Scientific Publications
AB1194 A proof of principle dietary intervention trial to examine the protective effect of broccoli bioactives, (specifically sulforaphane), on osteoarthritis
Annals of the Rheumatic Diseases
Background: Osteoarthritis (OA) accounts for more than a third of chronic moderate to severe pain in the UK. Sulforaphane (SFN) is a naturally occurring phytochemical derived from eating cruciferous vegetables, particularly broccoli. It has several biological activities that promote health, including anti-inflammatory properties. SFN has a potential role in limiting pain and cartilage destruction in OA.
Objectives: In a two centre, placebo-controlled, double-blind, two-arm parallel, RCT, proof of principle study with a primary outcome we evaluated whether dietary SFN (from consumption of broccoli) improves pain in participants with knee OA. Secondary objectives were to determine whether food containing SFN improves knee function, and the feasibility of the study design for an appropriately powered trial.
Methods: Participants with symptomatic and radiographic knee OA were recruited from regions of Norfolk and Leeds in the UK, were over 50 years of age with moderate to severe knee pain (at least 4 on 0-10 numeric rating scale), and radiological evidence of OA (Kellgren-Lawrence score 2-3). The intervention was a sensory-matched soup. Patients received either the intervention soup (300g high glucoraphanin soup (containing broccoli, and base vegetables)) or placebo soup (300g no glucoraphanin soup (base vegetables only)), once daily on 4 days per week. The study duration was 12 weeks with follow-up visits at 6 and 12 weeks. Knee pain and function outcome measures were obtained using WOMAC, an 11-point NRS, the ICOAP questionnaire and use of rescue analgesics/NSAIDS. Creatinine, metabolites of glucoraphanin, matrix metalloproteinase-3 and C-reactive protein were measured in plasma or 24-hour urine samples.
Results: Recruitment was severely curtailed due to the Covid-19 pandemic. In total n= 37 consented and n= 24 met screening criteria with one drop-out at week 12 (control n=17 and intervention n= 7), the recruitment aim was n= 64. Control group was 43% female, and mean age was 70.14 yrs [7.69 SD] (control), and 61.88 yrs [8.58 SD] (intervention). Mean BMI was 26.4[2.91 SD] (control) and 28.2[4.68 SD] (intervention). The intervention resulted in a trend towards a decrease in pain scores for each subscale of WOMAC, ICOAP and NRS measures: total WOMAC median difference from baseline (9.65 [CI:-0.78,20.09]), ICOAP constant (wk 12, 4.83 [CI: -1.99,11.64]) and intermittent pain (wk 12, 2.96 [CI: -0.31,6.24]), NRS overall pain (wk 12, 1.86 [CI: 0.13,3.58]), while a trend for increased satisfaction for knee function (wk12, -1.55 [CI: -3.55,0.44]) was also observed for this group. No change in rescue analgesics/NSAIDS for knee pain was seen in either group. Metabolites of glucoraphanin were detected in plasma and urine samples in the intervention group. Study compliance and intervention adherence was high (>70%-100%) while acceptability for randomisation was 100%.
Conclusion: Patient acceptability, adherence and retention was high and biomarkers and metabolites objectively confirming compliance with the intervention were detected. In this study, although underpowered to observe significant differences for pain, improving trends over time were observed across a range of clinical pain measures including the primary outcome, where the CI encompassed the minimal clinically important difference, requisite for progress to full trial. Given COVID impact on recruitment, the pilot achieved its aims, and we conclude that the intervention was feasible, and a full trial is justified.
Annals of the Rheumatic Diseases
View Publication
