Evidence Tables: Glucosamine and Chondroitin Studies

Roger Chou; Marian S McDonagh; Erika Nakamoto; Jessica Griffin

Trials

Author Year	Eligibility Criteria	Demographics (Age, Gender, Race)	Study Design/Type	Interventions (Drug, Dose, Duration)	Run-in/Washout Period	Allowed Other Medications/Interventions	Other Population Characteristics (Diagnosis, etc)	Number Screened/Eligible/Enrolled	Number Withdrawn/Lost to fu/Analyzed	Results	Adverse Effects Assessment: Pre-Specified, Active or Passive Ascertainment, Measured the Severity of Adverse Effect?	Adverse Effects Reported	Total Withdrawals; Withdrawals due to Adverse Events	Notes
Herrero-Beaumont, 2007²⁰⁶ (GUIDE)	Male and female outpatients, diagnosed with primary symptomatic knee OA in 1 or both knees according to the American College of Rheumatology criteria. Grade II or III on the Kellgren/Lawrence radiographic system. Discouraged enrollment of obese patients. Excluded patients with inflammatory joint disease.	Age: Mean age NR overall Placebo: 64.5 +/− 7.2 Acetaminophen: 63.8 +/− 6.9 Glucosamine sulfate: 63.4 +/− 6.9 Female: 278/318 (87.4%) Placebo: 89/104 (86%) Acetaminophen: 93/108 (86%) Glucosamine: 96/106 (91%) Race/Ethnicity NR	RCT	Glucosamine: 1500 mg glucosamine sulfate, oral solution, once daily. Rottapharm. Acetaminophen side comparator: 1 gram tablets 3 times per day Placebo 6 month treatment duration	Narcotic, non-narcotic analgesics or anti-inflammatory symptomatic medications including topical agents were discontinued for the duration of at least 5 half-lives or 72 hours, whichever was longer. Recommended washout for corticosteroids was 3 months and was 6 months for glucosamine or other drugs considered specific for OA.	Ibuprofen 400mg tablets as rescue medication. Physical and/or occupational therapy were allowed if the regimen had been stable for at least 3 months prior to randomization.	Duration of knee OA: 7.4+/−6.0 vs. 6.5 +/−5.3 vs. 7.2+/−5.8 Baseline Lequesne index: 11.0+/−3.1 vs. 11.1+/−2.7 vs. 10.8+/−2.6 Baseline Kellgren/Lawrence grade: Grade 2: 50% vs. 56% vs. 52% Grade 3: 41% vs. 31% vs. 36% Grade 2/3 unspecified: 9% vs. 12% vs. 11% Baseline WOMAC: Total: 38.3+/−15.2 vs. 40.4+/−14.8 vs. 37.9+/−14.3 Pain: 7.8+/−3.0 vs. 8.0+/−2.9 vs. 7.9+/−3.0 Function: 27.8+/−11.4 vs. 29.4+/−11.0 vs. 27.2+/−10.9	334 screened 325 randomized 7 excluded with no efficacy data 318 ITT population	4 adverse events; 7 Lack of efficacy; 5 loss to followup;12 Protocol violations Analyzed 78 protocol completers. 106 ITT population. 12 adverse events; 5 Lack of efficacy; 3 loss to fu; 8 protocol violations Analyzed 80 protocol completers. 108 ITT population 9 adverse events; 8 Lack of efficacy; 5 Loss to fu; 12 Protocol violations Analyzed 70 protocol completers 104 ITT population	Comparisons to Placebo. No head-to-head. 6 month change in Lequesne Index from baseline −3.1 (−3.8, −2.3); p=0.032 NS: −2.7 (−3.3,−2.1); p=0.18 −1.9 (−2.6, −1.2) 6 month change in WOMAC from baseline Total: −12.9 (−15.6, −10.1); p=0.039 NS: −12.3 (−14.9, −9.7); p=0.08 −8.2 (−11.3,−5.1) Pain: NS: −2.7 (−3.3, −2.1); p=0.12 NS: −2.4 (−3.0, −1.8); p=0.41 −1.8 (−2.6, −1.1) Function: −9.2 (−11.2, −7.2); p=0.022 −8.7 (−10.6, −6.8); p=0.049 −5.5 (−7.7, −3.3) OARSI-A responders: 39.6 (p=0.004) 33.3 (p=0.047) 21.2 OARSI-B, Pain MCII, Function MCII, Pain PASS, Function PASS also reported as secondary outcomes Per-protocol Completers- For all 3 treatments, the degree of improvement in per-protocol completers was higher than that in the ITT population.	Pre-specified: For non- lab AEs: No (general question): For lab AEs: Yes, laboratory tests including measurement of serum glucose and liver function tests were preformed at enrollment, 3 months and 6 months of treatment. Active or passive ascertainment: Active- asked a non leading question during clinic visits and drew labs Assessment of severity: Yes, MedDRA	A vs. B vs. C Total AEs: 95 vs. 96 vs. 89 Symptoms occurring in at least 3 patients during treatment: Dyspepsia: 5 vs. 2 vs. 4 Abdominal pain: 3 vs. 4 vs. 4 Diarrhea: 3 vs. 4 vs. 4 Respiratory tract infections: 8 vs. 4 vs. 9 Gastroenteritis: 4 vs. 0 vs. 2 Coughing and associated symptoms: 1 vs. 4 vs. 0 Headache: 2 vs. 6 vs. 4 Dizziness: 1 vs. 4 vs. 1 Back pain: 7 vs. 4 vs. 5 Neck pain: 3 vs. 2 vs. 0 Fall: 5 vs. 3 vs. 2 Injury: 2 vs. 4 vs. 0 Laboratory: Liver function (transaminases and/or GGT) : 2 vs. 21 vs. 6 Glucose: no change	Withdrawal due to AEs: 4 vs. 12 vs. 9
Kahan, 2009²⁰⁹	Male and female outpatients 45–80 years, primary knee OA of the medial tibiofemoral compartment diagnosed according to ACR.	