OVERVIEW
- There is good evidence supporting the use of antidepressants in functional somatic symptoms; patients do not have to be depressed to derive benefit
- Other atypical analgesics such as gabapentin and pregabalin can also be useful
- This group of patients often tolerate drugs poorly due to a combination of sensitivity to side effects and the nocebo response
Introduction
This chapter will address four points: how drugs appear to work for medically unexplained symptoms (MUS); choosing which drug to use; explaining treatment; and side effects including the nocebo response.
How drugs appear to work for symptoms
When considering drug treatment for patients with MUS it can be helpful to think of five symptom groups as targets for treatment; pain; functional disturbance (e.g. other abnormalities of sensation, movement disorders, palpitations); fatigue; depression; and anxiety. Occasionally, other psychiatric diagnoses such as hypochondriasis, post-traumatic stress disorder (PTSD), psychotic illness, obsessive–compulsive disorder or even dementia may need to be considered for specific treatment.
Reducing depression or anxiety
Obviously if depression and anxiety are prominent, benefit can arise through treatment of these. However, the benefits of antidepressants and some anticonvulsants extend beyond this, although the mechanisms by which they work have been most studied for pain.
Reducing central sensitisation to pain
The brain is not only important for interpreting incoming stimuli and relating them to previous memories, but also has an important buffering effect on ascending pain signals. The descending antinocioceptive system (and potentially analogous systems for other sensory stimuli) that acts as a filter or ‘pain barrier’ to incoming signals, may be defective in functional syndromes such as fibromyalgia and IBS. Antidepressants and anticonvulsants may act by enhancement of these descending antinocioceptive effects, normalising a system that has lost some natural filters and barriers.
Altering symptom appraisal and autonomic responses
Furthermore, it is increasingly recognised that many patients with functional symptoms have an exaggerated stress response (this likely contributing to the worsening of symptoms in the context of life stressors), and have an attentional bias for unpleasant or threatening stimuli. Drugs may thus work by normalising these exaggerated endocrine and/or autonomic stress responses and by inhibition of prefrontal cortical areas that underpin ‘attention’ to noxious stimuli.
This physiological evidence is supported by evidence from systematic reviews that it is not necessary to be depressed to benefit from ‘antidepressants’. Indeed, patients with a number of functional syndromes benefit from doses of tricyclic drugs which are sub-therapeutic for depression. People with a wide range of functional somatic symptoms appear to benefit and in many cases there is little to choose between classes except tolerability. Antidepressants with both serotonergic and noradrenergic activity (such as tricylic antidepressants other than clomipramine, venlafaxine and duloxetine) appear to have benefit in chronic pain that is independent of any mood-elevating effects.
Choosing which drug to use
The two main classes of centrally acting drugs for managing MUS are antidepressants and anticonvulsants, however there are a few other drugs with a potential role. Table 17.1 lists a range of conditions within the MUS spectrum and summarises the options for drug treatment. Precautions in prescribing for particularly patient groups are detailed in Table 17.2
Table 17.1 Prescribing tips for specific conditions.
| Condition | Medication | Notes |
| Palpitations | Propranolol | Increased awareness of normal heartbeat can be helped by propranolol. It can also be helpful if other symptoms of autonomic arousal, such as exaggerated physiological tremor, are contributing to health anxiety |
| Tension type headache | TCAs | If not tolerated then SSRIs are likely to be a more reasonable alternative than other antidepressants |
| Irritable bowel syndrome | Antispasmodics e.g. mebeverine | May relieve cramping pain |
| TCAs | Useful if pain and diarrhoea prominent, may worsen constipation | |
| SSRIs | Less useful for pain but will not worsen constipation | |
| SNRIs | May be good compromise if constipation and prominent pain symptoms | |
| Chronic pelvic pain | COCP, GnRH agonist (e.g. goserelin) | Hormonal treatment may be of some benefit in those whose pain is cyclical |
| Antispasmodics | If comorbid with irritable bowel syndrome | |
| TCAs, gabapentin | Atypical analgesia options are as for fibromyalgia (see below) | |
| NSAIDs | In contrast with fibromyalgia, however, NSAIDs are worth trialling in chronic pelvic pain | |
| Fibromyalgia | TCAs | Atypical analgesic, sleep promoting and gastrointestinal effects may all be beneficial. Often benefit from doses regarded as sub-therapeutic for treatment of depression |
| SNRIs | Preferable to SSRIs as greater atypical analgesic effects | |
| Pregabalin, gabapentin | Atypical analgesic effects and also some anxiolytic (licensed indication for pregabalin) and mood-elevating effects | |
| Non-epileptic attacks (dissociative convulsions) | SSRIs, trazodone | If panic disorder is clearly present it should be treated aggressively, with SSRIs as first-line treatment. If it is not, or treatment is not tolerated, sedative antidepressants with anxiolytic properties (e.g. trazodone or mirtazapine can be useful) |
| Dissociative motor or sensory disorders | TCAs | Pain is often prominent; treating as per fibromyalgia can be helpful |
| Chronic fatigue syndrome | As fibromyalgia | If pain prominent treat as for fibromyalgia If pain not prominent SSRIs are a reasonable (non-sedating) alternative. Conversely, if insomnia prominent and TCAs not tolerated mirtazapine or trazodone worth trying. No evidence to support the use of stimulants |
In these summaries it is assumed that depression and anxiety/panic disorder are not particularly prominent. If they are their treatment should be prioritised. COCP, combined oral contraceptive pill; GnRH, gonadotrophin-releasing hormone; NSAIDS, non-steroidal anti-inflammatory drugs; SNRI, Serotonin norepinephrine reuptake inhibitors; SSRI, selective serotonin reuptake inhibitors; TCA, tricyclic antidepressants.
