Triamcinolone, cartilage loss and knee pain in osteoarthritis

Triamcinolone intra-articular corticosteroid injection is approved as an adjunctive, short-term treatment for pain relief in patients with knee osteoarthritis (OA). Long-term use is not supported by current evidence, including published 2-year clinical data showing that repeat intra-articular triamcinolone injections every 12 weeks, as used in clinical practice, lead to cartilage loss and do not reduce pain in patients with knee OA.

• Triamcinolone, and several other intra-articular corticosteroid injections, can be used for short-term pain relief for people with knee OA.
  
• In a 2-year clinical study, intra-articular triamcinolone injections every 12 weeks led to cartilage loss and did not reduce pain for patients with knee OA.
  
• In OA, improvements in knee pain after intra-articular corticosteroid injections have been shown to be moderate at 1–2 weeks, small at 13 weeks and not evident at 26 weeks.
  
• Management of OA should aim to optimise the patient’s physical function, maximise their quality of life, enable their ability to cope with pain and reduce symptoms.
  
• If considered, intra-articular corticosteroid injections should generally be administered short term, and as an adjunct to core treatments.
  

Triamcinolone, and several other intra-articular corticosteroid injections, can be used for short-term pain relief for people with knee OA.^1-7

However, results of a 2-year clinical study that was published in 2017 do not support the use of intra-articular triamcinolone injected every 12 weeks for the treatment of symptomatic knee OA.⁸

What can we take from this study, and how should we manage knee OA? 

In the double-blind study, 140 patients with symptomatic knee OA (and ultrasonic features of synovitis) were randomised to receive either intra-articular injections of triamcinolone acetonide 40 mg or saline (0.9% sodium chloride; placebo) every 12 weeks for 2 years.⁸

Changes in knee cartilage volume loss (using cartilage thickness) and pain were co-primary outcomes.⁸

Study participants

Patients included in the study were aged ≥ 45 years with a diagnosis of OA defined using clinical diagnostic criteria from the American College of Rheumatology.⁸

Patients were required to have knee pain, radiological features of OA, ultrasonographic evidence of effusion synovitis, and pain from the knee joint confirmed by clinical examination.⁸

Patients were excluded from the study if they had symptoms suggestive of an alternative diagnosis (such as inflammatory joint disease or osteonecrosis).⁸

Based on the above eligibility criteria, the included patients can be considered as having ‘typical’ knee OA (see ‘How should knee OA be diagnosed?’ for more information).

Co-primary outcomes

The first co-primary outcome, change in cartilage volume, was assessed in the compartment with greatest joint space narrowing (the ‘index’ compartment), using cartilage thickness.⁸

Cartilage volume was assessed every 12 months, using MRI.⁸

The study authors advised that they used MRI to enable direct quantitation of cartilage and soft-tissue structures, because radiography is insensitive to OA progression and does not directly image critical soft-tissue structures or bone marrow lesions in OA.⁸

However, the amount of change in cartilage volume that represents a minimal clinically important difference has not been established.⁸

The second co-primary outcome, pain, was assessed every 12 weeks using the Western Ontario and McMaster Universities (WOMAC) OA Index pain subscale.⁸

Using the WOMAC pain subscale, a score of 0 represents no pain and a score of 20 represents extreme pain.⁸

A 3.94-point improvement is considered to represent a minimal clinically important difference.⁸

Cartilage loss was statistically significantly greater with triamcinolone than with the placebo, with a between-group difference of –0.11 mm (95% confidence interval [CI] –0.20 to –0.03, p = 0.01).⁸

Knee pain reduction was not significantly different between triamcinolone and the placebo, with a between-group difference of –0.64 (95% CI –1.6 to 0.29, p = 0.17).⁸

Triamcinolone’s effect on cartilage volume and pain are not unexpected (see the study limitations below and ‘How should knee OA be managed?’ for more information).

Secondary outcomes

A validated volumetric cartilage damage index was also used to quantitate cartilage loss.⁸

Using the cartilage damage index, cartilage loss was significantly greater with triamcinolone than with the placebo (between-group difference of −61.25 µm³).⁸

However, changes in area of denudation, bone marrow lesion volume and effusion volume were not significantly different between the two treatment groups.⁸

There were also no significant differences in knee function or stiffness between the treatment groups.⁸

Safety

There were significantly more adverse events with saline than with triamcinolone (63 versus 52 patients, 182 versus 131 events; p = 0.02 for both comparisons).⁸

Eight events were treatment-related, three with saline (one cellulitis, two injection site pain) and five with triamcinolone (one facial flushing, four injection site pain).⁸

Serious adverse events were similar between the two treatment groups (p = 0.06).⁸

Limitations

The study authors acknowledged that treatment responses could potentially have been affected by high expectations of active treatment, large placebo responses to saline injection, and patients being allowed to continue their usual medicines.⁸

In addition, because corticosteroids typically have only short-term effects on pain, transient responses could have been missed because pain was not measured within the 4-week period after each injection.⁸

A diagnosis of knee OA is nearly 100% likely if a patient has all of the following signs and symptoms:⁹

• persistent activity-related knee pain
  
• limited morning stiffness
  
• reduced function
  
• crepitus
  
• restricted movement
  
• bony enlargement.
  