Chondroitin Sulfate: Age: 62.9 ± 0.5; Female: 70%; Race: NR Placebo: Age: 61.8 ± 0.5; Female: 67%; Race: NR	RCT	Chondroitin Sulfates 4&6 800 mg sachet daily, every evening with glass of water Placebo sachet daily, every evening with glass of water 2 years	24 hours for acetaminophen, 5 days for NSAIDs prior to symptom assessments	Acetaminophen in 500 mg tablets (max dosage 4 gm/day) NSAIDs in cases of acute pain	Duration of knee OA: Left knee: 6.1 ± 0.3 vs. 6.5 ± 0.4; Right knee: 6.6 ± 0.4 vs. 6.3 ± 0.4 KL grade 1: 17.4% vs. 19.7%; KL grade 2: 26.2% vs. 21.6%; KL grade 3: 56.4 vs. 58.7% Minimum JSW, mm: 3.73 ± 0.08 vs. 3.81 ± 0.07 Pain score, 100 mm VAS: 57.2 ± 0.9 vs. 57.3 ± 1.0 WOMAC score, normalized 100mm scales: Total: 40.5 ±1.2 vs. 41.6 ± 1.2; Pain: 40.0±1.2 vs. 40.5±1.2; Function: 39.2±1.3 vs. 39.0±1.2; Stiffness: 42.3±1.5 vs. 43.5±1.5	1052/NR/622	103 vs. 96 withdrawals/18 vs. 18 lost to followup/ITT analysis 622	Interaction between time and treatment effect, indicating that the effect of treatment significantly increased over time (p<0.01) Decrease in minimum JSW loss: −0.07 ± 0.03 vs. −0.31 ±0.04, median effect of treatment 0.14mm (0.06–0.21mm), p<0.0001 Percentage of patient with radiographic progression: 28% vs. 41%, p<0.0005. Relative risk reduction: 33% (16%, 46%) Reduction in minimum JSW loss at 2 years: −0.11 ± 0.04mm vs. −0.39±0.04 mm. treatment effect= 0.20mm (0.11,0.30 mm), p<0.0001 Percentage of responder patients at 6 months: reduction in VAS pain score of at least 40%: 53% vs. 45%, p=0.04 reduction in VAS pain score of at least 60%: 41% vs. 32%, p=0.03 reduction in VAS pain score of at least 40mm: 28% vs. 19%, p=0.01 reduction in VAS pain score of at least 60mm: 9% vs. 4%, p<0.01 decreased WOMAC of at least 40%: 41% vs. 34%, p=0.05 patient assessed VAS at 6 months: 42.2 ± 1.8mm vs. 36.6 ± 1.7mm, p<0.02 doctor assessed VAS at 6 months: 39.6 ± 1.6mm vs. 34.8±1.7mm, p<0.04	Pre-specified: NR Active or passive ascertainment: NR Severity: NR	Gastrointestinal side effects were the most frequently reported, 6% vs. 5.9% No significant laboratory abnormalities	103 vs. 96 withdrawals. 16 vs. 17 withdrawals due to AE
Mazieres, 2007²¹⁰	Male and female outpatients 50–80 years with medial OA, defined according to ACR criteria. Patients with symptomatic knee OA that had lasted for >6 months, with pain during daily activity ≥ 40 mm on a 0–100 mm visual analogue scale, a Lequesne’s Index Score of between 6 and 12, and Kellgren/Lawrence grade 2 or 3 on an anterior-posterior view in an extended standing position taken within the previous 6 months. Exclusions: secondary knee OA, isolated patella-femoral OA and those requiring knee surgery in the coming year, know hypersensitivities to CS or paracetamol, NSAID use for >50% of the time during the previous 2 months, NSAID use within 48 hours before inclusion or SYSADOA, steroid by any route, intra-articular hyaluronic acid or arthroscopic debridement within 6 months before inclusion	CS: Age: 66 (8.8) Female 71% Race: NR Placebo: Age: 66 (7.7) Female: 69% Race: NR	RCT	Chondroitin Sulfate 500mg, twice daily by oral route Placebo, twice daily by oral route 24 weeks	NR	Start with paracetamol (up to 4 gm/day). NSAIDS allowed if paracetamol was not effective. NSAIDs not allowed 2 days and paracetamol not allowed 12 hours prior to evaluation visits.	Duration of disease: 6.2 (6.8) vs. 6.6 (7.6) VAS pain during activity: 62 (13) vs. 61 (12) VAS pain at rest: 40 (20) vs. 40 (22) Lequesne’s Index: 9.5 (2) vs. 9.4 (2) KL stages 2–3: 69% vs. 59%	322/NR/307 (153 CS, 154 Placebo)	14 vs. 14 withdrawals during treatment period, 12 vs. 11 withdrawals during washout period. 307 ITT population	Pain During Activity: VAS, mm; Mean (SD) Week 0: 61 (13) vs. 61 (13) Week 4: 48 (21) vs. 51 (20) Week 12: 40 (23) vs. 42 (21) Week 24: 36 (24) vs. 41 (23) Week 32: 33 (23) vs. 40 (24) Change from baseline to week 24: –26.2 (24.9) mm vs. −19.9 (23.5) mm, p= 0.029 Lequesne’s Index: Mean (SD): Week 0: 9.5 (2.1) vs. 9.4 (1.8) Week 4: 8.3 (2.8) vs. 8.4 (2.4) Week 12: 7.8 (3.6) vs. 7.9 (3.1) Week 24: 7.2 (3.7) vs. 7.7 (3.3) Week 32: 6.8 (3.9) vs. 7.5 (3.6) Change from baseline to week 24: −2.4 (3.4) vs. −1.7 (3.3), p=0.109. OMERACT-OARSI responders: 68% vs. 56% (p=0.03) Change in pain at rest (VAS; mm): −18.8 (23.8) vs. −16.6 (24.2), NS Patient’s global assessment: 3.1 (3.0) vs. 2.5 (3.1), NS Investigator’s global assessment: 3.1 (2.7) vs. 2.5 (3.0), p=0.044 Consumption of analgesics (days): 28 (29) vs. 28 (32), NS Consumption of NSAIDs (days): 6.9 (20.2) vs. 9.2 (24.6), NS QOL, mental: 1.2 (10.4) vs. 0.3 (11.3), NS QOL, physical: 5.8 (9.0) vs. 3.8 (10.2), p=0.021 Carry over effect: changes at the end of the follow-up (week 32) compared to the end of the treatment period (week 24): Change in pain on activity −1.9 (20.9) vs. −0.4 (18.7), NS Change in Lequesne’s index: −0.4 (2.3) vs. −0.2(2.6), NS	Pre-specified: No Active ascertainment: requested at visits Severity: NR	Total Number of AEs: 141 vs. 155, majority were gastro-intestinal troubles including dyspepsia, nausea, vomiting, abdominal pain and diarrhea. Patients with at least one AE: 49% vs. 49% Patients with at least on SAE: 6.5% vs. 5.2%, one in each group was considered related to treatment, eczema and urticaria	Total withdrawals: 26 vs. 25 due to AE: 13 vs. 8
