The variety of adverse events associated with low-dose methotrexate (MTX) therapy is as interesting and as important as its therapeutic benefits. Curiously, the therapeutic usefulness of MTX in rheumatoid arthritis (RA) was not widely appreciated until the late 1970s, nearly 30 years after it was released as an antineoplastic agent and 20 years after its use in psoriatic disease. Similarly, 20 years after its rediscovery as an effective antirheumatoid agent, our understanding of the nature and mechanisms of its adverse events in patients with RA is still expanding. To complement the other articles in this issue on MTX pulmonary, hepatic, and infectious complications, and on its oncogenic potential, this article reviews the current understanding of gastrointestinal, hematologic, neurologic, and potential vascular adverse events, as well as postdosing reactions, MTX osteopathy, and MTX-induced nodulosis.

A brief review of how medication-related adverse events may be identified, categorized, and quantified provides a useful context. As reviewed by Day et al, adverse drug effects may be classified as <sup>16Day R.O., Quinn D.I., Conaghan P.G., et al. Adverse drug reactions and their measurement in the rheumatic diseases J Rheumatol 1995 ;  22 : 983

Cliquez ici pour aller à la section Références</sup> Type A—dose dependent (e.g., MTX gastrointestinal toxicity); Type B—idiosyncratic (e.g., MTX pneumonitis); Type C—resulting from long-term therapy but anticipated, based on overall drug exposure (e.g., MTX hepatotoxicity); and Type D—delayed effects occurring even after discontinuation of the drug (e.g., MTX in the first trimester of pregnancy, inducing teratogenesis).

Although the most common adverse events are often identified in randomized, controlled trials prior to regulatory approval, uncommon, rare, and sometimes serious adverse events are most often identified post marketing, usually through spontaneous case reports by physicians.<sup>73Venning G.R. Identification of adverse drug reactions to new drugs. II How were 18 important adverse reactions discovered and with what delays? Br Med J 1983 ;  286 : 289 [cross-ref]

Cliquez ici pour aller à la section Références</sup> Because rheumatic diseases like RA require long-term drug therapy, it is particularly important to identify types B, C, and D adverse events. Identification, and therefore frequency and attribution, of an adverse event may be underestimated or in error when the adverse event is common in the healthy population (e.g., fatigue) or common in the disease population (e.g., nodulosis in patients with RA), or commonly associated with concomitant medications (e.g., nonsteroidal anti-inflammatory drug [NSAID] dyspepsia). Fatigue, nodulosis, and dyspepsia also may be caused by low-dose MTX therapy, however. Once an uncommon adverse event is identified, cohort, case–control, post marketing surveillance, and record linkage studies, as well as intensive monitoring, may help in establishing the frequency and risk factors.

Knowledge of MTX pharmacokinetics, reviewed elsewhere in this volume, may be helpful in predicting type A adverse events. Both the total amount and the length of exposure to MTX is correlated with an increased risk of some adverse events.<sup>23Gleyer W.A. The clinical pharmacology of methotrexate: New applications of an old drug Cancer 1978 ;  41 : 36

Cliquez ici pour aller à la section Références</sup> Length of exposure is more important than serum level for inducing some adverse events. Using observations from MTX-treated juvenile RA patients, it has been suggested that some adverse events (e.g., type A) occur at serum levels of MTX ≧ 1.0 × 10⁻⁷ mol/L 36 to 42 hours after ingestion.<sup>74Wallace C.A., Sherry D.D. Practical approach to avoidance of methotrexate toxicity J Rheumatol 1995 ;  22 : 1009

Cliquez ici pour aller à la section Références</sup> These authors suggest that an MTX serum level ≤ 1.0 × 10⁻⁷ mol/L 24 hours post dosage will greatly reduce type A adverse events, such as gastrointestinal symptoms and perhaps myelosuppression (Figure 1). The reality, however, is more complex because nausea may be unrelated to MTX serum levels, and the resolution of some MTX-associated adverse events by using supplemental folate cannot be explained in these terms.