Brachial-to-radial-systolic blood pressure amplification (Bra-Rad-SBPAmp) can affect central SBP estimated by radial tonometry. Patients with type 2 diabetes mellitus (T2DM) have vascular irregularities that may alter Bra-Rad-SBPAmp. By comparing T2DM with non-diabetic controls, we aimed to determine the (1) magnitude of Bra-Rad-SBPAmp; (2) haemodynamic factors related to Bra-Rad-SBPAmp; and (3) effect of Bra-Rad-SBPAmp on estimated central SBP. Twenty T2DM (64±8 years) and 20 non-diabetic controls (60±8 years; 50% male both) underwent simultaneous cuff deflation and two-dimensional ultrasound imaging of the brachial and radial arteries. The first Korotkoff sound (denoting SBP) was identified from the first inflection point of Doppler flow during cuff deflation. Bra-Rad-SBPAmp was calculated by radial minus brachial SBP. Upper limb and systemic haemodynamics were recorded by tonometry and ultrasound. Radial SBP was higher than brachial SBP for T2DM (136±19 vs 127±17 mm Hg; P<0.001) and non-diabetic controls (135±12 vs 121±11 mm Hg; P<0.001), but Bra-Rad-SBPAmp was significantly lower in T2DM (9±8 vs 14±7 mm Hg; P=0.042). The product of brachial mean flow velocity × brachial diameter was inversely and independently correlated with Bra-Rad-SBPAmp in T2DM (β=-0.033 95% confidence interval -0.063 to -0.004, P=0.030). When radial waveforms were calibrated using radial, compared with brachial SBP, central SBP was significantly higher in both groups (T2DM, 116±13 vs 125±15 mm Hg; and controls, 112±10 vs 124±11 mm Hg; P<0.001 both) and there was a significant increase in the number of participants classified with 'central hypertension' (SBP⩾130 mm Hg; P=0.004). Compared with non-diabetic controls, Bra-Rad-SBPAmp is significantly lower in T2DM. Regardless of disease status, radial SBP is higher than brachial SBP and this results in underestimation of central SBP using brachial-BP-calibrated radial tonometry.