Journal reference:Osani MC, Bannuru RR. Efficacy and safety of duloxetine in osteoarthritis: a systematic review and meta-analysis. Korean J Intern Med 2019 Mar 15

Link: https://dx.doi.org/10.3904/kjim.2018.460

Published: March 2019

Evidence cookie says…

Duloxetine (60-120 mg daily) appears to have a small to moderate beneficial effect on knee osteoarthritis pain and function at 3 months, compared to placebo. side-effects need to be considered – around a third experience gastrointestinal side-effects it does not appear to be more effective than NSAIDs people with chronic pain may have a bimodal response – either very poor or very good

An abridged version of this article was published in the July 2019 issue of Medical Observer.

Tam M. Duloxetine as treatment for knee osteoarthritis pain. Medical Observer. 2019 July 12.

Online version on AusDoc.Plus platform: “Is an SNRI effective in knee OA?”

Clinical scenario

Fatima, a 59-year-old lady living with painful knee osteoarthritis, was brought in by her daughter, a pharmacy assistant and one of my regular patients. During the consultation, duloxetine, a serotonin and noradrenaline reuptake inhibitor antidepressant, came up in conversation. I reflected afterwards that although it was frequently mentioned in pain discussions on an online GP forum, I had never previously commenced the drug for chronic pain. So, what is the evidence?

Clinical question

Is duloxetine an effective treatment for knee osteoarthritis pain (and loss of knee function)?

What does the research evidence say?

Step 1: The Cochrane Library

The Cochrane Library has a systematic review that largely focusses on peripheral neuropathy, but not other types of pain [1].

Step 2: TripDatabase and PubMed

I conducted a search using the TripDatabase PICO search tool (Participant: “knee osteoarthritis”, Intervention: “duloxetine”, Comparator: blank, Outcomes: blank). This identified quite a few clinical trials. To see whether there were any recent systematic reviews, I conducted a PubMed search using the search term “duloxetine osteoarthritis systematic review”. This identified two important systematic reviews – the first which examined the efficacy of duloxetine, NSAIDs, and opioids in osteoarthritis by Myers et al. in 2014 [2], and a very recent paper on the efficacy and safety of duloxetine in osteoarthritis by Osani and Bannuru, published in the Korean Journal of Internal Medicine in 2019 [3]. I will do a critical appraisal of the newer paper and include some insights from the earlier in the discussion.

Critical appraisal

I will use the systematic reviews critical appraisal sheet from the Centre for Evidence Based Medicine [4].

What PICO question does the systematic review ask?

In people with a clinical diagnosis of osteoarthritis of any joint (Participants); what is the effect of a duloxetine (60 to 120 mg) (Intervention); compared to a matching placebo (Comparator); on pain and functional status as measured on validated scales; and adverse events (Outcome).

Is it clearly stated?

Yes.

Is it unlikely that important studies were missed?

Yes, probably. The authors searched several appropriate databases, reviewed reference lists of relevant systematic reviews and supplements of conference proceedings. Given that much of the early research on duloxetine were internal pharmaceutical company trials, it is unknown whether there are important data that is unpublished.

Were the criteria used to select articles for inclusion appropriate?

Yes. The authors only included randomised trials (RCTs) with a matching placebo.

Were the included studies sufficiently valid for the question asked?

Possibly. The authors formally assessed risk of bias of the included studies and stated that most of the included trials were moderate to high quality. However, over half of the studies had high discontinuation rates, with differences between the duloxetine and placebo groups. The authors identified that the research team of one study has been previously flagged for research misconduct so opted to not include it in their analysis.

Were the results similar between studies?

Yes. The effect-size estimates for pain and function were similar across the included studies.

What were the results?

Duloxetine (60-120 mg) vs placebo treatment on pain at 12-14 weeks, 5 studies (1713 participants):

• Standardised mean difference: -0.38 (95% CI -0.48 to -0.28), I²=5%
• Small to moderate size effect that was seen consistently across the included studies.

Duloxetine vs placebo on function, 5 studies (1695 participants):

• SMD: -0.35 (95% CI -0.46 to -0.24), I²=23%
• Small to moderate effect size that was consistent across the studies.

Other results – the duloxetine groups has more adverse events. For duloxetine vs placebo:

• treatment-emergent adverse events: 55.1% vs 37.4%
• gastrointestinal adverse events: 35.5% vs 7.7%
• discontinuation: 12.4% vs 5.5%

Discussion and conclusion

Duloxetine appears to have a small to moderate beneficial effect (standardised mean difference of roughly 0.35) on knee osteoarthritis pain and function at 3 months, compared to placebo. This does need to be balanced with the cost of side-effects, especially gastrointestinal side-effects. Furthermore, many of the included studies had a differential in the rate of discontinuation due to adverse events between the duloxetine and placebo groups. The authors of this systematic review caution that this might bias their analysis towards overestimating the efficacy of duloxetine [3], something that had also been previously identified by other authors [5].

The systematic review by Myers et al. (2014) found that duloxetine was no different in efficacy to other analgesics (e.g., NSAIDs) for osteoarthritis after failure of paracetamol, though there is an indication that it is superior to both tramadol and hydromorphone [2]. Curiously, people with chronic painful conditions appear to have a bimodal (see Stat Facts) response to duloxetine – experiencing either a very good or very poor response [5].

There probably needs to be greater focus on non-pharmacological therapies (exercise, weight loss, psychological strategies) in the management of knee osteoarthritis in routine practice. Where pharmacotherapy is considered appropriate for chronic pain, duloxetine is a possible second-line option though the adverse effects need to be considered.

Stat Facts

“Mode” and bimodal distributions

“Mode” is another way of describing the central tendency of a distribution of values, in addition to the more common used mean and median. The mode of a dataset is the value that occurs most frequently. If the distribution of values has one “hump”, for instance, a “normal” distribution, then the dataset is unimodal. If the distribution of values has two “humps”, then it is bimodal. Importantly, the mean and median of a distribution of values is not meaningful when it is bimodal as there are two peaks rather than one.

References

1. Lunn MP, Hughes RA, Wiffen PJ. Duloxetine for treating painful neuropathy, chronic pain or fibromyalgia. Cochrane database of systematic reviews 2014 Jan 3(1):CD007115.
2. Myers J, Wielage RC, Han B, et al. The efficacy of duloxetine, non-steroidal anti-inflammatory drugs, and opioids in osteoarthritis: a systematic literature review and meta-analysis. BMC Musculoskelet Disord 2014 Mar 11;15:76.
3. Osani MC, Bannuru RR. Efficacy and safety of duloxetine in osteoarthritis: a systematic review and meta-analysis. Korean J Intern Med 2019 Mar 15.
4. Centre for Evidence Based Medicine. Systematic Review: Are the results of the review valid? Oxford: University of Oxford, 2005.
5. Moore RA, Cai N, Skljarevski V, Tolle TR. Duloxetine use in chronic painful conditions–individual patient data responder analysis. Eur J Pain 2014 Jan;18(1):67-75.