This review assessed the efficacy and safety of immune globulins in preventing infectious hepatitis and hepatitis A. The authors concluded that given the limitations of the evidence, immune globulins should only be used where there are inadequate supplies of the vaccine, or after the 8-day window of opportunity for vaccination is past. The studies were generally of a poor quality and the authors' conclusions reflect the limited evidence.

To assess the effectiveness and safety of immune globulins (IG) in preventing infectious hepatitis and hepatitis A virus (HAV) infection.

The Cochrane Library, MEDLINE, EMBASE, Biological Abstracts and the Science Citation Index were searched to December 2002 for reports in any language; the search terms were reported. IMSworld New Product Launches and a chemical sources chemical directory were searched to identify manufacturers, and vaccine manufacturers were contacted. The bibliographies of relevant reports were also checked. The Cochrane Database of Systematic Reviews and DARE were also searched.

Study designs of evaluations included in the review

Studies in which the participants were definitely or possibly prospectively randomly allocated to the treatment groups were considered to be randomised controlled trials (RCTs) and were eligible for inclusion.

Specific interventions included in the review

Studies that compared IG with no treatment or placebo for the prevention of HAV infection were eligible for inclusion. The studies could use IG in any dose, preparation or time schedule. The included studies used IG for primary prevention and for prevention after exposure.

Participants included in the review

Studies of any person, regardless of age, sex, gender, immune status or risk category, were eligible for inclusion. The included studies were conducted in children and adults.

Outcomes assessed in the review

Studies that assessed efficacy or safety were eligible for inclusion. The review assessed outcomes reported as 'infectious hepatitis' and 'hepatitis A' by the original authors. The review did not assess serological outcomes. The included studies used different definitions for the outcomes: the diagnosis of viral hepatitis was based on medial history, physical examination and laboratory results; the diagnosis of infectious hepatitis was based on clinical jaundice and laboratory results or undefined; and hepatitis A was defined as non-B viral hepatitis or not defined. The included studies followed patients for 2 to 12 months.

How were decisions on the relevance of primary studies made?

Three reviewers independently selected studies and resolved any disagreements on inclusions through recourse to a fourth author.

The studies were assessed for adequacy of randomisation, sequence generation, allocation concealment, blinding and missing data. These criteria were adapted from criteria in the Cochrane Reviewers' Handbook. Validity was assessed, but the authors did not state who performed the validity assessment.

The data were extracted using a structured form, but the authors did not state how many reviewers performed the data extraction.

How were the studies combined?

Pooled relative risks (RRs) and 95% confidence intervals (CIs) were calculated using both fixed-effect and random-effects models. Effectiveness was calculated as one minus the RR and converted into a percentage. Publication bias was assessed using a funnel plot and was tested using Egger's test.

How were differences between studies investigated?

Statistical heterogeneity was assessed using the chi-squared statistic, and random-effects models were used in the presence of significant heterogeneity (P<0.1). Potential causes of differences among the studies were discussed in the text of the review. A sensitivity analysis was used to assess the influence on the results of study quality (no details of this analysis were reported) and age of the participants (all ages, adults and children).

Six studies (n=279,944) provisionally classified as RCTs were included.

The studies were generally of a poor quality, and the reporting of study design was poor. Only one RCT was clearly both randomised and blinded.

There was some asymmetry in the funnel plot for primary prevention, suggesting the possibility of publication bias.

IGs were more effective than placebo or no treatment in the primary prevention of infectious hepatitis and hepatitis A across all ages; 4 RCTs suggested an effectiveness of 83% (fixed-effect RR 0.17, 95% CI: 0.15, 0.19, P<0.00001; random-effects RR 0.17, 95% CI: 0.06, 0.51, P=0.001). There was significant statistical heterogeneity among the studies (P<0.00001), although the studies showed the same direction of effect. The effectiveness of IG was lower in adults (52%) than in children (90%).

IGs were more effective than placebo or no treatment in the prevention of infectious hepatitis and hepatitis A after exposure across all ages; 2 RCTs suggested an effectiveness of 69% (fixed-effect and random-effects RR 0.31, 95% CI: 0.20, 0.47). No significant statistical heterogeneity was found (P=0.83).

Two studies reported very limited safety data.

IGs were effective in preventing infectious hepatitis and hepatitis A but there was no evidence on safety. Given the limitations of the evidence, IG should only be used where there are inadequate supplies of the vaccine, or after the 8-day window of opportunity for vaccination is past.

The review question was clear in terms of the study design, intervention and outcomes. However, the authors accepted possible as well as definite RCTs, and any definition of outcomes as reported in the original studies. Several relevant sources were searched and attempts were made to minimise publication and language bias. The authors also assessed the possibility of publication bias. Three reviewers independently selected studies, thus reducing the potential for bias and errors. However, the methods used to assess validity and extract the data were not described, so it is not known any whether efforts were made to reduce errors and bias. Validity was assessed using established criteria. Issues such as baseline comparability of the treatment groups were not discussed, while the validity of methods used to assess the outcomes was mentioned only briefly.

The studies were combined using a meta-analysis and statistical heterogeneity was assessed. Meta-analysis graphs were drawn and the influence of some variables on the results was explored. Although the direction of treatment effect was consistent among the studies, the effect size differed and significant statistical heterogeneity was identified; this suggests that a summary treatment effect was not appropriate. More comprehensive information on the characteristics of the participants and IG interventions would have been helpful in judging how generalisable these results are (though this information might not have been reported in the original studies). The studies were generally of a poor quality and the outcomes, in particular, were poorly defined. These facts, which weaken the evidence considerably, were acknowledged in the authors' cautious conclusions.

Practice: The authors stated that the only indications for use of IG in secondary prevention of HAV may be where supplies of the vaccine are inadequate, or after the 8-day window for vaccination is past. For other situations, the authors suggested that the vaccine be used.

Research: The authors did not state any implications for further research.

Viral Hepatitis Project and 'Programme Nazionale Linee Guida', Istituto Superiore di Sanita (D. Leg.vo. 30/12/1992 n. 502).

This is a critical abstract of a systematic review that meets the criteria for inclusion on DARE. Each critical abstract contains a brief summary of the review methods, results and conclusions followed by a detailed critical assessment on the reliability of the review and the conclusions drawn.