchyma is one of the important mechanisms implicated in development of chronic lung disease (CLD) in preterm neonates. Release of enzymes and other anti - oxidants following cell damage is considered to be responsible for the damage to lung tissue. Various strategies have been attempted to counteract enzymatic damage to pulmonary parenchyma and to prevent CLD.

Objectives

This review examines the effectiveness of alpha 1 proteinase inhibitor (a1PI) for the prevention of CLD defined as requirement of supplemental oxygen at 36 weeks post menstrual age (PMA) in preterm neonates.

Search strategy

A literature search was performed using the following databases: MEDLINE (1966 - February 2005), EMBASE (1980 -February 2005), CINAHL (1982 - February 2005), Cochrane Central Register of Controlled Trials (CENTRAL, The Cochrane Library, Issue 1, 2005) and abstracts from the annual meetings of the Society of Pediatric Research, American Pediatric Society and Pediatric Academic Societies published in Pediatric Research (1991-2004). No language restrictions were applied.

Selection criteria

Selection criteria applied to the clinical trials were: 1. The population had to be preterm neonates 2. The intervention had to be administration of a1PI compared to placebo or no treatment within the first week of life 3. The eligible studies had to include any of the following outcomes: a. prevention of CLD b. reduction in duration of assisted ventilation c. reduction in the duration of oxygen requirement d. reduction in the need for systemic antiinflammatory therapy e. reduction in mortality or f. reduction in long term adverse neurological sequelae. 4. The trial had to be randomized or quasi-randomized.

Data collection & analysis

The methodological quality of the trials was assessed using the information provided in the studies and by personal communication with the authors. Data on relevant outcomes were extracted and the effect size was estimated and reported as pooled relative risk (RR), risk difference (RD) and weighted mean difference (WMD) as appropriate. Two strategies were used for analysis : 1. Comparison of 60 mg/kg/dose of a1PI administered for 4 days in first 2 weeks after birth, versus placebo and 2. Comparison of any dose of a1PI administered in first 2 weeks after birth, versus placebo.

Main results

Two eligible studies were identified. The methodological qualities of identified studies were good. One study randomized infants to either placebo, or a1PI 60 mg/kg/dose for four doses while in the second study the same investigators explored the efficacy of different dose regimens of a1PI compared to placebo. There was no statistically significant difference in the development of CLD at 36 weeks PMA amongst all randomized infants. For 60 mg/kg/dose for four doses of a1PI compared to placebo the pooled RR was 0.64 [95% CI 0.35, 1.18] and RD was -0.10 [95% CI -0.23, 0.03] while for any doses of a1PI the pooled RR was 0.79 [95% CI 0.44, 1.41], RD was -0.05 [95% CI -0.17, 0.06].

There was a trend towards reduced risk of development of oxygen dependency at 28 days postnatal age, pooled RR 0.81 [95% CI 0.64, 1.01], RD -0.14 [95% CI -0.29, 0.00] for 60 mg/kg/dose for four doses of a1PI compared to placebo. The pooled RR was statistically significant, RR 0.80 [95% CI 0.65, 0.98], RD -0.15 (95 % CI -0.29, -0.01) when any doses of a1PI were combined. The number of patients needed to treat (NNT) with a1PI to prevent one infant developing oxygen dependency at 28 weeks was 7.

The benefits of short-term outcomes did not remain for long-term outcomes such as CLD and/or death at 36 weeks post menstrual age (pooled RR 0.84 [95% CI 0.53, 1.34], RD -0.06 [95% CI -0.20, 0.09] for 60 mg/kg/dose and pooled RR 0.95 [95% CI 0.61, 1.49], RD -0.02 [95% CI -0.15, 0.12] for any dose) or risk of development of long term neurodevelopmental abnormalities (pooled RR 1.29 [95% CI 0.43, 3.90], RD 0.03 [95% CI -0.11, 0.18]). In addition, no statistically significant difference was noted in other respiratory parameters such as duration of oxygen requirement or respiratory support.

Reviewers' conclusions

Prophylactic administration of a1PI did not reduce the risk of CLD at 36 weeks or long term adverse developmental outcomes in preterm neonates.

