BackgroundKnee osteoarthritis (KOA) is a progressive musculoskeletal disease that leads to physical impairment and disability. Several nonoperative therapeutic options have been proposed to reduce symptoms, improve physical function, and prevent disability. One of these therapeutic options is dextrose prolotherapy (DPT). The current study aimed to investigate the efficacy of DPT in the treatment of symptomatic KOA and to evaluate its potential proliferative role as assessed by musculoskeletal ultrasound (MSUS) and magnetic resonance imaging (MRI).ResultsSeventeen knees (21.25%) dropped out, leaving 63 knees in the final analysis (57 females and 6 males, aged 41–74 years) with KOA. At 24-week follow-up, the visual analogue scale and overall Western Ontario and McMaster Universities scores showed significant improvement (P < 0.001). Moreover, a highly significant increase in cartilage thickness with a decrease in effusion grade and meniscal extrusion was observed by MSUS (P < 0.001). By MRI, there was a statistically significant decrease in effusion grade (P = 0.013) and meniscal extrusion (P = 0.005) and a highly significant increase in cartilage thickness (P < 0.001). Logistic analysis showed that lower patients’ body mass index (BMI) at baseline was a significant predictor for clinical improvement following DPT.ConclusionDPT might be considered a relatively safe minimally invasive intervention for the treatment of KOA with high adherence and satisfaction. It may lead to significant clinical and structural improvement in patients with KOA. Furthermore, lower BMI is a predicting factor for better clinical response to DPT.Trial registrationThe study was registered prospectively in ClinicalTrials.gov: clinical trial registration number NCT04178304 on 21 November 2019, https://register.clinicaltrials.gov/XXX.