Abstract Background Switching from boosted PIs to dolutegravir in virologically suppressed HIV-infected patients with high cardiovascular risk significantly decreased total cholesterol and other proatherogenic lipid fractions at 48 weeks. The impact of this strategy on subclinical cardiovascular disease is unknown. Methods NEAT022 is a European, multicentre, open-label, randomized, non-inferiority trial. HIV-infected adults aged >50 years or with a Framingham score >10% were eligible if plasma HIV RNA was <50 copies/mL for >24 weeks on a boosted PI-based regimen. Patients were randomized 1:1 to switch from boosted PIs to dolutegravir or to continue on boosted PIs. Common carotid arteries intima–media thickness (CIMT) and pulse wave velocity (PWV) were measured following a standardized protocol in a subgroup of NEAT022 study participants at baseline and at Week 48. Results One hundred and fifty-six patients participated in the ultrasonography and arterial stiffness substudies, respectively. In each substudy, population characteristics did not differ between arms and matched those of the main study. At 48 weeks, patients who switched to dolutegravir had lower mean progression of both right (+4 versus +14.6 μm) and left (−6.1 versus +1.6 μm) CIMT and also a smaller increase in mean PWV (+0.18 versus +0.39 m/s) than patients continuing on boosted PIs, although differences were not statistically significant. CIMT trends were consistent across Framingham score, age and country. Inconsistent effects were seen in arterial stiffness. Conclusions Relative to continuing on boosted PIs, switching to dolutegravir in virologically suppressed patients with high cardiovascular risk showed consistent favourable although non-significant trends on CIMT progression at 48 weeks.