INTRODUCTION: Thiazide diuretics are associated with increased risk of diabetes mellitus. This risk may arise from K(+)-depletion. We hypothesised that a K(+)-sparing diuretic will improve glucose tolerance, and that combination of low-dose thiazide with K(+)-sparing diuretic will improve both blood pressure reduction and glucose tolerance, compared to a high-dose thiazide. METHODS AND ANALYSIS: This is a parallel-group, randomised, double-blind, multicentre trial, comparing hydrochlorothiazide 25-50 mg, amiloride 10-20 mg and combination of both diuretics at half these doses. A single-blind placebo run-in of 1 month is followed by 24 weeks of blinded active treatment. There is forced dose-doubling after 3 months. The Primary end point is the blood glucose 2 h after oral ingestion of a 75 g glucose drink (OGTT), following overnight fasting. The primary outcome is the difference between 2 h glucose at weeks 0, 12 and 24. Secondary outcomes include the changes in home systolic blood pressure (BP) and glycated haemoglobin and prediction of response by baseline plasma renin. Eligibility criteria are: age 18-79, systolic BP on permitted background treatment ≥ 140 mm Hg and home BP ≥ 130 mm Hg and one component of the metabolic syndrome additional to hypertension. Principal exclusions are diabetes, estimated-glomerular filtration rate <45 mL/min, abnormal plasma K(+), clinic SBP >200 mm Hg or DBP >120 mm Hg (box 2). The sample size calculation indicates that 486 patients will give 80% power at α=0.01 to detect a difference in means of 1 mmol/L (SD=2.2) between 2 h glucose on hydrochlorothiazide and comparators. ETHICS AND DISSEMINATION: PATHWAY-3 was approved by Cambridge South Ethics Committee, number 09/H035/19. The trial results will be published in a peer-reviewed scientific journal. TRIAL REGISTRATION NUMBERS: Eudract number 2009-010068-41 and clinical trials registration number: NCT02351973.