Clinical Inquiries

How do hyaluronic acid and corticosteroid injections compare for knee OA relief?

Author and Disclosure Information


Inconsistent evidence shows a small amount of pain relief early (one week to 3 months) with corticosteroid (CS) injections and an equally small improvement in pain relief and function later (3 to 12 months) with hyaluronic acid (HA) injections (strength of recommendation [SOR]: B, meta-analysis of a randomized controlled trial [RCT] and inconsistent RCTs).

Guidelines state that CS injections can be considered for symptomatic knee osteoarthritis (OA), but that insufficient evidence exists to recommend HA injections (SOR: B, evidence-based guidelines).




A 2015 network meta-analysis of 137 RCTs with 33,243 patients (ages 45-76 years) with knee OA compared the effectiveness of a variety of treatments including intra-articular CS and HA.1 At 3 months, the effect on pain was not significantly different between the CS and HA groups (12 trials; effect size [ES]=0.02; 95% confidence interval [CI], -0.12 to 0.17). However, a small but significant improvement in function was noted (scoring system not defined) at 3 months favoring HA (ES=0.24; 95% CI, 0.06-0.43; number of trials not specified).

At 3 and 6 months, HA improves pain, but not function, more than CS

Another meta-analysis published in 2015 examined the effectiveness of intra-articular CS and HA in 7 RCTs with 583 patients with knee OA.2 All 7 trials were included in the network meta-analysis and discussed separately to evaluate different time points.

Pain at one month wasn’t significantly different using a visual analog score (VAS) of one to 100 (4 trials; 245 patients; mean difference [MD]=1.66 points; 95% CI, -0.90 to 4.23). At 3 and 6 months, the HA group reported significantly reduced pain compared with the CS group (3 months: 3 trials; 320 patients; MD=12.58 points; 95% CI, -17.76 to -7.40; 6 months: 5 trials; 411 patients; MD=9.01 points; 95% CI, -12.62 to -5.40). There were no significant differences in function outcomes (Index of severity for OA of the knee by Lequesne et al; The Knee Society Clinical Rating System), maximum flexion, or adverse events.

Triamcinolone improves pain, function, but not for long

A 2016 double-blind RCT of 110 patients with knee OA compared intra-articular HA and triamcinolone, assessing pain and function at intervals between 24 hours and 6 months.3 Patients in the HA group received a single injection of 6 mL hylan G-F 20 (Synvisc); patients in the CS group received 1 mL of triamcinolone acetonide 40 mg and 5 mL of 1% lidocaine with epinephrine.

The CS group reported significantly less pain (VAS score 1 to 100) at 24 hours than the HA group (24 points vs 36 points; P=.002); relief lasted as long as one week (14 points vs 23 points; P=.018). After the first week, no difference was seen in pain between groups for as long as 6 months.

Function, assessed by a modified Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC 1 to 100; higher score indicates worse pain, stiffness, and function) showed a significant improvement with CS at 2 weeks (25 points vs 31 points; P=.03), but no difference at any other time point up to 6 months.


Evidence-based answers from the Family Physicians Inquiries Network

Next Article: