LETTER TO THE EDITOR - Reducing the risk of MTX pneumonitis in RA

[Guest guest]

Rheumatology 2004 43(8):1059-1060; doi:10.1093/rheumatology/keh223

Rheumatology Vol. 43 No. 8 © British Society for Rheumatology 2004;

all rights reserved

--------------------------------------------------------------------------------

Letter to the Editor

Reducing the risk of methotrexate pneumonitis in rheumatoid arthritis

A. R. Clewes and J. K. Dawson

Department of Rheumatology, St Helens and Knowsley Hospitals NHS

Trust, St Helens, Merseyside WA9 3DA, UK

SIR, We read with interest the article by Saravanan and [1]

concerning pre-existing lung disease and the risk of methotrexate

pneumonitis (MTX-P) in rheumatoid arthritis (RA). We agree that a

number of published studies aimed at identifying risk factors for

MTX-P in patients with RA seemed to show that pre-existing lung

disease is associated with an increased risk of MTX-P, but we wish to

emphasize that there have been no consistent findings concerning

interstitial lung disease (ILD) as a specific risk. Furthermore, we

would suggest that the meta-analysis by Saravanan and [1] at the

very least exaggerates the association between pre-existing lung

disease and MTX-P. Two large studies have not been included in the

meta-analysis. Cottin et al. [2] studied 124 patients commencing

methotrexate (MTX) and recorded four patients as having MTX-P; none

had pre-existing lung disease on chest X-ray (CXR) or baseline

abnormalities in full pulmonary function tests (PFTs). Carson et al.

[3] also found no association between existing lung disease and MTX-P

in a retrospective study of 168 subjects, although pre-existing lung

disease was only ascertained by questionnaire. It may seem that a

questionnaire is a superficial assessment of pre-existing lung

disease. However, Saravanan and do not mention that the

radiological studies have large numbers of missing baseline chest

radiographs. Only 26 of the 111 (23%) subjects in the study by Alarcon

et al. [4] had a baseline CXR. There was no baseline CXR in 19 of 125

(15%) patients in the study by Golden et al. [5]. This casts doubt on

the suggested odds ratio for this particular meta-analysis, as the

authors would appear to have assumed that an absent CXR is equivalent

to no pre-existing lung disease on CXR.

***********************************

Read the entire letter here:

http://rheumatology.oxfordjournals.org/cgi/content/full/43/8/1059

Not an MD