Editorials

Interrupting the sympathetic outflow in causalgia and reflex sympathetic dystrophy

A futile procedure for many patients 

Causalgia and reflex sympathetic dystrophy are poorly understood disorders that most commonly follow trauma to a limb, although they are also seen in other medical conditions. Patients typically develop chronic burning pain, together with various combinations of sensory disturbances, swelling, and vasomotor, sudomotor, and trophic changes.^1-3 Traditionally, the pain is treated by interrupting the sympathetic supply to the painful area. Is this an effective approach?

Periarterial sympathectomy was first used to treat causalgia, in which, by definition, major nerve injury occurs. Various forms of surgical sympathectomy have subsequently been carried out, ^{3 4} especially during war time, when controlled trials were not feasible, and so whether surgery was truly effective will never be known. Open surgical sympathectomy to relieve pain in causalgia and related conditions is rarely recommended now, not least because less invasive proceduresincluding endoscopic sympathectomy and percutaneous radiofrequency lesioning of the sympathetic trunkhave been developed, although critical evaluation of efficacy is awaited.⁴

For several decades, local anaesthetic sympathetic blockade has been undertaken with a variety of techniques.³ Unfortunately few adequately controlled trials have been carried out, and Kozin, in a review of 500 patients treated by sympathetic block, concluded: "The majority of patients have transient or no significant pain relief."² Furthermore, a meta-analysis of randomised controlled trials, retrospective and prospective case series, and controlled studies comprising 1144 patients showed that the benefit of sympathetic blockade with local anaesthetic was indistinguishable from that of placebo.⁵ It is therefore doubtful whether sympathetic blockade should be advocated for relief of chronic pain in causalgia and reflex sympathetic dystrophy.

There seem to be no controlled studies demonstrating efficacy of neurolytic sympathetic blocks. Possible side effects, ranging from trivial to devastating, are of even greater importance with these more permanent procedurespainful sequelae may include phenol or alcohol neuritis and postsympathectomy pain (sympathalgia), which can also occur after surgical sympathectomy.⁶

Peripheral sympathetic blockade with regional intravenous guanethidine infusion has been used for 25 years, but only recently have critical appraisals of benefit been undertaken. Jadad and colleagues foundfrom the few studies sufficiently robust to allow statistical assessment together with their own, subsequently abandoned, randomised controlled trialthat there was no evidence that regional intravenous guanethidine was better than placebo.⁷ Similar conclusions were obtained from a double blind, randomised, multicentre study comparing guanethidine with saline placebo in local anaesthetic.⁸ At present, the evidence seems insufficient to support the use of these peripheral sympatholytic procedures in the routine management of pain.

More recently the  adrenergic blocker phentolamine has been used intravenously as a test of sympathetic nerve involvement in these chronic pains in order to predict the outcome of longer lasting sympathetic blocks.⁹ There have been few studies of the reliability of this procedure, and the contribution of a placebo effect is much debated. ^{10 11} The usefulness of the phentolamine test as a prelude to procedures that are of uncertain benefit is currently unclear.

Thus, in contrast to the pain relief commonly achieved by sympathetic blockade in disorders such as pancreatic cancer and attributable to blocking visceral afferent nerves,⁶ there is little if any evidence that interrupting the sympathetic supply is more effective than placebo in alleviating the pain of causalgia and reflex sympathetic dystrophy. Some individual patients, however, may benefit from sympathetic blockade, and there may also be groups of patients with specific clinical features, in particular allodynia,¹² whose pain is more likely to respond and who perhaps account for those reports of successful relief of pain. Pain relief is, however, invariably unpredictable, of uncertain duration, and inconsistent between the different forms of treatment and when the same treatment is repeated. Even the dogma that early treatment is more successful has been disputed.¹³ The optimal number and frequency of anaesthetic or chemical blocks have not been established; one patient may receive 12 sympathetic blocks while another receives 39 regional guanethidine infusions.¹³ Perhaps offering treatments of even dubious efficacy, or obtaining pain relief by exploiting the placebo effect, is better than doing nothing. All these medical interventions, however, carry risks for the patient and financial implications for all. Efficacy and safety must first be assured, particularly when licensed drugs, such as guanethidine, are administered for unlicensed uses.

The involvement of the sympathetic nervous system in causalgia and reflex sympathetic dystrophy, which forms the rationale for treatment by sympathetic interruption, has been questioned,¹⁴ and the issues discussed here raise further questions. Contrary to predictions from experimental data, interrupting the sympathetic nervous system in practice seems futile for obtaining long term relief of pain in many if not most of these patients. How to identify the minority of patients whose pain might respond to these procedures is the next task, but fresh approaches to management are also required.