Table 17.2 Precautions in prescribing for particular patient groups.
| Medication | Precautions |
| SSRIs | If substantial comorbidity citalopram/escitalopram or sertraline are first choice as lower potential for drug interactions. Note recent revised dose limits of citalopram (40 mg rather than 60 mg in under 65s) with increased recognition of the potential of citalopram/escitalopram to lengthen the QTc interval (i.e. the time between the start of the Q wave and the end of the T wave corrected for heart rate as measured on an ECG) Bleeding risk; may want to prescribe gastroprotective drug if patient is older or on NSAIDs Initial increase in anxiety means often prudent to start on half dose Relatively safe in overdose, but need to review after initiation in case increase in suicidality |
| TCAs | Dangerous in overdose. Lofepramine relatively safe and (although still very dangerous) nortriptyline and imipramine less toxic than amitriptyline. Lofepramine and nortriptyline have relatively less serotonergic activity, so may not be quite as effective in pain symptoms as more dual acting TCAs. Avoid if recent myocardial infarction/unstable angina. Caution if co-prescribed with other QTc prolonging drugs Lower seizure threshold; try to avoid in epilepsy |
| Carbamazepine | Teratogenicity means should be avoided in women of childbearing age |
| Gabapentin | May worsen absence or myoclonic seizures. Evidence of safety in pregnancy lacking |
| Pregabalin | Evidence of safety in pregnancy lacking |
| Benzodiazepines | Addictive potential means best avoided in these often chronic conditions |
| Propranolol | Avoid in asthma. As may mask signs and symptoms of hypoglycaemia caution in diabetes. Caution in pregnancy |
NSAIDS, non-steroidal anti-inflammatory drugs; SSRI, selective serotonin reuptake inhibitors; TCA, tricyclic antidepressants.
Antidepressants
There is good evidence that antidepressants are beneficial in functional somatic syndromes, with a number needed to treat for short-term improvement of approximately four. The benefit is seen in patients with and without depression. Limited data guides antidepressant choice. Meta-analyses comparing the responses of patients with headaches, fibromyalgia, and chronic pain suggest tricyclic antidepressants (TCAs) are slightly more effective than selective serotonin reuptake inhibitors (SSRI). The difference is probably greatest in patients with chronic, unexplained pain. A reasonable principle is that when pain symptoms are prominent then it makes sense to choose an agent with combined noradrenergic and serotonergic activity.
Tricyclic drugs
These are the most commonly prescribed drugs for symptoms. Most GPs are familiar with using amitritpyline for a wide range of pain syndromes including post-herpetic neuralgia. Some specialists prefer imipramine and some patients seem to tolerate nortriptyline better.
Serotonin norepinephrine reuptake inhibitors (SNRI)
Duloxetine or venlafaxine (aiming for doses above 150 mg of the latter) are reasonable choices if there are contraindications to tricyclic use. If these drugs are not tolerated, however, there is still reason to be optimistic that SSRIs can provide benefit.
Selective serotonin reuptake inhibitors (SSRIs)
SSRIs probably have a class effect. Citalopram/escitalopram and sertraline are the most commonly used in this situation.
Trazodone
The sedative, anxiolytic, non-addictive agent trazodone can be very useful if insomnia is prominent. It can also be helpful, in split doses, if anxiety is particularly prominent. Although there is a theoretical risk of serotonin syndrome and an increased risk of gastrointestinal bleeds it is generally reasonably safe combined with SSRIs.