Michel, 2005²¹¹	Male and female patients 40–85 years with clinically symptomatic knee OA (knee pain while standing, walking, and/or on motion for at least 25 of the 30 days prior to study entry) diagnosed according to the ACR clinical and radiographic criteria for OA of the knee. Exclusion criteria: Kellgren/Lawrence grade 4, any causes of secondary OA, traumatic knee lesions, severe comorbidity (severe renal, heart, lung, or neurologic disease), previous joint surgery, intraarticular medications, including corticosteroids into he last month, and the foreseeable prospect of major surgery during the 2- year study period.	Chondroitin Group: Mean age: 62.5 ± 9.1 Female: 51% Race: NR Placebo Group: Mean age: 63.1 ± 10.7 Female: 52% Race: NR	RCT	Chondroitin Sulfates 4 & 6, 800mg tablet daily Placebo 2 years	3 month washout required for potentially longer acting substances such as Chondroitin Sulfate and Glucosamine	Acetaminophen in 500-mg tablets at a maximum dose of 3 gm/day. Secondary rescue with NSAIDs were allowed up to a maximum 5 consecutive days if the primary rescue analgesia with acetaminophen was insufficient. Physical therapy was limited to application of warmth and strengthening exercises No other interventions allowed	ITT Group: Minimum JSW, mm: 2.41 ± 0.14 vs. 2.35 ± 0.14 Mean JSW, mm: 3.04 ± 0.14 vs. 3.00 ± 0.15 WOMAC score, range 0–10: Total: 2.3 ± 1.6 vs. 2.6 ± 1.7 Pain: 2.5 ± 1.6 vs. 2.7 ± 1.8 Function: 2.1 ± 1.6 vs. 2.5 ±1.8 Stiffness: 3.0 ± 2.3 vs. 3.5 ± 2.5	341/300/300	40 vs. 41 withdrawals during treatment 300 ITT analysis	A vs. B, at 2 years JSN Minimum: 0.045 ± 0.48 vs. −0.07 ± 0.56, difference: 0.12 (95% CI 0.00 to 0.24), p=0.05 JSN Mean: 0.00 ± 0.53 vs. −0.14 ± 0.61, difference 0.14 (95% CI 0.01 to 0.27), p =0.04 NS changes in WOMAC: Total: −3.9% vs. 2.1% Pain: −11.0% vs. −6.2% Stiffness: −7.8% vs. −4.6% Function: −0.8% vs. 5.9%	Pre-specified: No Active ascertainment Assessment of severity: No	AEs with frequencies of at least 5% in one of the two study groups: Upper respiratory tract infection: 29% vs. 31% Headache: 7% vs. 9% Abdominal pain: 4% vs. 11% Allergic episode: 6% vs. 6% Cardiac problem: 6% vs. 5% Urinary tract infection: 5% vs. 5%	9 vs. 9 withdrawals due to AE 2 events judged to be related to Chondroitin: abdominal pain and nausea in 1 patient each.
Messier, 2007²¹³	Males and females ≥ 50 years with radiographic evidence of mild to moderate knee OA, Kellgren-Lawrence grade II–III; radiographic classification criteria or confirmation of mild to moderate radiographic evidence of knee OA from a personal physician; not participating in any other intervention study.	Mean Age Overall NR GH/CS: 70.0 ± 1.28 Placebo: 74.1 ± 1.32, p0.03 Female: GH/CS: 75.6% Placebo: 65.9% Race, GH/CS vs. Placebo: Caucasian: 68.9% vs. 77.3% African American: 20% vs. 11.4% Asian/Pacific Islander: 6.7% vs. 2.3% Native American: 4.4% vs. 6.8%	RCT with run-in/washout period, Phase 1 treatment. Phase 2 treatment plus exercise.	Glucosamine hydrochloride 1500mg/day and Chondroitin sulfate 1200mg/day taken either once or three times per day Placebo taken either once or three times per day 1 year treatment period	2-week discontinuation of all over-the-counter or prescription medications. Rescue medication with acetaminophen up to 4g per day and any other necessary medications unrelated to OA were permitted.	Rescue medication of acetaminophen up to 4g/day	--	865 screened/435 not interested/341 ineligible 89 randomized	17 withdrawn/89 analyzed using ITT last observation carried forward	Function (WOMAC physical function 0–68) Mean(SE): Baseline: 25.9 (1.7) vs. 21.1 (1.5), p=0.04 6 months: 21.9 (1.1) vs. 22.9 (1.1), NS 12 months: 19.4 (1.2) vs. 20.6 (1.2), NS Pain (WOMAC pain 0–20): Baseline: 7.1 (0.5) vs. 5.9 (0.5), NS 6 months: 6.2 (0.4) vs. 6.2 (0.4), NS 12 months: 6.0 (0.5) vs. 5.18 (0.5), NS 6 minute walk (meters): Baseline: 384.7 (17.6) vs. 398.7 (17.3), NS 6 months: 393.6 (8.0) vs. 396.5 (7.9), NS 12 months: 409.2 (8.7) vs. 410.5(8.6), NS Knee concentric extension strength (N): Baseline: 209.4 (31.2) vs. 163.9 (20.6), NS 6 months: 176.9 (16.3) vs. 202.7 (17.5), NS 12 months: 207.6 (14.1) vs. 209.7 (15.0), NS Knee concentric flexion strength (N): Baseline: 106.0 (16.1) vs. 83.0 (10.9), NS 6 months: 106.1 (7.3) vs. 106.7 (7.8), NS 12 months: 102.9 (7.7) vs. 124.8 (8.3), p=0.05 Balance (foot length): Baseline: 0.52 (0.04) vs. 0.53 (0.03) 6 months: 0.523 (0.014) vs. 0.583 (0.017), p=0.01 12 months: 0.538 (0.017) vs. 0.591 (0.020), p=0.05 During Phase II: Pill compliant GH/CS group had less pain than the non-compliant group (p=0.02) and a nonsignificant trend in function (p=0.06)	Pre-specified: NR Active or passive: NR Severity: NR		17 withdrawals, 0 due to adverse events 1 AE reported: Hair loss	Groups differ at baseline on age, BMI, gender, annual household income and WOMAC function
Moller, 2010²¹²	Males and females ≥ 40 years of age, with OA of the knee as defined by criteria of the American College of Rheumatology, with pain in the affected knee scoring ≥ 30 on Huskisson’s VAS, and a confirmatory knee X-ray diagnosis, Kellgren-Lawrence grades I–III, associated to cutaneous plaque-type psoriasis with a Psoriasis Area and Severity Index score of >5. Exclusion criteria were Kellgren-Lawrence grade IV, VAS ≥ 30 due to pain of any cause in other sites, non- plaque type psoriasis forms, concurrent arthritic conditions that could confound evaluation of the index joint, presence of any clinically significant cutaneous disease that may interfere with the assessment of lesions during the study and presence of any medical condition judged by the investigator to preclude the patient’s inclusion in the study.	