Estimated pooled differences (95% CI) between glucosamine and placebo on a 0–100 scale (positive values indicate a greater reduction in the outcome measure for glucosamine) and p values for treatment–subgroup interactions
| All studies (n=1625 in five studies) | Knee OA only (n=1403 in four studies) | GH in knee OA (n=1058 in three studies) | GS in knee and hip OA (n=567 in three studies) | GS in knee OA (n=345 in two studies) | ||
| Pain at short-term* | Estimated pooled differences and 95% CI | |||||
| Glucosamine vs placebo | 0.60 (−1.80 to 3.00) | 0.91 (−1.91 to 3.75) | 0.98 (−1.94 to 3.91) | −0.43 (−4.44 to 3.58) | 0.59 (−11.79 to 12.98) | |
| p Values for treatment–subgroup interactions | ||||||
| Pain subgroup† | 0.77 | 0.97 | 0.80 | 0.17 | −‡ | |
| BMI subgroup§ | 0.31 | 0.62 | 0.56 | 0.41 | 0.89 | |
| Sex subgroup¶ | 0.68 | 0.59 | 0.68 | 0.86 | 0.68 | |
| KL subgroup** | 0.75 | – | – | – | – | |
| Inflammation subgroup†† | 0.92 | – | – | – | – | |
| Pain at long-term* | Estimated pooled differences and 95% CI | |||||
| Glucosamine versus placebo | 0.98 (−1.76 to 3.73) | 0.19 (−2.83 to 3.22) | 0.78 (−4.33 to 5.89) | 1.22 (−1.90 to 4.33) | −0.38 (−3.67 to 2.90) | |
| p Values for treatment–subgroup interactions | ||||||
| Pain subgroup† | 0.26 | 0.28 | 0.42 | 0.44 | 0.86 | |
| BMI subgroup‡ | 0.55 | 0.10 | 0.51 | 0.72 | 0.10 | |
| Sex subgroup§ | 0.46 | 0.53 | 0.75 | 0.52 | 0.77 | |
| KL subgroup¶ | 0.72 | 0.40 | – | – | – | |
| Inflammation subgroup** | 0.23 | – | – | – | – | |
| Function at short-term‡‡ | Estimated pooled differences and 95% CI | |||||
| Glucosamine versus placebo | 1.74 (−0.45 to 3.96) | 1.80 (−0.81 to 4.04) | 1.92 (−0.77 to 4.61) | 1.23 (−2.11 to 4.57) | −0.39 (−10.88 to 10.09) | |
| p Values for treatment–subgroup interactions | ||||||
| Pain subgroup† | 0.47 | 0.34 | 0.37 | 0.69 | -‡ | |
| BMI subgroup‡ | 0.87 | 0.83 | 0.64 | 0.38 | 0.12 | |
| Sex subgroup§ | 0.47 | 0.30 | 0.39 | 0.91 | 0.34 | |
| KL subgroup¶ | 0.96 | – | – | – | – | |
| Inflammation subgroup** | 0.37 | – | – | – | – | |
| Function at long-term‡‡ | Estimated pooled differences and 95% CI | |||||
| Glucosamine versus placebo | 1.40 (−1.27 to 4.06) | 0.63 (−2.31 to 3.58) | 0.85 (−4.43 to 6.13) | 2.02 (−0.82 to 4.86) | 0.62 (−2.29 to 3.52) | |
| p Values for treatment-subgroup interactions | ||||||
| Pain subgroup† | 0.49 | 0.38 | 0.55 | 0.94 | 0.91 | |
| BMI subgroup‡ | 0.82 | 0.42 | 0.65 | 0.56 | 0.68 | |
| Sex subgroup§ | 0.72 | 0.61 | 0.80 | 1.00 | 0.94 | |
| KL subgroup¶ | 0.83 | 0.77 | – | – | – | |
| Inflammation subgroup** | 0.46 | – | – | – | – | |
*Measured using WOMAC pain (0–100) and adjusted for age sex, BMI, WOMAC pain at baseline and study number.
†WOMAC pain <70 versus ≥70 on a 0–100 scale.
‡Too few cases in high pain group for the software to test the interaction term.
§BMI <27 kg/m2 versus ≥27 kg/m2.
¶Male versus female.
**Kellgren and Lawrence grades 0–2 versus 3–4 (not available in McAlindon et al).24
††Measured using WOMAC function (0–100) and adjusted for age sex, BMI, WOMAC function at baseline and study number. Positive estimated pooled differences indicate a greater reduction in the outcome in the glucosamine group compared with the placebo group.
‡‡Presence of inflammation, defined as presence of swelling/effusion on clinical examination or an elevated ESR, defined as ESR ≥20 mm/h for men aged ≥50 years, ESR ≥15 mm/h for men aged <50 years, ESR ≥30 mm/h for women aged ≥50 years and ESR ≥20 mm/h for women aged <50 years, versus absence of inflammation (not available in McAlindon et al24 and Fransen et al).16
BMI, body mass index; ESR, erythrocyte sedimentation rate; GH, glucosamine hydrochloride; GS, glucosamine sulfate; KL, Kellgren and Lawrence grade; OA, osteoarthritis.