Anticonvulsants
Anticonvulsants are useful in pain management. Pregabalin, gabapentin and carbamazepine have a clear role and lamotrigine and topiramate may also have pain-reducing effects. As many of the neurophysiological processes in pain are common to both ‘explained’ and ‘unexplained’ pain syndromes it makes sense to try these effective drugs in patients for whom pain is a major symptom, regardless of cause.
Pregabalin also has anxiolytic effects, for which it is licensed, and this property may be shared by gabapentin. Controlled trials are needed, but clinical experience does suggest that pregabalin and gabapentin have a useful role in some functional somatic syndromes.
Explaining treatment
Many patients with MUS do not regard themselves as having depression (and a good proportion are right!). Consequently, if you are prescribing a psychotropic drug you need to explain why. If the first time your patient finds you have prescribed an ‘antidepressant’ is when they read the information leaflet, then it is too late. Your patient will probably feel deceived, diminished or dismissed.
You may wish to explain that antidepressants are frequently used to treat symptoms such as pain and headache and are effective even if people are not depressed. Consider explaining treatment in terms of correcting physiological processes: for instance restoring the nerve pathways that act as symptom filters or barriers. When prescribing an anticonvulsant, again make it clear that this is not for epilepsy. Some clinicians find it is useful to consider the analogy of other drugs that have multiple uses, for instance aspirin being used to treat a headache or to thin the blood.
Remember that many people still think of antidepressants as addictive, you may need to counter that.
As these patients are particularly prone to side effects (see below) drugs should be started at low dose and increased gradually. A ‘script’ for discussing the initiation of antidepressants is suggested below. As will be clear from the discussion above this does actually reflect what we know about the actions of these drugs rather than being disingenuous.
Pain like yours often needs something as well as painkillers in order to build up pain resistance in the nerves. X is a drug we often use to do this. It started out as a treatment for depression (and if you read the leaflet in the pack it says that), but it works just as well for pain in people who don’t have depression.
Reviewing and discontinuing drugs
Often, before the diagnosis became clear, a variety of unnecessary drugs have already been started. These can contribute significantly to symptom load. This is particularly apparent in patients with pain symptoms in whom opioid analgesia may result in fatigue, constipation and possibly intermittent withdrawal symptoms, while contributing little to symptom control.
There is no evidence that NSAIDs are beneficial in fibromyalgia, and these should be stopped. Up to 80% of people with non-epileptic attacks (dissociative seizures) in whom epilepsy has been excluded have been exposed to anticonvulsants. There is a comparable problem with anti-anginal drugs for patients with chest pain and normal arteries. Such prescriptions have the potential to cause considerable confusion for both doctors and patients, and in the case of non-epileptic attacks cessation of these drugs is associated with a reduction in frequency of attacks. Anticonvulsants should be stopped through a tapered reduction because of the risk of withdrawal seizures (see Table 17.3).
Table 17.3 How to stop anticonvulsant drugs (data from Oto et al. 2005).
| Drug | Withdrawal protocol |
| Phenytoin | 100 mg/week until dose is 100 mg/day, then 25 mg/week |
| Carbamazepine | 200 mg/week until dose is 1000 mg/day, then 100 mg/week |
| Sodium valproate | 500 mg/week until dose is 500 mg, then 200 mg/week |
| Vigabatrin | 500 mg every 2 weeks until dose is 500 mg/day, then 500 mg alternated days for 2 weeks |
| Lamotrigine | 100 mg/week until dose is 300 mg, 50 mg/week until dose is 50 mg, then 25 mg/week |
| Gabapentin | 800 mg/week until dose is 1200 mg, then 400 mg/week |
| Topiramate | 100 mg/week until dose is 200 mg, 50 mg/week until dose is 50 mg, then 25 mg/week |
| Levetiracetam | 500 mg/week until dose is 1000 mg, then 250 mg/week |
| Pregabalin | 200 mg/week until 200 mg, then 100 mg/week |
If you are the patient’s GP, you will be well placed to review why particular drugs were started, if they had any beneficial effect, and whether there is any ongoing rationale for their use. Starting an antidepressant can be a good opportunity to down-titrate and stop unnecessary drugs. In the case of functional pain it can be explained to the patient that atypical analgesics such as antidepressants and anticonvulsants are more effective than NSAIDs or opioid analgesia for the sort of pain that they have. Starting them should enable discontinuation of the side-effect causing agents they are currently taking.
Addiction to prescribed treatment
Detailed discussion of the relationship between chronic pain and addiction is beyond the scope of this chapter, but Box 17.1 summarises how to recognise addiction in chronic pain. Addiction requires the presence of aberrant behaviours, as physical dependence and tolerance alone are expected physiologic phenomena associated with chronic opioid or benzodiazepine treatment. Its prevalence in this population is estimated as 3–19%, above the population prevalence of substance addiction.
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
