Background: This review was performed to test the hypothesis that presymptomatic diagnosis, for example by newborn screening, and early treatment may prevent or reduce irreversible organ damage and thereby improve outcome and quality of life in patients with cystic fibrosis.
Objectives: To determine whether there is evidence that early diagnosis of cystic fibrosis by means of neonatal screening, followed by current treatment, improves survival and long term morbidity, without unacceptable adverse effects.
Search strategy: We searched the Cochrane Cystic Fibrosis and Genetic Disorders Trials Register. Additional studies were identified by one of the reviewers from handsearching conference proceedings not included in the Cochrane Register. Pharmaceutical companies manufacturing screening tests for cystic fibrosis were also contacted to identify any trials of neonatal screening for cystic fibrosis. Date of the most recent search of the Group's specialised register: January 2001.
Selection criteria: All randomised or pseudorandomised controlled trials, published and unpublished, comparing screening followed by early treatment to clinical diagnosis and later treatment in patients with cystic fibrosis.
Data collection and analysis: Four reviewers independently assessed trial eligibility and methodological quality and two of these reviewers independently extracted data.
Main results: Two trials involving a total of 1,124,483 neonates met inclusion criteria. A total of 210 patients with cystic fibrosis aged from zero to 11 years with a maximum follow-up of eleven years are included. Concealment of allocation was unclear in both studies. Sequence generation was adequate in one study and unclear in the other. Method to ascertain cases was similar in one study and not similar in the other. An intention-to-screen-analysis was possible in one study, but could not be made due to lack of data and was not performed in the other. Differences in study design, variation in outcomes reported and their summary measures precluded calculation of pooled screening estimates. Only data from one study could be analysed in this review. This study reported a reduced risk of weight and height below the fifth percentile among screened patients (odds ratio control compared with screened group for: weight 6.16, 95% CI 2.44, 15.57 and height 5.03, 95% CI 1.63, 15.63). Adverse effects among parents in the screened and control populations were examined, but it is difficult to assess how meaningful these results are as the timing of the administration of the questionnaire to each group was not clear. Estimation of direct medical costs of screening suggested it was cheaper to diagnose cystic fibrosis by screening rather than other methods. The costing methods used however were not fully described and costs have not been related to effect.
Reviewer's conclusions: There are few randomised controlled trials assessing the effectiveness of neonatal screening in cystic fibrosis. From the data available at this time, there is little evidence suggesting benefit from screening for cystic fibrosis in the neonatal period, although there is similarly little evidence of harm. This systematic review has identified the need for individual patient data from both included studies. Although we have not been able to perform a meta-analysis, this review provides a summary of all the information currently available from randomised controlled trials on the effectiveness of neonatal screening for cystic fibrosis.