Sweat test is the standard for cystic fibrosis (CF) diagnosis. Conductivity is an alternative method not yet approved, in spite of its good correlation with chloride concentration. The aim was to assess the capacity of sweat conductivity to discriminate between CF and non-CF subjects. Automated measurements of conductivity and chloride concentration were carried out on the same sweat samples from subjects with clinical suspicion of CF. Sweat samples from 3,834 subjects, median age 1.8 years (range 1 month-54 years) were analysed, and those with chloride titration >60 mmol/l were considered as CF patients (n=294). Conductivity median values in CF and non-CF subjects were 111 mmol/l (82-148) and 36 mmol/l (12-89), respectively. The Spearman correlation between chloride titration and conductivity was r=0.60 (P<0.001). The receiver operating characteristics (ROC) curve showed very high agreement between two methods. The best conductivity cut-off value to diagnose CF was > or =90 mmol/l (sensitivity 99.7%, specificity 100%, positive and negative predictive values of 100% and 99.97%, respectively, and kappa=0.998). Likewise, the best conductivity cut-off value to exclude CF was <75 mmol/l. The sweat conductivity method showed good correlation with chloride titration, and accurately discriminated between subjects with and without CF. In accordance with this, CF diagnosis might be confirmed for conductivity values > or =90 mmol/l and excluded for <75 mmol/l. Values between 75 and 89 mmol/l should correspond to an equivocal range. However, more studies are needed to confirm the role of conductivity in definitive CF diagnosis.