What Does the Evidence Say? (By Injection Type)

Below we summarise the highest-quality evidence from major guidelines and randomised trials for common joint injections. We cite sources directly to help you (and your GP/physio) understand where each recommendation comes from.

Corticosteroid (Steroid) Injections

  • Knee Osteoarthritis (OA):
    • Guidelines: International guidelines consistently support short-term symptom relief from intra-articular corticosteroids when first-line care (exercise, education, weight management, NSAIDs) is insufficient. The 2019 OARSI guideline conditionally recommends steroid injections for knee OA; it emphasises exercise/education as foundational care.
    • Randomised trial signal of harm with frequent, long-term use: In a 2-year RCT of 140 patients receiving 40 mg triamcinolone every 3 months, steroids led to greater cartilage thickness loss versus saline (−0.21 mm vs −0.10 mm; between-group −0.11 mm; 95% CI −0.20 to −0.03) and no pain benefit. This supports limiting frequency and duration.
    • Practical take: Best for an inflamed, irritable knee needing a short-term flare settle (weeks), not as a repetitive long-term strategy.
  • Hip & Shoulder:
    • Hip OA: OARSI allows consideration similar to knee OA in select patients; benefits are short-term and patient-specific.
    • Subacromial/Rotator Cuff–Related Pain: Earlier systematic reviews show modest, short-term pain relief compared with NSAIDs or placebo; effect wanes over weeks to a few months. Repeated injections risk tendon effects; use judiciously.

Hyaluronic Acid (HA, “Gel”) Injections

  • Knee OA—mixed/negative overall effect at population level:
    • BMJ 2022 systematic review/meta-analysis of randomised trials found viscosupplementation yielded a small, clinically unimportant reduction in pain versus placebo and reported an increased risk of serious adverse events; authors concluded results do not support broad use.
    • AAOS Knee OA CPG (non-arthroplasty): Not recommended for routine use in symptomatic knee OA.
    • Practical take: Some individuals report benefit (especially milder OA phenotypes), but average effect is small; shared decision-making is key.
  • Hip OA: Evidence is less robust than knee; comparative trials are limited. Some reviews compare PRP vs HA with variable findings; heterogeneity is high.

Platelet-Rich Plasma (PRP)

  • Knee OA:
    • Network meta-analyses (2023–2025) generally show PRP (and some biologic protocols) achieving greater pain and function improvement at 6–12 months than corticosteroids and, in many analyses, than HA; durability beyond 12 months varies and protocols differ (single vs series, leucocyte-poor vs rich).
    • Guideline posture: Major public guidelines (ACR/OARSI/AAOS) have been cautious historically due to heterogeneity and cost; they prioritise core treatments first and consider PRP on a case-by-case basis.
    • Practical take: PRP can be a reasonable option for mild–moderate knee OA seeking longer relief than steroids, understanding out-of-pocket costs and variability in preparation.
  • Hip OA: Early comparative evidence suggests PRP may perform as well as or better than HA for symptoms in some studies, but trials are smaller and heterogeneous.

Genicular Nerve Procedures (Diagnostic Blocks & Radiofrequency)

  • For chronic knee OA pain not yet proceeding to arthroplasty:
    • Randomised data suggest cooled radiofrequency ablation (RFA) of genicular nerves improves pain/function compared with intra-articular HA in selected patients; real-world and sham-controlled trials support benefit at up to 6–12 months in appropriate candidates.
    • Ongoing trials continue to refine indications and technique (e.g., phenol neurolysis vs RFA vs conservative care).

Bursitis & Tendinopathy (Shoulder, Hip)

  • Subacromial pain/rotator cuff–related shoulder pain: Systematic reviews indicate small, short-term benefits from subacromial steroid injections; benefits diminish by months. Consider as part of a plan anchored in rehab.
  • Greater Trochanteric Pain Syndrome (GTPS): Evidence supports short-term pain relief from corticosteroid injections; comparisons with PRP, physio, or shockwave therapy show variable results across small RCTs; ultrasound guidance can improve accuracy.

How Often, and Safety Considerations

  • Frequency: Given the 2-year RCT signal of accelerated cartilage loss with quarterly knee steroid injections, we avoid frequent, repeated steroid use; we individualise intervals and aim for the fewest injections necessary.
  • Diabetes: Steroids can transiently raise blood glucose; we advise monitoring and coordination with your GP for insulin/oral agent adjustments.
  • Hyaluronic Acid: Population-level benefit is small; not routinely recommended by AAOS for knee OA, and 2022 BMJ meta-analysis questions broad use. Consider selectively after shared decision-making.
  • PRP: Autologous product with low adverse-event rates in trials; outcomes vary with preparation (e.g., leucocyte-poor vs rich), dose, and number of injections. We use standardised, evidence-aligned protocols to reduce variability.
  • Procedure guidance: Ultrasound or fluoroscopic guidance improves accuracy (hip, shoulder, GTPS, deep joints), which can enhance outcomes and reduce complications.