Background

Advances in perinatal and neonatal care such as use of antenatal steroids (Crowley 2001), surfactant replacement therapy (Jobe 1993) and modification in ventilatory techniques have led to improved survival of preterm neonates and resulted in higher number of survivors with chronic lung disease (CLD) (Shaw 1993). Pulmonary sequelae after respiratory distress syndrome were identified initially by Shepard 1964; Robertson 1964 and later Northway 1967, who coined the term "bronchopulmonary dysplasia". Shennan 1988 defined CLD as oxygen requirement (with or without characteristic x-ray findings) persisting beyond 36 weeks post menstrual age (PMA). CLD is associated with prolonged hospitalization, increased risk of re-hospitalization and adverse neurodevelopmental outcomes amongst survivors.

The etiology of CLD is multi factorial. Inflammation of pulmonary parenchyma plays a major role in evolution of CLD (Ogden 1982; Rosenfeld 1984). As a result of the inflammation there is a massive influx of neutrophils in to the alveoli (Ogden 1982), liberation of pro - inflammatory mediators like cytokines (Jones 1996) and release of oxygen free radicals and proteinases in the alveoli. Merritt 1983 has shown that neutrophil elastase released from neutrophils is elevated in tracheal lavage fluid of neonates who go on to develop CLD. The proteinases in the pulmonary parenchyma can digest fibrillar collagen, elastin fibers, laminin and fibronectin (Gerdes 1988; Fletcher 1990).

Strategies or interventions for the prevention of CLD include antioxidants like superoxide dismutase (Davis 1998), vitamin E (Watts 1991), vitamin A (Darlow 2001), ceruloplasmin and anti proteinases (Rosenfeld 1986). Antioxidants neutralize the initial burst of free oxygen radicals. Failure of release of antioxidants results in tissue destruction with release of proteolytic enzymes such as elastase. Alpha-1 proteinase inhibitor (a1PI), a major inhibitor of elastase, prevents its proteolytic action on fibronectin and elastin (Rosenfeld 1984). Activity of a1PI has been evaluated in preterm and term neonates. Rosenfeld 1986 observed that a1PI activity and ceruloplasmin levels were low in all neonates, especially in less mature neonates who had higher incidence of CLD (Fletcher 1990; Evans 1972; Singer 1976). Furthermore Bruce 1981 showed that the levels of a1PI may be normal in preterm neonates but exposure to high oxygen concentration may result in inactivation of the methionine terminal of a1PI rendering it physiologically inactive.

a1PI in the form of alpha-1 antitrypsin has been used since 1987 in adult patients with alpha-1 antitrypsin deficiency as replacement therapy (Wewers 1987). Aerosolized alpha-1 antitrypsin has been used in patients with cystic fibrosis who have normal alpha-1 antitrypsin levels but excessive elastase activity (McElvaney 1991; Hubbard 1989). Koppel 1994 demonstrated that early administration of alpha-1 antitrypsin to newborn rats exposed to hyperoxia protects them from developing the anticipated pulmonary and vascular changes and the resultant right ventricular hypertrophy. In preclinical animal experiments, Fournel 1988 demonstrated a1PI extravasation into pulmonary alveolar spaces with intravenous administration of a1PI resulting in sustained and proportionate increase in alveolar antielastase activity without abnormal clearance or significant inactivation of a1PI. No physiological or systemic adverse responses to anticipated clinical doses were observed and the toxicologic studies also failed to reveal any acute or subchronic effects. Known side effects attributable to a1PI administration are delayed fever and transient leucocytosis (Hebel 2000).

With these observations, it seems that improving the proteinase - antiproteinase balance may have beneficial effects on CLD (Gadek 1981; Yoder 1991). However, Gerdes 1988 failed to show sustained clinical improvement after administration of dexamethasone which reduced broncheoalveolar lavage fluid elastase concentration.

The aim of this review was to assess the effectiveness of a1PI therapy when administered to ventilated preterm neonates in the first week of life in the prevention of CLD.

Objectives

The primary objective was to compare the effectiveness of a1PI versus placebo or no treatment administered to ventilated preterm neonates beginning in the first week of life for the prevention of CLD [defined as requirement for supplemental oxygen at 36 weeks PMA (Shennan 1988)].

Secondary objectives were to compare the effectiveness of a1PI versus placebo on:

1. Other indicators of CLD including:
Requirement for supplemental oxygen at 28 days of postnatal age
Duration of requirement for supplemental oxygen
Duration of assisted ventilation
Requirement for systemic antiinflammatory therapy
Changes in pulmonary function tests (lung compliance and airway resistance)

2. Incidence of adverse events including:
Mortality (expressed as early neonatal mortality <7 days; neonatal mortality <28 days and mortality prior to hospital discharge)
Growth impairment (weight, head circumference and length) at 36 weeks PMA
Side effects including delayed fever, transient leucocytosis and other important side effects

3. Long term adverse neurodevelopmental outcome (defined as presence of cerebral palsy and/or mental retardation [Bayley Scales of Infant Development- Mental Developmental Index<70] and/or legal blindness [<20/200 bilateral visual acuity] and /or deafness [aided or <60 dB on audiometric testing or the need for hearing aid]).

Criteria for considering studies for this review

Types of studies

Randomized or quasi-randomized clinical trials comparing a1PI administered via inhalation or by systemic route (regardless of dose and duration) in preterm neonates (<37 weeks PMA) enrolled within the first week of life.

Types of participants

Preterm neonates (<37 weeks PMA) receiving assisted ventilation and postnatal age < 1 week.

Types of interventions

Alpha-1 proteinase inhibitor (a1PI) versus placebo or no treatment

Types of outcome measures

The primary outcome measure was development of CLD [defined as requirement for supplemental oxygen at 36 weeks PMA (Shennan 1988)] amongst all randomized patients.

Secondary outcome measures were

1. Amongst all randomized

a. CLD at 28 days (defined as requirement for supplemental oxygen at 28 days of age)
b. Death by 36 weeks PMA
c. Death by 28 days postnatal age
d. Death by 7 days postnatal age
e. CLD or death by 36 weeks PMA
f. CLD or death by 28 days of age postnatal age

2. Amongst survivors:

a. CLD by 36 weeks PMA
b. CLD by 28 days postnatal age
c. Duration of requirement for supplemental oxygen
d. Duration of assisted ventilation
e. Requirement for systemic antiinflammatory therapy
f. Changes in pulmonary function tests

3. Incidence of adverse events including:

a. Mortality (expressed as early neonatal mortality < 7 days; neonatal mortality < 28 days and mortality prior to hospital discharge)
b. Growth impairment (weight, head circumference and length < 3rd percentile) at 36 weeks PMA
c. Side effects including delayed fever, transient leucocytosis or other side effects

4. Long term neurodevelopmental outcome (frequency of cerebral palsy and/or mental retardation, legal blindness and /or deafness)

Search strategy for identification of studies

See: Collaborative Review Group search strategy
Randomized or quasi randomized controlled trials comparing a1PI versus placebo were identified from MEDLINE (1966 - February 2005) using MeSH headings: infant-newborn, chronic lung disease, bronchopulmonary dysplasia; lung inflammation, antiinflammatory agents, alpha-1 proteinase inhibitor, alpha-1 antitrypsin; administration, inhalation, intravenous.

Other databases that were searched included: EMBASE (1980 to February 2005), CINAHL (1982 to February 2005), Cochrane Central Register of Controlled Trials (CENTRAL, The Cochrane Library, Issue 1, 2005), reference lists of published trials, and abstracts from the annual meetings of the Society of Pediatric Research, American Pediatric Society and Pediatric Academic Societies published in Pediatric Research (1991-2005). No language restriction were applied. No dosage regimen was specified. The candidate articles were screened by the two review authors (PS, AO) to identify articles eligibility for inclusion in the review.

Methods of the review

Standardized methods of the Cochrane Neonatal Collaborative Review Group was used to assess methodological quality of the studies.

All published articles identified as potentially relevant by the literature search were assessed for inclusion in the review.
In order to be included the trial had to meet the following criteria:

1. The study population had to be preterm neonates admitted to a neonatal intensive care unit
2. The intervention had to be a1PI administered via inhalation or intravenous route
3. The study had to be a randomized or quasi randomized controlled trial
4. At least one primary or secondary outcome was reported.

Quality of included trials was evaluated by the following questions:
1. Masking of randomization
2. Masking of intervention
3. Complete follow-up
4. Masking of outcome measurement.

Data from the primary author were obtained if published data provided inadequate information for the review. Retrieved articles were assessed and data were abstracted independently by the review authors. Discrepancies were resolved by consensus among the two review authors.

Subgroup analyses were planned a priori based on route of administration (intravenous or inhalation). Statistical methods included relative risk (RR), risk difference (RD), number needed to treat (NNT) and weighted mean difference (WMD) when appropriate as recommended by the Cochrane Collaboration Neonatal Review Group. A fixed effect model was used for meta-analysis.

Two studies contributed to this review. The analyses were performed using two different strategies:
1. Comparing 60 mg/kg/dose for four doses of a1PI with placebo
2. Comparing any dose of a1PI with placebo
The decision to conduct separate analyses according to dose was made after the protocol was written.