Age (mean ± SD): Overall: 59.8±10.8 CS: 58.6±11.4 Placebo: 61.0±10.4 Gender (% female): Overall: 52.6% CS: 48.3% Placebo: 57.1% Race: NR	Randomized controlled trial	Chondroitin Sulfate 800mg daily Placebo 3 months duration	Washout periods: 6 months for intra-articular hyaluronic acid; 3 months for intra-articular corticosteroids and SYSADOAs; 1 month for oral corticosteroids, 1 week for oral NSAIDs; 1 month for high-potency topical corticosteroids, psoralen photochemotherapy and systemic treatment for psoriasis; 2 weeks for ultraviolet and topical treatment for psoriasis.	Acetaminophen allowed for osteoarthritic symptoms. Syndet soap and moisturizing body milk for daily skin care were provided by the study.	Kellgren-Lawrence grade I: A: 6.7%, B: 7.1% Kellgren-Lawrence grade II: A: 78.3%, B: 75.0% Kellgren-Lawrence grade III: A: 15.0%, B: 21.4% Pain intensity, VAS score, mm, mean ± SD: A: 58.1 ± 16.7, B: 56.8 ± 16.0	181 screened, 129 randomized	4- No data for at least one follow-up 2- Non compliance 60- ITT population analyzed 58- Per protocol population analyzed 5- No data for at least one follow-up 3- No fulfillment of inclusion criteria 3- Non-compliance 2- Use of medication other than study drug or paracetamol 56- ITT population analyzed 51- Per protocol population analyzed	Pain intensity, VAS, mm, mean ± SD: Baseline: A: 58.8 ± 1.7, B: 56.8 ± 15.3 1 month: A: 43.5 ± 2.8, B: 50.3 ± 2.4; Mean difference: −6.8; 95% CI −13.3 to −0.3; p = 0.041 2 months: A: 36.5 ± 2.7, B: 42.0 ± 2.8; Mean difference: −5.5; 95% CI −13.3 to 2.3; p = 0.165 3 months: A: 31.3 ± 2.8, B: 43.2 ± 2.9; Mean difference: −11.8; 95% CI −19.9 to −3.7; p = 0.0004 Lequesne index, mean ± SD: Baseline: A: 9.0 ± 3.5, B: 9.9 ± 3.5 1 month: A: 7.5 ± 0.3, B: 7.3 ± 0.3; Mean difference: 0.19, 95% CI −0.8 to 1.1, p = 0.700 2 months: A: 5.4 ± 0.4, B: 6.3 ± 0.4; Mean difference: −0.93, 95% CI −2.1 to 0.2, p = 0.109 3 months: A: 4.5 ± 0.5, B: 6.1 ± 0.5; Mean difference: −1.7, 95% CI −3.0 to −0.4, p = 0.013 Acetaminophen, number pills/month, mean ± SD: 1 months: A: 29.5 ± 31.4, B: 29.5 ± 29.6; Mean difference: 3.4, 95% CI 22.7 to 36.3, p = 0.991 2 months: A: 32.3 ± 33.9, B: 28.8 ± 28.2; Mean difference: 3.9, 95% CI 22.6 to 38.7, p = 0.668 3 months: A: 38.2 ± 42.6, B: 30.2 ± 33.8; Mean difference: 5.1, 95% CI 23.1 to 43.7, p = 0.453	Prespecified: Yes for laboratory blood tests, NR for others Active ascertainment for blood tests, NR for other Severity NR		2 withdrawals, 0 for adverse events 13 withdrawals, 0 for adverse events
Rozendaal, 2008²⁰⁷	Patients met the American College of Rheumatology clinical criteria for hip osteoarthritis and were able to complete questionnaires in Dutch. Excluded patients who had undergone or were awaiting hip replacement surgery, Kellgren and Lawrence score of 4, renal disease, liver disease, diabetes mellitus, or a disabling comorbid condition that would make visits to the research center impossible, patients receiving glucosamine.	Age: Mean age NR overall Placebo: 63.7 (8.5) Glucosamine sulfate: 63.1 (9.5) Female: Placebo: 70.3% Glucosamine: 68.5% Race/Ethnicity NR	RCT	1500mg oral glucosamine sulfate, administered once daily or as two 750 mg tablets Placebo 24 months treatment duration	NR	Baseline Pain Med use: Placebo overall: Daily 18.9% Sometimes: 27.9% None: 53.2% Glucosamine overall: Daily: 28.8% Sometimes: 25.2% None: 46.0% Interventions NR, except Total Hip Arthroplasty was collected and used in analyses.	Kellgren and Lawrence Score (%): 1: 49.5 vs. 53.2 ≥2: 50.5 vs. 46.8 Mean minimum JSW (SD), mm: 2.13 (1.00) vs. 2.33 (0.90) Mean WOMAC score (SD): Pain: 35.9 (23.0) vs. 32.4 (23.2) Function: 36.0 (24.1) vs. 34.1 (21.7) Stiffness: 44.2 (27.2) Mean pain in past week (SD), mm: 34.3 (26.5) vs. 30.5 (25.2)	Screened: 387 Eligible & Randomized: 222	Withdrawals during treatment period: NR Lost to follow-up: 7 vs. 8 ITT analysis: 111 vs. 111	Primary Outcomes: WOMAC (negative difference favors glucosamine): Pain overall (SE): −1.90 ± 1.6 vs. −0.30 ± 1.6; Unadjusted difference: −1.60 (−5.60, 2.40); Adjusted difference: −1.54 (−5.43, 2.36) Function overall (SE): −1.69 ± 1.3 vs. 0.38 ± 1.3; Unadjusted difference: −2.07 (−5.53, 1.39); Adjusted difference: −2.01 (−5.38, 1.36) JSN, mm (positive difference favors glucosamine sulfate): Minimal: −0.094 (0.32) vs. −0.057 (0.32); Unadjusted difference: −0.038 (−0.130, 0.055); Adjusted difference: −0.029 (−0.122, 0.064) Lateral: −0.180 (0.34) vs. −0.159 (0.36); Unadjusted difference: −0.020 (−0.124, 0.083); Adjusted difference: −0.017 (−0.121, 0.088) Superior: −0.123 (0.36) vs. −0.129 (0.30); Unadjusted difference: 0.006 (−0.090, 0.101); Adjusted difference: 0.016 (−0.079, 0.111) Axial: −0.070 (0.48) vs. −0.079 (0.30); Unadjusted difference: 0.009 (−0.108, 0.124); Adjusted difference: −0.005 (−0.118, 0.108) Secondary Outcomes: WOMAC (Negative difference favors glucosamine): Pain, 3mos. −2.50 (19.2) vs. −1.79 (16.2); Unadjusted difference: −0.71 (−5.47, 4.05); Adjusted difference: 0.06 (−4.11, 4.22). 12 mos. −0.54 (19.9) vs. −0.89 (23.3); Unadjusted difference: 0.35 (−5.66, 6.36); Adjusted difference: 1.42 (−3.82, 6.67). 24 mos. −1.47 (20.7) vs.0.88 (26.4); Unadjusted difference: −2.34 (−9.16, 4.48); Adjusted difference: −0.77 (−6.53, 4.98) Function, 3 mos. −3.29 (14.9) vs. −1.08 (12.7); Unadjusted difference: −2.22 (−5.97, 4.05); Adjusted difference: −2.04 (−5.48, 1.40). 