Article Text

Download PDFPDF

Peripheral neuropathy in patients with systemic rheumatic diseases treated with leflunomide
  1. C Metzler1,
  2. A C Arlt2,
  3. W L Gross1,
  4. J Brandt1
  1. 1Department of Rheumatology, University of Schleswig-Holstein, Campus Luebeck and Rheumaklinik Bad Bramstedt, Germany
  2. 2Department of Neurology, Rheumaklinik Bad Bramstedt, Germany
  1. Correspondence to:
    Dr C Metzler
    University Hospital of Schleswig-Holstein, Campus Luebeck, Department of Rheumatology, Ratzeburger Allee 160, D-23538 Luebeck, Germany;

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Leflunomide (LEF) was introduced as a new disease modifying antirheumatic drug in 1998. Up to now hepatotoxicity, hypertension, and diarrhoea have been reported as adverse events. Peripheral neuropathy (PNP) has been described in eight patients with rheumatoid arthritis (RA) and psoriatic arthritis.1–,3 In the postmarketing surveillance PNP does not appear as a side effect.4

This first retrospective study aimed at evaluating whether LEF might be associated with PNP in a large cohort of patients with inflammatory rheumatic diseases.

All inpatients of a primary rheumatology clinic with systemic rheumatic diseases and treatment with LEF between August 1998 and May 2004 were retrospectively screened for a new onset of PNP. Patients with definite reasons for PNP—for example, active vasculitis or collagenosis, diabetes mellitus, lack of vitamins, alcohol abuse, and neoplasms, were excluded. PNP was screened clinically and confirmed by electrophysiological examination.

Seven hundred and eighty five patients with LEF treatment were identified. PNP was diagnosed in 106 (13.5%) patients. Ninety five patients were excluded with diabetes (22 patients), malignoma with cytotoxic treatment (2 patients), vasculitis (67 patients), and connective tissue diseases (4 patients). None of the patients with PNP and systemic rheumatic diseases potentially associated with PNP had any history of neuropathy or any clinical or serological signs of disease activity at the time of the first manifestation of PNP and were, for example, in remission.

Eleven of 785 (1.4%) patients were identified as having PNP that was possibly related to LEF (table 1). These 11 patients received LEF in a dose of 20–50 mg/day by mouth daily for RA in six patients, Wegener’s granulomatosis in two, and for other inflammatory rheumatic diseases in three patients, respectively. PNP had occurred within a median of 9 months (range 2–48) after the start of LEF. Electroneurography and electromyography showed that PNP related to LEF had an acrodistal symmetric pattern in all patients. There was evidence of a mostly axonal affection in seven patients and demyelising changes in two. Patients with PNP induced by LEF were followed up for a median of 12 months (range 2–45). In eight patients LEF was stopped and, in addition, washed out in three of them. PNP improved in 3/8 (37.5%) patients when LEF was stopped and remained stable in eight, independently of stopping or continuing treatment with LEF.

Table 1

 Characteristics of 11 patients with peripheral neuropathy (PNP) related to treatment with leflunomide (LEF)

This retrospective analysis supports an association between treatment with LEF and the occurrence of PNP. Comparable data about the epidemiology of PNP are rare; in 1992 Walters et al described a prevalence of 2.9% in a cohort of 480 healthy controls.5 Whether there is a dose dependent neurotoxic effect of LEF is unclear. One patient of our cohort established a PNP after an accidental intake of a higher dose LEF (50 mg/day). In contrast with this experience, no increase of PNP was seen in two different studies in which 20 patients with Wegener’s granulomatosis were treated with 30–40 mg/day LEF for 2 years6 and in a preliminary report of 11 patients with RA who received LEF in a dosage of 40 mg for a mean period of 4.4 months.7

The underlying pathogenetic mechanism is still unknown. When LEF was first used for the treatment of RA the possibility that LEF might induce vasculitis was discussed after a report of a new onset of secondary vasculitis in two patients with RA.8 In the meantime further safe and effective treatment of primary systemic vasculitides did not support this hypothesis.6 Thus one might argue that LEF was not effective enough to suppress a secondary vasculitis in these two cases. Interestingly, in three patients with RA and LEF related PNP nerve biopsies were performed showing a predominant axonopathic process and vasculitis of the arterioles.

In conclusion, PNP is a new adverse event of LEF. Therefore we recommend close neurological monitoring during treatment with LEF. In suspect cases stopping and washing out of LEF by cholestyramine should be considered.