Recombinant human DNase has been shown to reduce sputum viscosity and improve pulmonary function in patients with cystic fibrosis.1 The individual response is, however, highly variable with between one third and two thirds of patients showing an improvement in their forced expiratory volume in one second (FEV₁) of more than 10%.1 In view of the high cost of this treatment and the lack of long term studies confirming its safety, prediction of the response of individual patients is an important issue. In a larger series of patients both baseline pulmonary function and the clinical status of the patient were unable to predict the outcome.

As part of a multicentre phase IIIb trial we have compared the in vitro response assessed by total DNA content and fragment length in sputum with the response in pulmonary function in 92 patients with cystic fibrosis after six weeks of treatment with DNase. The study population consisted of 42 males and 50 females with a mean (SD) age of 20.6 (7.7) years with moderate pulmonary disease reflected by a forced vital capacity (FVC) of 30–70% predicted (mean (SD) 54.9 (9.8)%). All patients were clinically stable and did not receive intravenous antibiotic therapy in the six weeks prior to the study. Most of the patients were chronically colonised withPseudomonas aeruginosa. Sputum samples and pulmonary function tests were obtained at baseline and after six weeks of treatment with 2.5 mg DNase inhaled via a sidestream jet nebuliser with a Portaneb 50 compressor once daily. The median improvement in pulmonary function for the total group was +8% of baseline FEV₁ (range –25% to +138%) and +5.6% of baseline FVC (range –22 to +102%), similar to that reported in previous studies.1 The median DNA content decreased from 0.53 to 0.3 mg/ml sputum, as reported previously.2 A marked reduction in DNA fragment length was observed in the majority of cases but it remained unchanged in 18 patients. There was a small but significant difference in the change in FVC for patients with reduced DNA content compared with those in whom the DNA content remained unchanged (median FVC 8.1% versus 3.9%, p = 0.04, Wilcoxon test). No difference was found in the improvement in pulmonary function after six weeks of treatment in those patients with biochemical evidence of DNA fragmentation when compared with those with unaltered DNA fragment length (table 1). In addition, baseline DNA content and fragment length had no influence on the individual patient’s pulmonary function after six weeks of treatment.

Assessment of sputum DNA content and fragment length are therefore not helpful in predicting the clinical response of patients receiving DNase treatment. Heterogeneity of the sputum sample or non-uniform distribution of inhaled DNase within the lung may be among the factors responsible for this finding.