Acurácia do exame de condutividade no diagnóstico de Fibrose cística em recém-nascidos e lactentes jovens com triagem neonatal positiva

Abstract

Objective: this Thesis proposes to answer four questions: 1) verify the agreement between two concomitant sweat conductivity (SC) essays; 2) verify the rate and predictors for quantity not sufficient (QNS) of sweat collection to perform SC; 3) verify the rate and predictors for QNS of sweat collection to perform coulometry; 4) verify the SC accuracy as a diagnostic method for cystic fibrosis. Methods: This is a prospective study, with newborns and infants aged less than 3 months, identified by the statewide newborn screening program through two previous positive immunoreactive trypsinogen (IRT) results. For each question, a different methodology was adopted, described below. The results obtained were analyzed by appropriate statistical test. 1) two sweat samples were obtained simultaneously from each forearm of each patient at the same laboratory; 2) QNS was defined as a volume lower than 15 µl of sweat collection and the explanatory variables for this outcome were analyzed; 3) QNS was defined as a weight lower than 75 mg of sweat collection and the explanatory variables of this outcome were analyzed and 4) according to the STARD guidelines, the SC properties were calculated. Results 1) 320 patients were recruited. The Kappa index value was 1.0 and the Spearman's coefficient was 0.83; 2) 1020 patients were included. The QNS rate was 8.9%. A higher QNS rate was found in preterm infants and children weighing less than 3845g on the day of sweat collection (OR 5.8 and 5.3 respectively); 3) 1016 patients were recruited. The QNS rate was 3.3%. A higher QNS rate was found in preterm infants (OR 5.4), infants weighting less than 3800g (OR 4.0) on the day of sweat collection and a daily weight gain lower than 26g (OR 3.6). 4) 1176 patients were admitted. SC had sensitivity, specificity, PPV, NPV, accuracy, positive and negative likelihood ratios of 98.5%, 99.9%, 98.5%, 99.9%, 99.8%, 1091.7 and 0.01 respectively. Kappa coefficient and the area under the ROC curve were 0.99 and 98.8%, respectively.

Conclusion 1) SC showed an excellent agreement when performed repeatedly, under the same conditions, and in the same individual. 2) SC showed acceptable QNS rate, according to the CFF recommendation for this group. To avoid unnecessary SC repetition, it is suggested wait until 37 weeks (corrected age) and a weigh no less than 38450g on the day of the collection. 3) Coulometry showed acceptable QNS rate, according to the CFF recommendation for this group. For a success coulometry analysis, it would be indicated wait until 37 weeks (corrected age), presents a daily weight gain equal to or greater than 26g, and especially weight of not less than 3800g on the day of sweat collection. 4) Regarding SC accuracy excellent sensitivity, specificity, positive and negative predictive values, as well as positive and negative likelihood values were evidenced in a group of 1176 infants younger than three months. Thus, the results suggest that SC can be used safely as CF diagnosis in newborns and young infants.