12 mos. −0.98 (14.9) vs. −0.88 (17.6); Unadjusted difference: −0.11 (−4.63, 4.42); Adjusted difference: 0.11 (−4.14, 4.35). 24 mos. −0.84 (19.1) vs. 1.92 (19.7); Unadjusted difference: −2.76 (−8.35, 2.84); Adjusted difference: −1.63 (−6.73, 3.47). Stiffness, 3 mos. −4.59 (22.6) vs. −3.39 (17.7). Unadjusted difference: −1.20 (−6.66, 4.26); Adjusted difference: −0.12 (−4.94, 4.71). 12 mos. −1.38 (22.1) vs. −3.43 (21.6); Unadjusted difference: 2.06 (−4.00, 8.12); Adjusted difference: 3.11 (−2.07, 8.28). 24 mos. −3.43 (26.2) vs. −2.19 (24.1); Adjusted difference: −1.24 (−8.47, 5.98); Unadjusted difference: 0.66 (−5.27, 6.59). VAS pain also reported.	Prespecified: yes, used a checklist Active ascertainment; used a checklist at baseline and every 3 months Severity measured: NR	Serious Adverse Events: 4 vs. 2 AE resulting in treatment termination: 4 vs. 6 Abdominal pain: 14 vs. 10 Stomach symptoms: 25 vs. 19 Intestinal symptoms: 19 vs. 17 Increased blood pressure: 11 vs. 19 Decreased blood pressure: 4 vs. 3 Fatigue: 24 vs. 18 Headache: 16 vs. 26 Vertigo: 16 vs. 18 Cardiac problems: 6 vs. 9 Depressive mood: 10 vs. 6 Allergic episode: 7 vs. 5	Lost to follow up: 7 vs. 8, withdrawal during treatment NR. Withdrawal of treatment due to AE: 4 vs. 6
Rozendaal, 2009²⁰⁷	See Rozendall, 2008	See Rozendall, 2008	RCT, subgroup analysis of Rozendall, 2008 data Predefined subgroups: KL=1, KL ≥ 2, localized OA, generalized OA Exploratory subgroups: VAS ≤ 30, VAS > 30, No pain medication, pain medication, no knee OA, knee OA, JSN ≥ 2.5mm, <2.5 mm	See Rozendall, 2008	See Rozendall, 2008	See Rozendall, 2008	See Rozendall, 2008	See Rozendall, 2008	See Rozendall, 2008	The predefined subgroup analyses based on radiographic severity of OA and type of OA did not yield differences between GS and placebo in WOMAC pain, function and JSN. The exploratory analyses showed no difference in WOMAC pain, function and JSN. WOMAC Pain ( Negative value favors glucosamine): No Knee OA: 0.3 (21.5) vs. 0.1 (26.2); Unadjusted difference: 0.3 (−7.9, 8.5); Adjusted difference: −0.1 (−4.9, 4.7). WOMAC pain: Concomitant Knee OA: −5.8 (18.1) vs. 2.9 (27.1); Unadjusted difference: −8.7 (−21.2, 3.8); Adjusted difference: −5.68 (−12.62, 1.26).	See Rozendall, 2008	See Rozendall, 2008	See Rozendall, 2008
Sawitzke, 2008²¹⁴ (GAIT)	Males and females ≥ 40 years of age, had knee pain for at least 6 months occurring on the majority of days in the month preceding their enrollment in GAIT, and had Kellgren/Lawrence gade 2 or 3 knee OA determined on a screening AP radiograph of the knee in a weight bearing position. Exclusion: Minimum baseline medial tibiofemoral JSW of <2mm, predominant lateral compartment OA on any film of the MTP joints, history of significant trauma or surgery to the knee	Age (mean ± SD years): Glucosamine: 56.7± 10.4 CS: 56.4± 9.2 Glucosamine + CS: 56.5± 9.9 Celecoxib: 58.3± 10.7 Placebo: 56.6± 8.4 Female (%): Glucosamine: 61.0 CS: 71.8 Glucosamine + CS: 55.9 Celecoxib: 63.8 Placebo: 64.3 Race: NR	Prospective observational study of GAIT enrollees; ancillary study to assess structural changes in knee OA	Glucosamine 500 mg 3 times daily Chondroitin sulfate (400mg 3 times daily) Combination of Glucosamine and Chondroitin Celecoxib 200mg daily Placebo 24 months	NR-check other GAIT pubs	NR- check other GAIT pubs	Kellgren/Lawrence Grade 2, %: 80.5 vs. 81.0 vs. 69.2 vs. 72.6 vs. 80.5 Kellgren/Lawrence Grade 3, %: 19.5 vs. 19.0 vs. 30.9 vs. 27.4 vs. 19.5	662 GAIT participants consented to this study	(177 initial): 33/NR/77 (123 initial): 30/NR/71 (128 initial): 40/NR/59 (143 initial): 32/NR/80 (134 initial): 36/NR/70	Mean loss in JSW over 2 years: All NS 0.013 vs. 0.107 vs. 0.194 vs. 0.111 vs. 1.166 Difference from placebo (negative value = less JSW loss): −0.153 (−0.379, 0.074) vs. −0.059 (−0.287, 0.169) vs. 0.028 (−0.214,0.271) vs. −0.055 (−0.279, 0.170) Disease progression over 2 years, % of patients: All NS 18.6 vs. 21.4 vs. 24.4 vs. 20.2 vs. 22.4 OR vs. placebo for disease progression: 0.79 (0.48,1.3) vs. 0.94 (0.57,1.55) vs. 1.12 (0.67,1.88) vs. 0.87 (0.53,1.43)	NR- check earlier GAIT pub		Withdrawals: 33 vs. 30 vs. 40 vs. 32 vs. 36 Technical Loss: 9 vs. 6 vs. 11 vs. 10 vs. 8 Withdrawals due to AE: see earlier GAIT report
Sawitzke, 2010²¹⁵	Males and females ≥ 40 years of age, had knee pain for at least 6 months occuring on the majority of days in the month preceding their enrollment in GAIT, and had Kellgren/Lawrence gade 2 or 3 knee OA determined on a screening AP radiograph of the knee in a weight bearing position. Exclusion: Minimum baseline medial tibifemoral JSW of <2mm, predominant lateral compartment OA on any film of the MTP joints, history of signicant trauma or surgery to the knee	Age (mean ± SD years): Glucosamine: 56.7± 10.5 CS: 56.3± 8.8 Glucosamine + CS: 56.7± 10.7 Celecoxib: 57.6± 10.6 Placebo:56.9± 9.8 Female (%): Glucosamine: 68.7 CS: 73.0 Glucosamine + CS: 65.1 Celecoxib: 65.5 Placebo: 65.7 Race: NR	Prospective observational study of GAIT enrollees; ancillary study to assess structural changes in knee OA	Glucosamine 500mg 3 times daily Chondroitin sulfate (400mg 3 times daily) Combination of Glucosamine and Chondroitin Celecoxib 200mg daily Placebo 24 months	NR	NR	Kellgren/Lawrence Grade 2, %: 59.7 vs. 66.7 vs. 51.9 vs. 62.0 vs. 61.1, p=0.19 Duration of OA symptoms, mean years (SD): 9.7 (10.3) vs. 9.0 (9.0) vs. 10.0 (9.4) vs. 10.2 (9.2) vs. 10.1 (9.4)	662 GAIT participants consented to this study	See Sawitzke, 2008	Odds of pain response over 24 months versus placebo by WOMAC, OR (95% CI): 1.16 (0.65 to 2.04) 0.69 (0.40 to 1.21) 0.83 (0.51 to 1.34) 1.21 (0.71 to 2.07) reference Odds of a pain response over 24 months versus placebo by OMERACT/OARSI, OR (95% CI): 1.16 (0.74 to 1.83) 0.89 (0.53 to 1.50) 0.85 (0.55 to 1.31) 1.45 (0.86 to 2.42) reference Change in WOMAC pain and function score over 24 months	NR in ancillary study		NR in ancillary study
