Association between disease-specific quality of life and magnetic resonance imaging outcomes in a clinical trial of prolotherapy for knee osteoarthritis

Objective: To assess the relation between knee osteoarthritis (KOA)-specific quality of life (QOL) and intra-articular cartilage volume (CV) in participants treated with prolotherapy. KOA is characterized by CV loss and multifactorial pain. Prolotherapy is an injection therapy reported to improve KOA-related QOL to a greater extent than blinded saline injections and at-home exercise, but its mechanism of action is unclear. Design: Two-arm (prolotherapy, control), partially blinded, controlled trial. Setting Outpatient. Participants: Adults with ≥3 months of symptomatic KOA (N=37). Interventions: Prolotherapy: 5 monthly injection sessions; Control: blinded saline injections or at-home exercise. Main Outcome Measures: Primary: KOA-specific QOL scores (baseline, 5, 9, 12, 26, and 52wk; Western Ontario and McMaster University Osteoarthritis Index). Secondary: KOA-specific pain, stiffness, function (Western Ontario McMaster University Osteoarthritis Index subscales), and magnetic resonance imaging-assessed CV (baseline, 52wk). Results: Knee-specific QOL improvement among prolotherapy participants exceeded that among controls (17.6±3.2 points vs 8.6±5.0 points; P=.05) at 52 weeks. Both groups lost CV over time (P<.05); no between-group differences were noted (P=.98). While prolotherapy participants lost CV at varying rates, those who lost the least CV ("stable CV") had the greatest improvement in pain scores. Among prolotherapy participants, but not control participants, the change in CV and the change in pain (but not stiffness or function) scores were correlated; each 1% CV loss was associated with 2.7% less improvement in pain score (P<.05). Conclusions: Prolotherapy resulted in safe, substantial improvement in KOA-specific QOL compared with control over 52 weeks. Among prolotherapy participants, but not controls, magnetic resonance imaging-assessed CV change (CV stability) predicted pain severity score change, suggesting that prolotherapy may have a pain-specific disease-modifying effect. Further research is warranted.