Fifteen RCTs (n=1,775) were included: seven evaluated glucosamine sulphate and eight evaluated chondroitin sulphate.
The quality scores ranged from 60 to 100%, with a mean of 78.4% (SD=17.2%). The mean quality of glucosamine RCTs (90%) was significantly higher than that of chondroitin RCTs (68.4%) (Mann-Whitney U-test, adjusted z=2.27, P=0.02). None of the pooled analyses were reported to have statistical heterogeneity.
Glucosamine statistically significantly decreased JSN (2 RCTs; ES 0.41, 95% CI: 0.21, 0.60, P<0.001)).
LI (10 RCTs): active treatment (with either glucosamine or chondroitin) was found to significantly reduce the LI in comparison with placebo (ES 0.43, 95% CI: 0.32, 0.54, P<0.001). There was no significant difference between glucosamine and chondroitin RCTs (heterogeneity, P=0.68).
WOMAC (2 RCTs): 1,500 mg/day glucosamine for 3 years significantly reduced the WOMAC (ES 0.30, 95% CI: 0.11, 0.49, P<0.001).
Pain (12 RCTs): active treatment (with either glucosamine or chondroitin) was found to significantly reduce pain measured by VAS in comparison with placebo (ES 0.45, 95% CI: 0.33, 0.57, P<0.001).
Mobility (3 RCTs): active treatment (with either glucosamine or chondroitin) was found to significantly increase mobility (ES 0.59, 95% CI: 0.25, 0.92, P<0.001).
Responders (9 RCTs): there were significantly more responders in the active treatment groups (either glucosamine or chondroitin) than in the placebo group (ES 1.59, 95% CI: 1.39, 1.83, P<0.001).
Safety (11 RCTs): there was no significant difference in the overall number of adverse events reported for active treatment compared with placebo (ES 0.80, 95% CI: 0.59, 1.08, P=0.15).
Funnel plot asymmetry was observed (in addition, P=0.08), suggesting publication bias in favour of the intervention.