Description of studies

Two studies were included in this review, Stiskal 1998 and Dunn 2000. Dunn 2000 has been published only in abstract form. Clinical details concerning the participants, interventions and outcomes are given in the table, Characteristics of Included Studies.

Stiskal 1998 randomized patients to receive a1PI (60 mg/kg) or placebo (Human Albumin 5%, equal volume- 2.4 ml/kg). Infusion was administered over 5 minutes within 24 hours of birth and at four, seven and fourteen days of age. Other clinical management decisions were left to the clinical care team responsible for the patients. A total of 106 patients were enrolled (53 in each group). Recruitment was halted after 2 deaths occurred in one of the centres after a review by the safety committee the study was allowed to progress. There were four protocol violations and one family refused the participation of their neonate after the first dose had been given.

The primary outcome assessed in the Stiskal 1998 study was the incidence of CLD among the survivors (oxygen requirement at 28 postnatal days). Secondary outcomes assessed were oxygen requirements at 36 weeks PMA and the use of systemic steroids. Strict criteria were laid out to define whether patients met the criteria for steroid use (requirement for mechanical ventilation and FiO2> 0.3 for at least 48 hours after 16 days of life) but achievement of these criteria didn't mandate the use of steroid. The decision to give steroids or not was left at the discretion of the team taking care of the infant. Other efficacy parameters monitored were age at final extubation, age off supplemental oxygenation, age off respiratory support, x-ray changes at 28 days and evidence of right ventricular hypertrophy on electrocardiogram. The incidence of pulmonary hemorrhage, intraventricular hemorrhage, periventricular leucomalacia, necrotizing enterocolitis, retinopathy of prematurity, sepsis and PDA were recorded for both groups.

The follow-up of the cohort was published in abstract form (Kelly 1998). Authors were contacted to confirm that the patients were the same as in the original trial. Eighty three of 94 surviving infants were assessed. Five infants were assessed at <12 months age but the rest were assessed at >18 months of age. Patients were assessed at standard interval to >18 months of age for developmental progress. Patients were classified in 3 groups: 1. Normal 2. Mildly impaired (motor abnormality without significant functional limitation and/or mild hearing loss and/or severe myopia and/or developmental quotient > 1 < 2 SD below mean and 3. Severely impaired (motor abnormality with significant functional limitation and/or deafness and/or blindness and/or developments quotient > 2 SD below the mean).

Dunn 2000 performed an open randomized study to determine the effect of different dosing regimens of a1PI. Ventilated neonates weighing 500 - 1250g were randomized to four groups.

The placebo used in this study was normal saline solution (2.4 ml/kg).
Group 1. Placebo administered on days 1, 4, 7, 14 (24 patients)
Group 2. a1PI at 60 mg/kg administered on days 1, 4, 7, 14 (21 patients)
Group 3. a1PI at 120 mg/kg administered on days 1, 4, 7, 14 (21 patients)
Group 4. a1PI at 60 mg/kg administered on days 1, 2, 3, 4, 7, 14 (23 patients)
The outcomes assessed were supplemental oxygen requirement at 28 days postnatal age and 36 weeks PMA amongst survivors.

Methodological quality of included studies

For each study included in the review, assessments of methodological quality are given in the table, Characteristics of Included Studies.

The methodological details for both studies were extracted from the published information and personal contact with the authors.

In the study by Stiskal 1998 randomization was performed centrally by the pharmacist using computer generated random table numbers. Randomization was stratified according to birth weight and whether patient was out born or inborn. Allocation was concealed in sequential opaque, sealed envelopes. Randomization was done in blocks of four. Masking of the intervention was done to ensure that investigators, parents and care takers were unaware of treatment allocation. Crossover was not allowed and contamination was not possible because study drug was not available outside the trial. Co-interventions were possible for the use of other interventions which may be effective in reducing the incidence of chronic lung disease and were monitored. Short term outcome assessments (need for oxygen at 28 days postnatal age and 36 weeks PMA) were blinded as the assessors were not aware of treatment allocation. Roentgenograms were performed at 28 days age and reviewed by an independent radiologist.

For the long term outcomes of this group of patients as reported in Kelly 1998, personal communication with authors revealed that the outcome assessments were blinded. Outcome assessments were performed in all the major areas of infant development. Outcome data were presented amongst infants assessed at follow up.