Wilkens, 2010²⁰⁸	INCLUSION: Nonspecific chronic LBP (defined as the area below the 12th rib and above the gluteal folds); LBP for at least 6 months with summed score of at least 3 out of 24 points on the Roland Morris Disability Questionnaire, older than 25 years of age. Patients with concomitant leg pain were included as long as the LBP pain rating was higher than the leg pain rating. MRI scans no older than 1 year prior to inclusion consisting of at least 1 axial view and 2 sagittal views were required. MRI confirmed degenerative process. At least one of the following MRI criteria: disk signal intensity changes, reduced disk height compared with adjacent superior disk, facet joint changes, modic changes, or high-intensity zone. EXCLUSION: symptomatic intervertebral disk herniation or spinal stenosis, previous lumbar fracture or surgery, pregnancy or breastfeeding, seafood allergy, ongoing psychiatric or somatic disease potentially influencing a patient’s pain and use of any type of glucosamine 1 year prior to enrollment.	Age; mean (SD): Total: 48.5 (11.24) Glucosamine: 47.5 (11.5) Placebo: 49.4 (11.0) Female: Total: 121/250 (48.4%) Glucosamine: 54/125 (43.2%) Placebo: 67/125 (53.6%) Race: NR	RCT	Glucosamine sulfate 1500mg or placebo administered as three 500-mg capsules per day. Could be taken as one pill 3 times per day or all at once. Placebo 6 month treatment period	NR	Rescue medication: Pain killers or NSAIDs, existing analgesics, or usual LBP therapy (e.g., manipulation, physiotherapy, massage)	Mean (SD) Roland Morris Disability Questionnaire (RMDQ) (0–24): 9.2 (3.9) vs. 9.7 (4.5) Numeric Rating Scale (NRS) (0–10): LBP at rest: 3.7 (2.6) vs. 3.9 (2.4) Leg pain at rest: 1.8 (2.2) vs. 2.0 (2.3) LBP when active: 4.9 (2.5) vs. 5.1 (2.3) Leg pain when active: 2.4 (2.6) vs. 2.7 (2.6) EuroQol-5 Dimensions (EQ-5D) (−0.59 to 1.0): 0.57 (0.3) vs. 0.63 (0.2) EuroQol-Visual analog scale (EQ-VAS) (0–100): 5.8 (2.2) vs. 6.4 (2.0)	473 screened/250 randomized and enrolled	Withdrawals during treatment period: 7 vs. 10 Loss to followup: 4 vs. 4 Primary analysis is ITT and includes all 250 randomized patients	Mean SD (95% CI); All results NS: RMDQ (0–24): 6 weeks: 7.0 (6.1, 7.8) vs. 7.1 (6.3, 7.9); 3 months: 5.8 (5.0, 6.6) vs. 6.5 (5.7, 7.3); 6 months: 5.0 (4.2, 5.8) vs. 5.0 (4.2,5.8); 1 year: 4.8 (3.9, 5.6) vs. 5.5 (4.7, 6.4) NRS LBP at rest (0–10): 6 weeks: 2.9 (2.5, 3.3) vs. 2.9 (2.5, 3.3); 3 months: 2.7 (2.4, 3.1) vs. 2.9 (2.5, 3.3); 6 months: 2.5 (2.1, 2.9) vs. 2.4 (2.0, 2.8); 1 year: 2.5 (2.1, 2.9) vs. 2.8 (2.4, 3.1) NRS Leg pain at rest (0–10): 6 weeks: 1.3 (1.0, 1.7) vs. 1.5 (1.2, 1.9); 3 months: 1.4 (1.0, 1.8) vs. 1.7 (1.4, 2.1); 6 months: 1.4 (1.0, 1.7) vs. 1.5 (1.1, 1.8); 1 year: 1.5 (1.1, 1.8) vs. 1.6 (1.3, 2.0) NRS LBP when active (0–10): 6 weeks: 3.7 (3.2, 4.1) vs. 3.6 (3.2, 4.0); 3 months: 3.3 (2.9, 3.7) vs. 3.2 (2.8, 3.6); 6 months: 3.1 (2.7, 3.5) vs. 2.9 (2.5, 3.3); 1 year: 3.0 (2.5, 3.4) vs. 2.9 (2.5, 3.3) NRS Leg pain when active (0–10): 6 weeks: 1.8 (1.4, 2.2) vs. 1.9 (1.5, 2.3); 3 months: 1.7 (1.2, 2.1) vs. 1.9 (1.5, 2.3); 6 months: 1.6 (1.2, 2.0) vs. 1.9 (1.5, 2.3); 1 year: 1.7 (1.3, 2.1) vs. 2.0 (1.5, 2.4) EQ-5D (−0.59 to 1.0): 6 weeks: 0.68 (0.64, 0.72) vs. 0.69 (0.65, 0.72); 3 months: 0.73 (0.70, 0.78) vs. 0.69 (0.65, 0.73); 6 months: 0.74 (0.70, 0.78) vs. 0.76 (0.65, 0.74); 1 year: 0.74 (0.70, 0.78) vs. 0.70 (0.65, 0.74) EQ-VAS (0–100): 6 weeks: 6.8 (6.2, 7.3) vs. 6.7 (6.1, 7.2); 3 months: 7.2 (6.7, 7.8) vs. 6.8 (6.2, 7.3); 6 months: 7.2 (6.6, 7.8) vs. 7.1 (6.7, 7.4); 1 year: 7.4 (7.0, 7.7) vs. 6.6 (6.3, 7.0) Global perceived effect: No. (%): 6 weeks: 22 (18.6) vs. 27 (22.0); 3 months: 26 (21.5) vs. 26 (22.2); 6 months: 39 (33.1) vs. 42 (36.2); 1 year: 14 (30.9) vs. 32 (29.4)	Pre-specified: NR Ascertainment: NR Severity: NR	OR (95% CI) All NS differences AEs resulting in treatment discontinuation: 0.66 (0.48–1.36) All AEs: 0.83 (0.49–1.40) Skin problems: 0.79 (0.35–1.76) Neurological: 0.65 (0.31–1.38) Heartburn: 0.99 (0.06–15.9) Flatulence: 0.55 (0.21–1.44) Abdominal pain: 1.32 (0.29–6.04) Nausea/vomiting: 1.77 (0.50–6.21) Constipation: 4.03 (0.44–36.69) Diarrhea: 0.55 (0.16–1.92) Headache/vertigo: 0.98 (0.28–3.49) Musculoskeletal concerns: 0.42 (0.14–1.25) 10 AEs resolved with treatment discontinuation; 7 resolved with continuation of study drug; 2 Serious AEs (death and surgery) were considered unrelated to study drug. Fasting blood glucose, cholesterol, and blood pressure levels did not deviate from normal fluctuations during the trial	Total during treatment period: 7 vs. 10 Withdrawals due to AE: Glucosamine: 6 vs. 6

