Re: Efficacy, safety, and tolerability of antidepressants for pain in adults: overview of systematic reviews

Dear Editor

It is over 25 years since the first systematic review of the evidence for antidepressants and chronic pain concluded that a small minority of patients have excellent pain relief, but most are unaffected [1]. The overview by Ferreira and colleagues tells us little more and takes us no further [2].

We have subsequently learned that randomised trials in Cochrane reviews often have a high risk of bias [3] and that a pain trial with unequivocally low risk of bias is extremely rare [4]; most trials in most chronic pain reviews will have high risk of bias for at least one item. Moreover, most systematic reviews in pain and anaesthesia (85%) provide only critically low or low confidence in their results [4]. Seventeen of 26 reviews included in this overview were judged of a quality to provide critically low or low confidence in their results, begging the question of why they were included. Cochrane reviews usually provide high or moderate confidence; there are many for chronic pain – 50 for neuropathic pain, 20 for fibromyalgia, plus others.

With discrimination we find that large, good quality trial results exist for duloxetine only, with additional data from individual patient-level analysis demonstrating that for duloxetine different imputation methods (known to be a major source of bias in chronic pain trials) have little influence effect size estimates in any of four chronic pain conditions [5]. The use of mean data in pain efficacy trials is a common error and obscures the findings.

People with pain want to achieve a state of ‘no more than mild pain’ as quickly as possible [6], and the outcome of at least 50% pain intensity reduction with tolerable adverse events at the end of treatment has become a common outcome in chronic pain trials [7]: it is strange that it was not used here.

We should not treat anti-depressants as a class for chronic pain management. The differences amongst them are greater than their similarities. The BMJ already has a summary of effects of drugs by individual drug and dose in various acute and chronic pain conditions from mainly Cochrane reviews [8].

References

1. McQuay HJ, Tramèr M, Nye BA, Carroll D, Wiffen PJ, Moore RA. A systematic review of antidepressants in neuropathic pain. Pain. 1996 Dec;68(2-3):217-27. doi: 10.1016/s0304-3959(96)03140-5. PMID: 9121808.

2. Ferreira G E, Abdel-Shaheed C, Underwood M, Finnerup N B, Day R O, McLachlan A et al. Efficacy, safety, and tolerability of antidepressants for pain in adults: overview of systematic reviews BMJ 2023; 380 :e072415 doi:10.1136/bmj-2022-072415

3. Pirosca S, Shiely F, Clarke M, Treweek S. Tolerating bad health research: the continuing scandal. Trials 2022; 23: 458

4. Moore A, Fisher E, Eccleston C. Flawed, futile, and fabricated – features that limit confidence in clinical research in pain and anaesthesia: a narrative review. Br J Anaesth. 2022 Nov 8:S0007-0912(22)00574-8. doi: 10.1016/j.bja.2022.09.030. Epub ahead of print. PMID: 36369016.

5. Moore RA, Cai N, Skljarevski V, Tölle TR. Duloxetine use in chronic painful conditions–individual patient data responder analysis. Eur J Pain. 2014 Jan;18(1):67-75. doi: 10.1002/j.1532-2149.2013.00341.x. Epub 2013 Jun 3. PMID: 23733529; PMCID: PMC4302330.

6. Moore RA, Straube S, Aldington D. Pain measures and cut-offs – ‘no worse than mild pain’ as a simple, universal outcome. Anaesthesia. 2013 Apr;68(4):400-12. doi: 10.1111/anae.12148. Epub 2013 Jan 24. PMID: 23347230.

7. Finnerup NB, Haroutounian S, Baron R, Dworkin RH, Gilron I, Haanpaa M, Jensen TS, Kamerman PR, McNicol E, Moore A, Raja SN, Andersen NT, Sena ES, Smith BH, Rice ASC, Attal N. Neuropathic pain clinical trials: factors associated with decreases in estimated drug efficacy. Pain. 2018 Nov;159(11):2339-2346. doi: 10.1097/j.pain.0000000000001340. PMID: 30015707; PMCID: PMC6193835.

8. Moore A, Derry S, Eccleston C, Kalso E. Expect analgesic failure; pursue analgesic success. BMJ. 2013 May 3;346:f2690. doi: 10.1136/bmj.f2690. PMID: 23645858.