In the study by Dunn 2000 randomization was performed centrally by the pharmacist using computer generated random table numbers (personal communication). Allocation was not concealed. Randomization was done in blocks of 4. Groups were stratified according to birth weight. Masking of the intervention was not performed as one of the groups received 120 mg/kg/dose of a1PI and it was not possible to control the volume of placebo (personal communication). Crossover was not allowed and contamination was not possible because study drug was not available outside the trial. Co-interventions were possible for the use of other interventions which may be helpful in reducing the incidence of CLD but was monitored. Short term outcome assessments (need for oxygen at 28 days postnatal and 36 weeks PMA) were not blinded. Long-term follow up for this study was not performed (personal communication).

Results

PRIMARY OUTCOME:

Development of CLD at 36 weeks PMA (defined as requirement for supplemental oxygen at 36 weeks PMA) amongst all randomized patients:

There was no statistically significant difference in risk of developing CLD between treatment and placebo either with 60 mg/kg/dose for four doses (pooled RR 0.64 [95% CI 0.35, 1.18], RD -0.10 [95% CI -0.23, 0.03]) or when any doses of a1PI were combined (pooled RR 0.79 [95% CI 0.44, 1.41], RD -0.05 [95% CI -0.17, 0.06]).

SECONDARY OUTCOMES:

1. Amongst all randomized

a. CLD at 28 days (defined as requirement for supplemental oxygen at 28 days of age):

There was no statistically significant difference in risk of developing CLD between treatment and placebo with 60 mg/kg/dose for four doses (pooled RR 0.81 [95% CI 0.64, 1.01], RD -0.14 [95% CI -0.29, 0.00]). When all doses of a1PI were combined a statistically significant reduction in risk was noted (pooled RR 0.80 [95% CI 0.65, 0.98], RD -0.15 [95% CI -0.29, -0.01]). The number of patients needed to treat with a1PI to prevent the development of CLD at 28 days in one infant was seven.

b. Death by 36 weeks PMA:

No statistically significant difference were noted in the risk of death by 36 weeks between treatment and placebo either with 60 mg/kg/dose for four doses (pooled RR 1.56 [95% CI 0.58, 4.17], RD 0.04 [95% CI -0.05, 0.14]) or when any doses of a1PI were combined (pooled RR 1.48 [95% CI 0.59, 3.72], RD -0.04 [95% CI -0.05, 0.12]).

c. Death by 28 days

There was no statistically significant difference in risk of death by 28 days between treatment and placebo with 60 mg/kg/dose for four doses (pooled RR 1.71 [95% CI 0.56, 5.24], RD 0.04 [95% CI -0.04, 0.13]) or when any doses of a1PI were combined (pooled RR 1.81 [95% CI 0.58, 5.67], RD 0.04 [95% CI -0.03, 0.12]).

d. Death by 7 days

Similarly, there was no statistically significant difference in the risk of death before 7 days between treatment and placebo with 60 mg/kg/dose for four doses (RR 1.33 [95% CI 0.31, 5.67], RD 0.02 [95% CI -0.08, 0.11]).This outcome was reported by Stiskal 1998 only. The information was not available for the second study.

e. CLD or death by 36 weeks PMA

There was no statistically significant difference in risk of developing CLD or death by 36 weeks PMA between treatment and placebo either with 60 mg/kg/dose for four doses (pooled RR 0.84 [95% CI 0.53, 1.34], RD 0.06 [95% CI -0.20, 0.09]) or when any doses of a1PI were combined (pooled RR 0.95 [95% CI 0.61, 1.49], RD -0.02 [95% CI -0.15, 0.12]).

f. CLD or death by 28 days of age:

There was no statistically significant difference in the risk of developing CLD or death by 28 days between treatment and placebo either with 60 mg/kg/dose for four doses (pooled RR 0.65 [95% CI 0.36, 1.20], RD -0.10 [95% CI -0.24, 0.04]) or when any doses of a1PI were combined (pooled RR 0.87 [95% CI 0.73, 1.04], RD -0.10 [95% CI -0.23, 0.03]).