Although age of onset, symptoms and progression of OA will vary between patients, there are several ‘typical features’ of OA.^10,11

These are considered to be age > 40 years, activity-related joint pain, short duration of (or no) morning stiffness, and symptoms affecting only one or a few joints.¹¹

Knee OA can be diagnosed using these typical features, without X-ray or MRI, assuming no alternative or additional diagnoses are suspected.^11-13

Radiological features of OA typically have a low correlation to OA symptoms, and MRI should be considered only if there is suspicion of serious pathology not detected by X-ray.^12,14

Overall, management of OA should aim to maintain or optimise the patient’s physical function, maximise their quality of life, enable their ability to cope with pain and reduce symptoms.¹⁴

If analgesics are considered, then they should be recommended to help enable physical function, rather than to abolish pain.¹⁴

Although management will differ for each patient – because patients’ needs and responses to different OA management options will vary – weight management and exercise are recommended for all people with OA.¹⁴

If required, weight loss and regular exercise can help to reduce pain and improve physical function in people with knee OA.¹⁴

Whatever intervention is selected, it should be assessed regularly against individual goals, to determine whether it is safe and beneficial.¹⁴

What about intra-articular corticosteroid injections for knee OA?

Corticosteroid injections approved in Australia are generally indicated as adjunctive therapies for short-term administration (see Table 1).^1-7

For example, because of their rapid onset of action, intra-articular corticosteroid injections may be useful if patients have to travel and are concerned about pain from their OA.¹⁴

According to Australian guidelines on the use of intra-articular corticosteroid injections:

• no more than four injections should be administered to a single joint in a year, due to the potential for increased risk of cartilage damage¹⁵
  
• if there is no response after two consecutive injections, further injections should be avoided¹⁵
  
• a single injection may provide symptom relief lasting 4–12 weeks for patients with knee OA¹⁴
  
• clinical trials supporting intra-articular corticosteroid use may have a high or unclear risk of bias¹⁴
  
• corticosteroids are used for a wide range of conditions, often without substantial evidence of their effectiveness.¹⁵
  

In a meta-analysis of 26 randomised controlled trials, intra-articular corticosteroid injections in patients with knee OA were found to lead to small or moderate improvements in knee pain.¹⁶

Improvements were moderate at 1–2 weeks after the end of treatment (standardised mean difference [SMD] –0.48, 95% CI –0.70 to –0.27), small to moderate at 4–6 weeks (SMD –0.41, 95% CI –0.61 to –0.21), and small at 13 weeks (SMD –0.22, 95% CI –0.44 to 0.00).¹⁶

However, the quality of evidence was low, results were inconclusive and there was no evidence of reduced pain 26 weeks after injection (SMD –0.07, 95% CI –0.25 to 0.11).¹⁶

What about ‘regenerative’ injectables for knee OA?

Despite their increased use for OA, Australian therapeutic guidelines do not recommend intra-articular platelet-rich plasma (PRP), adipocyte cell suspensions or mesenchymal stem cell injections for treatment of OA.¹⁴

They advise that results from studies investigating intra-articular PRP injections should be interpreted with caution due to the high risk of bias.¹⁴

In addition, evidence to support the use of adipocyte cell suspensions and mesenchymal stem cells is also weak.¹⁴

In 2014, the Australian Rheumatology Association released a position statement on stem cell therapies.²³

It advised that there is currently insufficient evidence to recommend stem cell therapy/autologous cell-based interventions for OA outside of a clinical trial setting.²³

In October 2017, the TGA advised that it will change how autologous human cell and tissue products (including stem cell therapies) will be regulated in Australia.²⁴

These products have previously not been regulated by the TGA.²⁴

However, there is growing concern about the advertising of unproven treatments to consumers, and potential safety risks due to increasing treatment complexity.²⁴

The TGA has therefore proposed the following changes for the regulation of autologous human cell and tissue products:²⁴

• prohibit direct advertising to consumers
  
• exclude from regulation only those products manufactured and used in a hospital by a medical or dental practitioner, for a patient in the care of the same practitioner
  
• regulate products that are minimally manipulated, and for homologous use only, and manufactured and used outside a hospital by a medical or dental practitioner, and for a patient in the same practitioner’s care
  
• regulate under the Biologicals Regulatory Framework products that are manufactured and used outside an accredited hospital, and more than minimally manipulated, or for non-homologous use.
  

All patients with knee OA are encouraged to seek education about their condition and advice on self-management.¹⁴

You may find the following information useful when discussing knee OA and its management with your patients.

• NPS MedicineWise information for patients developed as part of the 2017 therapeutic program, Osteoarthritis: Practical tools for diagnosis and management.
  
• MyJointPain free resource intended to help patients relieve symptoms of joint pain or OA, and better manage their condition.
  
• Arthritis Australia website and information sheets (including translated versions).
  
• Healthdirect health service finder and question builder to help patients manage their OA and ask the right questions.
  
• TGA quick guide for consumers on stem cell treatments and their regulation.