Recommendation ID

What is the clinical and cost effectiveness of antidepressants for the management of sciatica?

Any explanatory notes
(if applicable)

Why the committee made the recommendations

The evidence showed that gabapentinoids did not improve sciatica symptoms, and oral corticosteroids did not improve pain or function, but may have an impact on quality of life. Both increased the risk of adverse events in the long-term. While there was no evidence of increased risk of adverse events associated with benzodiazepines, there was evidence of poorer response than placebo in terms of pain reduction. The committee considered:

The committee agreed that although the evidence about lack of effectiveness was limited, the harms would outweigh the benefits for most people with sciatica and therefore agreed to recommend against the use of gabapentinoids, oral corticosteroids and benzodiazepines for sciatica.

There was no evidence on the use of antiepileptics (other than gabapentinoids) for sciatica. Given the lack of evidence, and the committee's knowledge of potential harms, they agreed to recommend that antiepileptics (including gabapentinoids) should not be used for sciatica.

There was no evidence on the use of opioids for sciatica. Given the lack of evidence and the committee's knowledge of potential harms when used long term, the committee agreed to recommend against the use of opioids for chronic sciatica. However, the committee discussed whether opioids might be effective when used short term for acute sciatica, so made a research recommendation on this topic.

There was no evidence on the use of antidepressants for sciatica. The committee agreed that antidepressants were commonly prescribed for sciatica, and clinical experience suggests they may be of benefit in some people. The committee considered the potential for harm to be less than the harms of prolonged use of opioids. On this basis, the committee made a research recommendation to determine if there was any clinical benefit for their use to treat sciatica.

Limited evidence showed no benefit from NSAIDs for sciatica. The committee discussed that there were also known risks of harms from NSAIDs that most clinicians were aware of so they were unlikely to be continued if they were not helpful. They agreed there was not sufficient evidence to make a recommendation for or against the use of NSAIDs for sciatica, but agreed to include a recommendation highlighting the risk of harms and lack of evidence of benefit as well as a research recommendation on this topic.

The committee were aware that some people may already be using opioids, antiepileptics (including gabapentinoids) and benzodiazepines for long periods for sciatica. Given the potential harms from sudden withdrawal of these medicines, based on consensus, they recommended discussing with the person the potential harms of long-term use and the need to withdraw safely if they chose to do so.

No evidence was identified for paracetamol, nefopam or muscle relaxants other than benzodiazepines for the management of sciatica. The committee agreed that none of these are widely prescribed for sciatica. They noted that advice is already included in this guideline for the use of paracetamol for people with low back pain. Therefore no further recommendations were made regarding management of sciatica alone, and these medicines do not warrant further research.

How the recommendations might affect practice

These recommendations are expected to reduce the use of gabapentinoids and other antiepileptics, corticosteroids, benzodiazepines and long-term opioid analgesics for sciatica. This will reduce the chance of adverse events and dependence on medicines that are unlikely to provide clinical benefit. It might lead to an increased use of other recommended treatments.

Full details of the evidence and the committee's discussion are in evidence review A: pharmacological management of sciatica.

Source guidance details

Comes from guidance
Low back pain and sciatica in over 16s: assessment and management
Date issued
November 2016

Other details

Is this a recommendation for the use of a technology only in the context of research? No  
Is it a recommendation that suggests collection of data or the establishment of a register?   No  
Last Reviewed 01/09/2020