Phase IV noninferiority controlled randomized trial to evaluate the impact on diagnostic thinking and patient management and the test-retest reproducibility of the Gaxilose test for hypolactasia diagnosis.

Background: The diagnostic accuracy of the Gaxilose test (GT) for hypolactasia diagnosis has already been proved. The objectives of this clinical trial were to demonstrate the noninferiority of the GT compared to the hydrogen breath test (HBT) on the impact on diagnostic thinking and patient management, to evaluate the GT reproducibility with urine accumulated from 0 to 4 hours and from 0 to 5 hours and to assess test safety.

Methods: We conducted a randomized, parallel, noninferiority clinical trial. Patients with clinical symptoms suggestive of lactose intolerance were screened for inclusion and randomly assigned to the GT arm or the HBT arm of the study. The impact on diagnostic thinking and patient management was analyzed with pretest and posttest questionnaires in which the investigators indicated their estimated probability of hypolactasia diagnosis and the intended management before and after the GT or the HBT (noninferiority margin: -10%). The primary outcome of the study was the impact on diagnostic thinking, expressed as the mean of the absolute values of the differences between the pretest and posttest probabilities of hypolactasia diagnosis. Patients randomized to the GT arm performed also the retest to evaluate the reproducibility of the GT.

Results: A total of 147 patients were included in the intend-to-treat (ITT) population. Among them, 74 performed the HBT and 73 performed the GT. The results proved the noninferiority of the GT compared to the HBT on the impact on diagnostic thinking (ImpactGT = 31.74 ± 23.30%; ImpactHBT = 24.28 ± 19.87%; ΔGT-HBT = 7.46%; 95% confidence interval of ΔGT-HBT: 1.55%, infinite) and on patient management. The test-retest reproducibility was better for the GT with urine accumulated from 0 to 5 h: the intraclass correlation coefficient (ICC) was 0.5761, and the Kappa coefficient was 0.7548, indicative of substantial agreement between both tests. No serious adverse events were reported during the study.

Conclusions: The GT has an impact on diagnostic thinking and patient management noninferior to that of the HBT, is reproducible and well tolerated. These results prove the clinical benefit of its use in the clinical practice (ClinicalTrials.gov identifier: NCT02636413).