: ACR = American College of Rheumatology; AE = adverse event; BMI = body mass index ; CI = confidence interval; CS = chondroitin sulfate; EQ-VAS = EuroQol visual analogue scale; GAIT = Glucosamine/chondroitin Arthritis Intervention Trial; GH = glucosamine hydrochloride; GS = glucosamine sulfate; GUIDE = Glucosamine Unum-in-Die (Once a Day) Efficacy trial; ITT = intention to treat; JSN = joint space narrowing; JSW = joint space width; KL = Kellgren-Lawrence scale; LBP = low back pain; MCII = minimal clinically important improvement; MRI = magnetic resonance imaging; MTP = metatarsophalangeal; NR = not reported; NRS = nonrandomized study; NS = not significant; NSAID = nonsteroidal anti-inflammatory drug; OA = osteoarthritis; OARSI = Osteoarthritis Research Society International; OMERACT-OARSI = Outcomes Measures in Arthritis Clinical Trials-Osteoarthritis Research Society International; PASS = Patient Acceptable Symptom Scale; QOL = quality of life; RCT = randomized controlled trial; RMDQ = Roland Morris Disability Questionnaire; SD = standard deviation; SE = standard error; SYSADOA = Symptomatic Slow Acting Drugs in Osteoarthritis; UTI = urinary tract infection; VAS = visual analogue scale; WOMAC = Western Ontario and McMaster Universities Osteoarthritis Index