How We Choose the Right Injection

We base recommendations on your diagnosis, OA severity, signs of active inflammation, comorbidities, and your goals. In general:

  • Need rapid flare control: Consider single, judicious corticosteroid injection while advancing rehab.
  • Seeking longer relief (mild–moderate knee OA): Discuss PRP after core care; set realistic expectations (6–12 months typical in trials).
  • Considering HA: We discuss modest average effect and guideline positions; may trial in selected patients who prefer it and understand uncertainties.
  • Persistent knee pain not ready for replacement: Consider genicular nerve RFA after diagnostic blocks if appropriate.

Injection Options Compared: Benefits, Durability, and Guideline Position

We individualise recommendations based on your diagnosis, severity, and goals. Use this table to compare onset of relief, typical duration, evidence strength, and guideline stance.

Summary of common injection options for osteoarthritis and periarticular pain.
Injection Best For Onset of Relief Typical Duration Evidence Summary Guideline Stance Key Risks / Considerations
Corticosteroid (Steroid) Inflamed, irritable joints (e.g., knee OA flare); selected hip/shoulder bursitis or subacromial pain Fast (24–72 hours) Weeks to a few months; effect wanes with time Consistent short-term pain relief. A 2-year RCT of quarterly steroid in knee OA showed more cartilage loss vs saline and no pain advantage with frequent use. Supported for short-term relief after core care (exercise, weight, NSAIDs). Use judiciously; avoid frequent repeat courses. Post-injection flare, transient glucose rise (diabetes), skin changes; limit frequency to reduce cartilage risk.
Hyaluronic Acid (HA, “Gel”) Knee OA (select patients) Gradual (1–4 weeks) Variable; often up to 3–6 months in responders Large 2022 BMJ meta-analysis shows small, clinically trivial average benefit vs placebo and higher serious adverse events; individual response varies. AAOS: Not recommended for routine use in knee OA. Consider selectively after shared decision-making. Injection-site pain/swelling; cost considerations; manage expectations around modest average effect.
Platelet-Rich Plasma (PRP) Mild–moderate knee OA aiming for longer relief than steroids; select hip OA cases Gradual (2–6 weeks) Often 6–12 months in trials; durability varies by protocol Multiple recent systematic reviews/network meta-analyses show PRP often outperforms HA and steroids for pain/function at 6–12 months; prep methods (e.g., leucocyte-poor) may influence outcomes. Major guidelines remain cautious due to heterogeneity and cost; reasonable option case-by-case after core care. Low adverse-event rate (autologous). Out-of-pocket costs; protocol standardisation matters (number of injections, LP-PRP).
Genicular Nerve Radiofrequency Ablation (RFA) Chronic knee OA pain not ready for arthroplasty, after diagnostic nerve blocks Fast to intermediate (days to 2 weeks) 6–12 months in responders; repeatable Randomised and cohort data suggest superior pain/function vs intra-articular comparators in select patients; ongoing trials refining technique and selection. No universal OA guideline consensus; growing supportive evidence where expertise is available. Neuropathic symptoms, numbness, rare neuritis; requires accurate targeting and experienced operator.
Diagnostic / Therapeutic Nerve Blocks To confirm nerve-mediated knee pain; temporary relief; pre-RFA planning Immediate (minutes to hours) Hours to days Useful as part of a pathway toward RFA; limited durability alone. Adjunctive/diagnostic tool rather than stand-alone therapy. Transient numbness/weakness; short-acting by design.

References (External Links)

  1. American Academy of Orthopaedic Surgeons (AAOS). Management of Osteoarthritis of the Knee (Non-Arthroplasty), 3rd ed. (2021 CPG; updated online).
    PDF
    Guideline hub
  2. Osteoarthritis Research Society International (OARSI). 2019 Guidelines for the Non-Surgical Management of Knee, Hip, and Polyarticular OA.
    PDF
    Overview
  3. McAlindon TE, et al. Effect of intra-articular triamcinolone vs saline on knee cartilage volume and pain in knee OA: a 2-year RCT. JAMA 2017.
    Article
  4. Pereira H, et al. Viscosupplementation for knee OA: systematic review and meta-analysis of randomised trials. BMJ 2022.
    PDF
  5. Qiao X, et al. Efficacy and safety of corticosteroids, hyaluronic acid, PRP and combinations for knee OA: network meta-analysis. BMC Musculoskeletal Disorders 2023.
    PDF
  6. Jawanda H, et al. PRP, BMAC, and HA outperform corticosteroids at ≥6 months as intra-articular injections for knee OA: systematic review & network meta-analysis. Arthroscopy 2024.
    Article
  7. Zhou Q, et al. Critical overview of SRs/MAs of PRP vs HA in knee OA. Clinical Rheumatology 2024 (published online 2025 print).
    Article
  8. Iannaccone F, et al. PRP vs HA for knee OA: meta-analysis of RCTs. Journal of Orthopaedic Surgery & Research 2020.
    Article
  9. Chen AF, et al. Genicular nerve radiofrequency ablation for painful knee OA: evidence & patient selection (narrative review with RCT summary). Pain Medicine 2021.
    Article
  10. de Wit PR, et al. Genicular nerve radiofrequency ablation, phenol neurolysis or conservative medical management in patients with knee osteoarthritis: protocol for the RADIOPHENOL randomised controlled multicentre trial. BMJ Open 2025;15(7):e094576.
    Article