Relation of sweat chloride concentration to severity of lung disease in cystic fibrosis

In cystic fibrosis (CF), sweat chloride concentration has been proposed as an index of CFTR function for testing systemic drugs designed to activate mutant CFTR. This suggestion arises from the assumption that greater residual CFTR function should lead to a lower sweat chloride concentration, as well as protection against severe lung disease. This logic gives rise to the hypothesis that the lower the sweat chloride concentration, the less severe the lung disease. In order to test this hypothesis, we studied 230 patients homozygous for the DeltaF508 allele, and 34 patients with at least one allele associated with pancreatic sufficiency, born since January 1, 1955, who have pulmonary function data and sweat chloride concentrations recorded in our CF center database, and no culture positive for B. cepacia. We calculated a severity index for pulmonary disease, using an approach which takes into account all available pulmonary function data as well as the patient's current age and survival status. Patients with alleles associated with pancreatic sufficiency had significantly better survival (P = 0.0083), lower sweat chloride concentration (81.4 +/- 23.8 vs. 103.2 +/- 14.2 mEq/l, P < 0.0001), slower rate of decline of FEV(1) % predicted (-0.75 +/- 0.34 vs. -2.34 +/- 0.17% predicted per year), and a better severity index than patients homozygous for the DeltaF508 allele (median 73rd percentile vs. median 55th percentile, P = 0.0004). However, the sweat chloride concentration did not correlate with the severity index, either in the population as a whole, or in the population of patients with alleles associated with pancreatic sufficiency, who are thought to have some residual CFTR function. These data suggest that, by itself, sweat chloride concentration does not necessarily predict a milder pulmonary course in patients with cystic fibrosis.