Performance evaluation of a new coulometric endpoint method in sweat testing and its comparison with classic Gibson & Cooke and chloridometer methods in cystic fibrosis

Abstract

Background: The objective of the study was to assess the diagnostic efficacy of the coulometric endpoint method and compare it with classic Gibson & Cooke and chloridometer methods. Methods: This study is a prospective clinical study comparing two conventional sweat testing methods with the coulometric endpoint method in previously diagnosed cystic fibrosis (CF) patients and a non-CF control group. All individuals underwent two simultaneous sweat collections. One sample of sweat, collected by the CFΔ collector coil system, was analyzed by two methods: the titrimetric Cl- measurement (Sherwood® Chloridometer 926S, Sherwood Scientific Ltd., Cambridge, UK) and the coulometric endpoint method (CF Δ Collection System®, UTSAT/Turkey); the second sample was collected from the other forearm by the Gibson & Cooke method and the collected sweat was analyzed by manual titration in accordance with the Schales & Schales method. Within-run and between-run imprecisions were evaluated via Cl- concentrations of 40, 70, and 130 mmol/L samples. Results: One hundred and seventy (60 CF and 110 controls) subjects were included in the study. All three sweat test methods discriminated CF subjects from the healthy individuals. The mean difference between the coulometric endpoint and titrimetric Cl- measurement methods was -1.5 mmol/L, (95% confidence limits of agreement, ranging from -8.9 to 15.9 mmol/L); the mean difference between manual titration vs. coulometric endpoint methods was 12.8 mmol/L, (95% confidence limits of agreement ranging from -9.7 to 45.3 mmol/L) and the mean difference between the manual titration and titrimetric Cl- measurement methods was 11.3 mmol/L, (95% confidence limits of agreement ranging from -7.8 to 40.5 mmol/L) based on a Bland-Altman analysis. In the Receiver operating characteristic (ROC) analysis, made on the basis that Cl- concentration values < 40 mmol/L exclude the CF diagnosis, the coulometric endpoint method resulted in 96.7% sensitivity and 100% specificity for a cut-offvalue of 58.5 mmol/L (AUC: 0.994; 95% CI = 0.986-1.000; p < 0.001). Conclusions: The coulometric endpoint method can be as reliable as quantitative sweat Cl- analysis and may be considered as a definitive diagnostic tool for CF.