2. Amongst survivors:

a. CLD at 36 weeks PMA:

No statistically significant difference was noted in the risk of developing CLD at 36 weeks between treatment and placebo either with 60 mg/kg/dose for four doses (pooled RR 0.67 [95% CI 0.37, 1.24], RD -0.10 [95% CI -0.24, 0.04]) or when any doses of a1PI were combined (pooled RR 0.83 [95% CI 0.47, 1.47], RD -0.05 [95% CI -0.18, 0.08]).

b. CLD by 28 days:

There was no statistically significant difference in the risk of developing CLD at 28 days between treatment and placebo either with 60 mg/kg/dose for four doses (pooled RR 0.84 [95% CI 0.68, 1.04], RD -0.12 [95% CI -0.27, 0.02]) or when any doses of a1PI were combined (pooled RR 0.84 [95% CI 0.69, 1.02], RD -0.12 [95% CI -0.26, 0.01]).

c. Duration of requirement for supplemental oxygen:

There was no reduction in the duration of requirement of supplemental oxygen amongst the survivors (MD -11.2 days [95% CI -26.34, 3.94]). This was reported by Stiskal 1998 only.

d. Duration of assisted ventilation:

There was no reduction in the duration of the requirement of any kind of respiratory support amongst survivors (WMD -5.61 days [95% CI -18.04, 6.82]) between treatment and placebo with 60 mg/kg/dose for four doses.

e. Requirement for systemic antiinflammatory therapy:

There was no reduction in the need for systemic anti-inflammatory therapy amongst the survivors (RR 1.02 [95% CI 0.76, 1.38], RD 0.01 [95% CI -0.18, 0.20]). This outcome was reported only by Stiskal 1998, who set strict and uniform criteria for administration of systemic anti-inflammatory therapy.

f. Changes in pulmonary function tests: not tested in either study.

3. Incidence of adverse events including:

a. Mortality prior to hospital discharge:

There was no statistically significant difference in risk of death before discharge from the hospital with 60 mg/kg/dose for four doses (RR 1.75 [95% CI 0.54, 5.63], RD 0.06 [95% CI -0.06, 0.17]). This outcome was is reported by Stiskal 1998 only.

b. Growth impairment (weight, length and head circumference below 3rd percentile) at 36 weeks PMA: not reported in either study.

c. Side effects:

No disturbance in hematologic, renal hepatic parameters was observed by Stiskal 1998.

4. Long term neurodevelopmental outcome (frequency of cerebral palsy and/or mental retardation, legal blindness and /or deafness)

There was no difference in risk of severe neurodevelopmental abnormality amongst infants assessed in either group for 60 mg/kg/dose for four doses (RR 1.29 [95% CI 0.43, 3.90], RD 0.03 [95% CI -0.11, 0.18]). The long term outcome for the Dunn 2000 study is not available.

Discussion

An extensive literature search was conducted to identify trials that might qualify for this review. However, only two trials, both performed in a similar setting and by the same group of investigators, were identified. Even after combining trials appropriately the number of neonates providing data on outcomes remained very small (n=195). As a result, the fact that there was no statistically significant difference found between groups may be due to lack of statistical power.

The quality of included studies was good suggesting that extracted data were valid. However, there were some differences in design. The first study of (Stiskal 1998) compared the drug with placebo in a blinded randomized controlled trial while the second study (Dunn 2000) compared different dose regimens with placebo in an open randomized controlled trial. Methodological rigor was strict in both instances.

There was no statistically significant difference in the development of CLD at 36 weeks PMA. The difference remained insignificant even after higher dose and increased frequency of administration. The incidence of CLD at 28 days was reduced by a1PI but it was apparent that the difference was lost at 36 weeks PMA. The other outcome measures such as duration of oxygen requirement or duration of respiratory assistance was not statistically significantly different amongst the groups irrespective of the dose of a1PI.

Long term outcomes were evaluated in the Stiskal 1998 study only and were not found to be statistically significantly different between treatment and control groups.

Reviewers' conclusions

Implications for practice

The prophylactic administration of a1PI for prevention of CLD in preterm neonates may have short term benefits but to date there are no long term benefits reported. The use of a1PI cannot be recommended for routine use to prevent CLD in preterm infants.

Implications for research

With no trend towards long term benefits of a1PI it may be argued that further placebo-controlled trials are not warranted. However, the total number of patients enrolled to date is very small. The apparent lack of benefit may be due to a type II error.

Acknowledgements

Authors would like to thank Dr. Michael Dunn, Chief, Department of Newborn and Developmental Pediatrics, Sunnybrook and Women's College Health Sciences Centre, Toronto, Canada for providing information related to both trials included in this review.

Potential conflict of interest

Characteristics of included studies

References to studies

References to included studies

Dunn M, Stiskal J, O'Brien K, Ito S, Cox DW, Kelly E, Shennan A et al. a1Proteinase inhibitor (A1PI) therapy for the prevention of chronic lung disease of prematurity (CLD) - a dose ranging study and meta analysis with previous randomized clinical trial (RCT). Pediatric Research 2000;47:397A.

