Background Hypertension is prevalent in children on dialysis and associated with cardiovascular disease. We studied the blood pressure (BP) trends and the evolution of BP over 1 year in children on conventional hemodialysis (HD) vs. hemodiafiltration (HDF). Methods This is a post hoc analysis of the "3H - HDF-Hearts-Height" dataset, a multicenter, parallel-arm observational study. Seventy-eight children on HD and 55 on HDF who had three 24-h ambulatory BP monitoring (ABPM) measures over 1 year were included. Mean arterial pressure (MAP) was calculated and hypertension defined as 24-h MAP standard deviation score (SDS) >= 95th percentile. Results Poor agreement between pre-dialysis systolic BP-SDS and 24-h MAP was found (mean difference - 0.6; 95% limits of agreement -4.9-3.8). At baseline, 82% on HD and 44% on HDF were hypertensive, with uncontrolled hypertension in 88% vs. 25% respectively; p < 0.001. At 12 months, children on HDF had consistently lower MAP-SDS compared to those on HD (p < 0.001). Over 1-year follow-up, the HD group had mean MAP-SDS increase of +0.98 (95%CI 0.77-1.20; p < 0.0001), whereas the HDF group had a non-significant increase of +0.15 (95%CI -0.10-0.40; p = 0.23). Significant predictors of MAP-SDS were dialysis modality (beta = +0.83 [95%CI +0.51 - +1.15] HD vs. HDF, p < 0.0001) and higher inter-dialytic-weight-gain (IDWG)% (beta = 0.13 [95%CI 0.06-0.19]; p = 0.0003). Conclusions Children on HD had a significant and sustained increase in BP over 1 year compared to a stable BP in those on HDF, despite an equivalent dialysis dose. Higher IDWG% was associated with higher 24-h MAP-SDS in both groups.