Systematic Reviews

Author, Year	Aims	Time Period Covered	Eligibility Criteria	Number of Patients	Characteristics of Identified Articles: Study Designs	Characteristics of Identified Articles: Populations	Characteristics of Identified Articles: Interventions	Main Results	Subgroups
Bjordal, 2007¹⁹⁷	To determine the short-term pain-relieving effects of seven pharmacological agents for OA knee pain	MEDLINE, Embase, PedRo, Cochrane Controlled Trials Register 1996 through November 2005	Diagnosis: Knee OA verified by clinical exam and/or by x ray. If less than 4 trials available for an intervention, trials also including hip OA were considered, if more than 2/3 of their patients had knee OA; Symptom duration: 3 months; Trial designs: Blinded, placebo-controlled parallel groups RCTs; Outcome measures: Pain intensity within 4 weeks of treatment start on WOMAC or on a 100mm VAS for global or walking pain. Pain intensity at 8–12 weeks follow-up; Intervention groups: Identical placebo drug and adequate daily defined drug dosage equal to or exceeding set dosages per drug: paracetamol 4g, diclofenac 100mg, etodolac 400mg, ibuprofen 2400 mg, nabumetone 1500mg, naproxen 1000mg, oxaprozin 1200mg, tiaprofenic acid 600mg, valdecoxib 10mg, celecoxib 200mg, meloxicam 7.5mg, etoricoxib 30mg, lumiracoxib 200mg, rofecoxib 12.5mg, topical diclofenac, piroxicam or meloxicam 1%, ibuprofen gel 3%, triamcinolone 20mg, methylprednisolone 40mg, cortivazol 3.75mg, glucosamine sulfate 1500mg, chondroitin sulfate 800mg, codeine 50mg, oxymorphone 20mg, oxycodone 20mg, morphine sulfate 30mg, tramadol 100mg	14,060 patients for all included drugs. 9964 patients received Oral NSAIDs including coxibs, 749 received topical NSAIDs, 401 received glucosamine sulfate, 362 received chondroitin sulfate	64 RCTs total. 25 RCTs of oral NSAIDs (including coxibs), 9 topical NSAIDs, 7 glucosamine sulfate, 6 chondroitin sulfate	Mean Age: Oral NSAIDs: 62.6 years Topical NSAIDs: 64.2 years Glucosamine sulfate: 58.6 years Chondroitin sulfate: 63.0 years Mean baseline pain on 100mm VAS: Oral NSAIDs: 64.3 Topical NSAIDs: 54.7 Glucosamine sulfate: 57.8 Chondroitin sulfate: 50.7	Trials of included Oral NSAIDs:^* 6 celecoxib studies; 2 naproxen studies; 2 diclofenac studies; 3 etodolac studies; 1 diflunisal study; 1 meloxicam study; 2 nabumetone studies; 1 oxaprozin study Trials of included Topical NSAIDs: 7 diclofenac, 2 eltenac, 1 ibuprofen Trials of glucosamine: 7 Trials of chondroitin: 6	Best mean difference of change over placebo (100mm VAS): Glucosamine: 4.7 (95% CI 0.3 to 9.1) Chondroitin: 3.7 (95% CI 0.3 to 7.0) Glucosamine and chondroitin did not have effect size or 95% CI exceeding the mean threshold for minimal clinical important improvement, slight improvement, or minimal perceptible improvement	NR
Hochberg, 2010¹⁹⁹	To update the 2008 systematic review and meta-analysis with results of an updated meta-analysis that includes data from two recently published studies and limits the pooling to studies of 2- year duration	1996–October 2007	RCTs of 2-year duration that compared orally administered chondroitin sulfate to placebo and reported structural outcomes in the form of change in minimum joint space. No language restriction was applied.	1179	Randomized controlled trials	Michel et al., 2005: mean age 63, 52% women Sawitzke et al., 2008: mean age 57, 68% women Kahan et al., 2009: mean age 62, 68% women	Michel et al., 2005: 800 mg chondroitin sulfate once daily, 24 month duration Sawitzke et al., 2008: 400 mg chondroitin sulfate three times daily, 24 month duration Kahan et al., 2009: 800 mg chondroitin sulfate once daily, 24 month duration	Joint space narrowing (mm ± SD): Michel et al., 2005: CS: −0.045 ± 0.48, PBO: 0.07 ± 0.56; Mean difference (mm (95% CI)): 0.12 (0.00 to 0.23); Effect size (95 % CI): 0.22 (0.01 to 0.45) Sawitzke et al., 2008: CS: 0.107 ± 0.68, PBO: 0.166 ± 0.68; Mean difference (mm (95% CI)): 0.06 (−0.17 to 0.28); Effect size (95% CI): 0.09 (−0.24 to 0.42) Kahan et al., 2009: CS: 0.07 ± 0.03, PBO: 0.31 ± 0.04; Mean difference (mm (95% CI)): 0.14 (0.06 to 0.21); Effect size (95% CI): 0.26 (0.11 to 0.42) Pooled analysis: Mean difference (mm (95% CI)): 0.13 (0.06 to 0.19); Effect size (95% CI): 0.23 (0.11 to 0.35)	NR
Lee, 2010²⁰⁰	To assess the structural efficacies of daily glucosamine sulfate and chondroitin sulfate in patients with knee OA	Through July 2008	English language RCTs that compared glucosamine sulfate or chondroitin sulfate with a placebo in patients with OA, and utilized JSN as an outcome variable after treatment commencement. Studies were excluded if they did not contain a placebo group, if the OA site was not the knee joint, they did not contain adequate data, or if they were cross-sectional.	749	Randomized controlled trials	Pavelka et al, 2002: mean age patients: 61.2, controls: 63.5; 79% female patients, 76% female controls Reginster et al, 2001: mean age patients: 66.0, controls: 65.5; 75% female patients, 78% female controls Kahan et al, 2006: mean age not available; 68% female patients, 68% female controls Michel et al, 2005: mean age patients: 62.5, controls: 63.1; 51% female patients, 52% female controls Uebelhart et al, 2004: mean age patients: 63.2, controls: 63.7; 79.6% female patients, 82.1% female controls Uebelhart et al, 1998: mean age patients: 60.13, controls: 57.11; 47.8% female patients, 56.5% female controls	Pavelka et al, 2002: GS 1,500 mg qd Reginster et al, 2001: GS 1,500 mg qd Kahan et al, 2006: CS 800 mg qd Michel et al, 2005: CS 800 mg qd Uebelhart et al, 2004: CS 800 mg 2 periods of 3 months during 1 year Uebelhart et al, 1998: CS 400 bid	Glucosamine Sulfate: std diff in means (95% CI): Follow-up for 1 year (2 studies): 0.078 (−0.116 to 0.273), p=0.429 Follow- up for 3 years (2 studies): 0.432 (0.235 to 0.628), p=0.000 JSN > 0.5 mm (2 studies): OR 0.361 (0.204–0.640), p=0.000 Chondroitin Sulfate: std diff in means (95% CI): Minimum JSW (3 studies): 0.317 (0.136–0.497), p=0.001 Mean JSW (4 studies): 0.236 (0.148–0.386), p=0.000 Follow- up for 1 year (2 studies): 0.295 (0.000–0.590), p=0.050 Follow- up for 2 years (2 studies): 0.261 (0.131–0.392), p=0.000	NR
Vlad, 2007²⁰¹	To identify factors that explain heterogeneity in trials of glucosamine	1966–2006	Randomized, double-lind, placebo- controlled trials of parenteral or oral glucosmine for pain from OA of the knee or hip, and subjects were followed for >4 weeks.	2613	Randomized controlled trials	Mean age, % female NR	Glucosamine preparation: Clegg et al, 2006: Hydrochloride Herrero- Beaumont et al, 2005: GS Usha and Naidu, 2004: GS McAlindon et al 2004: Hydrochloride and GS Cibere et al, 2004: GS Pevelka et al, 2002: GS Hughes and Carr, 2002: GS Reginster et al, 2001: GS Rindone et al, 2000: GS Houpt et al, 1999: Hydrochloride Reichelt et al, 1994: GS Noack et al, 1994: GS Vajaradul, 1981: GS Pujalte et al, 1980: GS Rovati et al, 1999: GS	Pooled estimates of heterogeneity and pooled effect estimates (95% CI), P for difference: All studies (15 studies): 0.35 (0.14, 0.56) Glucosamine hydrochloride (3 studies): 0.06 (−0.08, 0.20) Glucosamine sulfate (12 studies): 0.44 (0.18, 0.70) Industry funding absent (4 studies): 0.05 (−0.32, 0.41) Industry funding present (11 studies): 0.47 (0.24, 0.70), p=0.05 Industry participation absent (7 studies): 0.11 (−0.16, 0.38) Industry participation present (8 studies): 0.55 (0.27, 0.84), p=0.02 Industry-affiliated author absent (8 studies): 0.16 (−0.11, 0.42) Industry-affiliated author present (7 studies): 0.55 (0.27, 0.84), p=0.04 Rottapharm product absent (7 studies): 0.11 (−0.16, 0.38) Rottapharm product present (8 studies): 0.55 (0.29, 0.82), p=0.01 Allocation concealment adequate (5 studies): 0.09 (−0.24,0.42) intermediate (6 studiess): 0.47 (0.14, 0.80) inadequate (4 studies): 0.54 (0.14, 0.94), p=0.09 No ITT analysis (5 studies): 0.44 (0.03, 0.84) ITT analysis (10 studies): 0.31 (0.05, 0.58), p=0.62 Jadad score 1–3 (4 studies): 0.30 (−0.14, 0.73) Jadad score 4–5 (11 studies): 0.37 (0.11, 0.63) No rescue medication (3 studies): 0.55 (0.01, 1.10) Rescue medication use (12 studies): 0.31 (0.07, 0.55), p=0.42	NR
Wandel, 2010¹⁹⁸	To determine the clinical effect of glucosamine, chondroitin, or the two in combination on joint pain and on radiological progression of disease in OA of the hip or knee	MEDLINE, EMBASE, CINAHL, and Cochrane Controlled Trials Register through June 2010.	Randomized trials with an average of at least 100 patients with knee or hip osteoarthritis per arm. Comparisons included chondroitin sulphate, glucosamine sulphate, glucosamine hydrochloride, or the combination of any two with placebo or head to head. Excluded trial arms with sub-therapeutic doses (<800mg/day of chondroitin, <1500mg/day glucosamine.	3803 to the interventions or placebo. Glucosamine sulphate vs. Placebo: 5 trials, 1104 randomized patients; Glucosamine sulphate or hydrochloride vs. Placebo: 1 trial, 205 patients; Chondroitin sulphate vs. Placebo: 3 trials 1229 patients; Glucosamine hydrochloride, chondroitin sulphate, and their combination vs. placebo: 1 trial, 1265 patients	10 RCTs: designs not specified	8 trials with knee OA only, one trial with hip or knee OA, one trial with hip OA only. Mean age: 58–66 years % Female: 27–86 (median = 68%) Average duration of symptoms: 6 months–10 years	6 glucosamine vs. placebo 3 chondroitin vs. placebo 1 glucosamine, chondroitin, combination vs. placebo	Pain Intensity (10cm VAS): Glucosamine vs. Placebo: −0.4 cm (−0.7 to −0.1) Chondroitin vs. Placebo: −0.3 cm (−0.7 to 0.0) Glucosamine and Chondroitin vs. Placebo: −0.5 cm (−0.9 to 0.0) Radiological joint space difference (negative number favors intervention): Glucosamine vs. Placebo: −0.2 mm (−0.3 to 0.0) Chondroitin vs. Placebo: −0.1mm (−0.3 to 0.1) Glucosamine and Chondroitin vs. Placebo: 0.00 mm (−0.2 to 0.2) Adverse Events, OR (95% CI): Glucosamine vs. Placebo: 0.94 (0.59 to 1.47) Chondroitin vs. Placebo: 0.99 (0.49 to 2.00) Glucosamine and Chondroitin vs. Placebo: no data Withdrawals due to AE, OR (95% CI) Glucosamine vs. Placebo: 0.99 (0.61 to 1.50) Chondroitin vs. Placebo: 0.92 (0.56 to 1.51) Glucosamine and Chondroitin vs. Placebo: 0.90 (0.43 to 1.85)	Estimated differences in pain intensity between supplements and placebo were on average 0.5 cm (0.1 to 0.9) higher in industry sponsored trials (p=0.02 for interaction)

: AE = adverse event; CI = confidence interval; CS = chondroitin sulfate; ITT = intention to treat; GS = glucosamine sulfate; NSAID = nonsteroidal anti-inflammatory drug; OA = osteoarthritis; OR = odds ratio; RCT = randomized controlled trial; VAS = visual analogue scale; WOMAC = Western Ontario and McMaster Universities Osteoarthritis Index

*: Characteristics of oral NSAID trials of included drugs for the current systematic review.

Appendix IEvidence Tables: Glucosamine and Chondroitin Studies

Publication Details

Copyright

Publisher

NLM Citation

Trials

Systematic Reviews