Stiskal 1998 {published data only}

Kelly EN, Azstalos E, O'Brien K, Stiskal J, Wylie L, Wylie G, Shennan A, Rabinovitch M, Dunn M. Neurodevelopmental outcome of babies <1250g enrolled in a randomised controlled trial (RCT) of alpha-1-proteinase inhibitor (a1PI) therapy. Pediatric Research 1998;43:219A.

* Stiskal JA, Dunn MS, Shennan AT, O'Brien KK, Kelly EN, Koppel RI et al. alpha 1- Proteinase inhibitor therapy for the prevention of chronic lung disease of prematurity: a randomized, controlled trial. Pediatrics 1998;101:89-94.

* indicates the primary reference for the study

Other references

Additional references

Bruce MC, Boat TF, Martin RJ, Dearborn DG, Fanaroff A. Inactivation of alpha1 proteinase inhibitor in infants exposed to high concentrations of oxygen. American Review of Respiratory Disease 1981;123:166.

Crowley 2001

Crowley P. Prophylactic corticosteroid for preterm delivery. In: The Cochrane Database of Systematic Reviews, Issue 1, 2001.

Darlow 2001

Darlow BA, Graham PJ. Vitamin A supplementation for preventing morbidity and mortality in very low birthweight infants. In: The Cochrane Database of Systematic Reviews, Issue 1, 2001.

Davis 1998

Davis JM. Superoxide dismutase: a role in the prevention of chronic lung disease. Biology of the Neonate 1998;74:29-34.

Evans 1972

Evans HE, Keller S, Mandl I. Serum trypsin inhibitory capacity and the idiopathic respiratory distress syndrome. Journal of Pediatrics 1972;81:588-92.

Fletcher 1990

Fletcher DS, Osinga DG, Hand KM, Dellea PS, Ashe BM, Mumford RA et al. A comparison of alpha 1-proteinase inhibitor methoxysuccinyl-Ala-Ala-Pro-Val-chloromethylketone and specific beta-lactam inhibitors in an acute model of human polymorphonuclear leukocyte elastase-induced lung hemorrhage in the hamster. American Review of Respiratory Disease 1990;141:672-7.

Fournel 1988

Fournel MA, Newgren JO, Betancourt CM, Irwin RG. Preclinical evaluation of alpha-1-proteinase inhibitor. Pharmacokinetics and safety studies. American Journal of Medicine 1988;84:43-7.

Gadek 1981

Gadek JE, Fells GA, Zimmerman RL, Rennard SI, Crystal RG. Antielastases of the human alveolar structures. Implications for the protease-antiprotease theory of emphysema. Journal of Clinical Investigation 1981;68:889-98.

Gerdes 1988

Gerdes JS, Harris MC, Polin RA. Effects of dexamethasone and indomethacin on elastase, alpha-1 proteinase inhibitor, and fibronectin in bronchoalveolar lavage fluid from neonates. Journal of Pediatrics 1988;113:727-31.

Hebel 2000

Hebel SK, Burnham TH, editor(s). Drug facts and comparisons. 2000 edition. Missouri: Facts and Comparisons:691.

Horbar 1993

Horbar JD, Wright EC, Onstad L. Decreasing mortality associated with the introduction of surfactant therapy: an observational study of neonates weighing 601 to 1300 grams at birth. Pediatrics 1993;92:191-6.

Hubbard 1989

Hubbard RC, McElvaney NG, Sellers SE, Healy JT, Czerski DB, Crystal RG. Recombinant DNA-produced alpha 1-antitrypsin administered by aerosol augments lower respiratory tract antineutrophil elastase defenses in individuals with alpha 1-antitrypsin deficiency. Journal of Clinical Investigation 1989;84:1349-54.

Jobe 1993

Jobe AH. Pulmonary surfactant therapy. New England Journal of Medicine 1993;328:861-8.

Jones 1996

Jones CA, Cayabyab RG, Kwong KY, Stotts C, Wong B, Hamdan H et al. Undetectable interleukin (IL)-10 and persistent IL-8 expression early in hyaline membrane disease: a possible developmental basis for the predisposition to chronic lung inflammation in preterm newborns. Pediatric Research 1996;39:966-75.

Koppel 1994

Koppel R, Han RN, Cox D, Tanswell AK, Rabinovitch M. Alpha1-antitrypsin protects neonatal rats from pulmonary vascular and parenchymal effects of oxygen toxicity. Pediatric Research 1994;36:763-70.

