Background: Optimal target blood pressure to reduce adverse cardiac remodelling in children with chronic kidney disease is uncertain. We hypothesised that lower blood pressure would reduce adverse cardiac remodelling. Methods: HOT-KID, a parallel-group, open-label, multicentre, randomised, controlled trial, was done in 14 clinical centres across England and Scotland. We included children aged 2–15 years with stage 1–4 chronic kidney disease—ie, an estimated glomerular filtration rate (eGFR) higher than 15 mL/min per 1·73 m 2—and who could be followed up for 2 years. Children on antihypertensive medication were eligible as long as it could be changed to angiotensin-converting enzyme (ACE) inhibitors or angiotensin receptor blockers (ARBs) if they were not already receiving these therapies. Participants were randomly assigned (1:1) to standard treatment (auscultatory office systolic blood pressure target between the 50th and 75th percentiles) or intensive treatment (systolic target <40th percentile) by the chief investigator using a rapid, secure, web-based randomisation system. ACE inhibitors or ARBs were used as first-line agents, with the dose titrated every 2–4 weeks to achieve the target blood pressure levels. The primary outcome was mean annual difference in left ventricular mass index (LVMI) by echocardiography measured by a masked observer and was assessed in the intention-to-treat population, defined as all the children who underwent randomisation irrespective of the blood pressure reached. Secondary and safety outcomes were the differences between groups in mean left ventricular relative wall thickness, renal function, and adverse effects and were also assessed in the intention-to-treat population. This trial is registered with ISRCTN, ISRCTN25006406. Findings: Between Oct 30, 2012, and Jan 5, 2017, 64 participants were randomly assigned to the intensive treatment group and 60 to the standard treatment group (median age of participants was 10·0 years [IQR 6·8–12·6], 69 [56%] were male and 107 [86%] were of white ethnicity). Median follow-up was 38·7 months (IQR 28·1–52·2). Blood pressure was lower in the intensive treatment group compared with standard treatment group (mean systolic pressure lower by 4 mm Hg, p=0·0012) but in both groups was close to the 50th percentile. The mean annual reduction in LVMI was similar for intensive and standard treatments (–1·9 g/m 2·7 [95% CI –2·4 to –1·3] vs –1·2 g/m 2·7 [–1·5 to 0·8], with a treatment effect of –0·7 g/m 2·7 [95% CI –1·9 to 2·6] per year; p=0·76) and mean value in both groups at the end of follow-up within the normal range. At baseline, elevated relative wall thickness was more marked than increased LVMI and a reduction in relative wall thickness was greater for the intensive treatment group than for the standard treatment group (–0·010 [95% CI 0·015 to –0·006] vs –0·004 [–0·008 to 0·001], treatment effect –0·020 [95% CI –0·039 to –0·009] per year, p=0·0019). Six (5%) participants reached end-stage kidney disease (ie, an eGFR of <15 mL/min per 1·73 m 2; three in each group) during the course of the study. The risk difference between treatment groups was 0·02 (95% CI −0·15 to 0·19, p=0·82) for overall adverse events and 0·07 (−0·05 to 0·19, p=0·25) for serious adverse events. Intensive treatment was not associated with worse renal outcomes or greater adverse effects than standard treatment. Interpretation: These results suggest that cardiac remodelling in children with chronic kidney disease is related to blood pressure control and that a target office systolic blood pressure at the 50th percentile is close to the optimal target for preventing increased left ventricular mass. Funding: British Heart Foundation.