Mathis 1973

Mathis RK, Freier EF, Hunt CE, Krivit W, Sharp HL. Alpha-antitrypsin in the respiratory-distress syndrome. New England Journal of Medicine 1973;288:59-64.

McElvaney 1991

McElvaney NG, Hubbard RC, Birrer P, Chernick MS, Caplan DB, Frank MM, Crystal RG. Aerosol alpha1- antitrypsin treatment for cystic fibrosis. Lancet 1991;337:392-4.

Merritt 1983

Merritt TA, Cochrane CG, Holcomb K, Bohl B, Hallman M, Strayer D, Edwards DK 3d, Gluck L. Elastase and alpha 1-proteinase inhibitor activity in tracheal aspirates during respiratory distress syndrome. Role of inflammation in the pathogenesis of bronchopulmonary dysplasia. Journal of Clinical Investigation 1983;72:656-66.

Northway 1967

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

Ogden 1982

Ogden BE, Murphy SA, Saunders GC, Johnson JD. Lung lavage of newborns with respiratory distress syndrome (RDS): prolonged neutrophil influx is associated with bronchopulmonary dysplasia (BPD). American Review of Respiratory Disease 1982;125:194.

Ogden 1984

Ogden BE, Murphy SA, Saunders GC, Pathak D, Johnson JD. Neonatal lung neutrophils and elastase/proteinase inhibitor balance. American Review of Respiratory Disease 1984;130:817-21.

Robertson 1964

Robertson B, Tunnell R, Rudhe U. Late stages of pulmonary hyaline mambranes of the newborn. Acta Pediatrica Scandinavica 1964;53:433-46.

Rosenfeld 1984

Rosenfeld W, Concepcion L, Evans H, Jhaveri R, Brunot V. Alpha1 antitrypsin (AAT) activity in development of bronchopulmonary dysplasia (BPD). Pediatric Research 1984;18:403A.

Rosenfeld 1986

Rosenfeld W, Concepcion L, Evans H, Jhaveri R, Sahdev S, Zabaleta I. Serial trypsin inhibitory capacity and ceruloplasmin levels in prematures at risk for bronchopulmonary dysplasia. American Review of Respiratory Disease 1986;134:1229-32.

Schwartz 1994

Schwartz RM, Luby AM, Scanlon JW, Kellogg RJ. Effect of surfactant on morbidity, mortality, and resource use in newborn infants weighing 500 to 1500 g. New England Journal of Medicine 1994;330:1476-80.

Shaw 1993

Shaw NJ, Gill AB, Weindling AM, Cooke RW. The changing incidence of chronic lung disease. Health Trends 1993;25:50-3.

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.

Shepard 1964

Shepard F, Gray J, Stahlman M. The occurence of pulmonary fibrosis in children who had idiopathic respiratory distress syndrome. Journal of Pediatrics 1964;65:1078-9.

Singer 1976

Singer AD, Thibeault DW, Hobel CJ, Heiner DC. Alpha1-antitrypsin in amniotic fluid and cord blood of preterm infants with the respiratory distress syndrome. Journal of Pediatrics 1976;88:87-93.

Watts 1991

Watts JL, Milner R, Zipursky A, Paes B, Ling E, Gill G, Fletcher B, Rand C. Failure of supplementation with vitamin E to prevent bronchopulmonary dysplasia in infants less than 1,500 g birth weight. European Respiratory Journal 1991;4:188-90.

Wewers 1987

Wewers MD, Casolaro MA, Sellers SE, Swayze SC, McPhaul KM, Wittes JT, Crystal RG. Replacement therapy for alpha 1-antitrypsin deficiency associated with emphysema. New England Journal of Medicine 1987;316:1055-62.

Yoder 1991

Yoder MC Jr, Chua R, Tepper R. Effect of dexamethasone on pulmonary inflammation and pulmonary function of ventilator- dependant infants with bronchopulmonary dysplasia. American Review of Respiratory Disease 1991;143:1044-8.

Yost 2001

Yost CC, Soll RF. Early versus late selective surfactant treatment for neonatal respiratory distress syndrome. In: The Cochrane Database of Systematic Reviews, Issue 1, 2001.

Other published versions of this review

Shah P, Ohlsson A. Alpha-1 proteinase inhibitor (a1PI) for preventing chronic lung disease in preterm infants. In: The Cochrane Database of Systematic Reviews, Issue 3, 2001.

Comparisons and data

Notes

Published notes

Amended sections

Contact details for co-reviewers