Arthritis & Rheumatism, Volume 62,
November 2010 Abstract Supplement

Abstracts of the American College of
Rheumatology/Association of Rheumatology Health Professionals
Annual Scientific Meeting
Atlanta, Georgia November 6-11, 2010.

Observational Study To Evaluate the Relationship of Methotrexate to Interstitial Lung Disease in Rheumatoid Arthritis.

Caplan7,  Liron, Hines15,  Anne E., McDonald10,  Jay R., Zeringue10,  Angelique L., Mann6,  Alyse, Sufka13,  Paul H., Osgood2,  Kevin G.

Albancy VAMC
St. Louis VAMC
Univ of Colorado
Univ of Florida Jacksonville, Saint Augustine, FL
Univ of Minneapolis, Minneapolis, MN
Univ of Nebraska Med Ctr, Omaha, NE
University of Colorado
Austin Diagnostic Clinic, Austin, TX
Birmingham VAMC, Minneapolis, MN
Dallas VAMC, Dallas, TX
Denver Health
Denver VAMC
Denver VAMC, Aurora, CO
Portland Rheum Clinic LLC, Lake Oswego, OR
St. Louis VAMC, St. Louis, MO


The relationship of methotrexate (MTX) to interstitial lung disease (ILD) remains controversial, with some studies supporting an association, while others dispute this association. We sought to identify risk factors for incident ILD in a population of veterans with rheumatoid arthritis (RA), with special attention on the disease modifying anti-rheumatic drug (DMARD) methotrexate (MTX).


We used a retrospective case-control study design where controls consist of subjects with RA and cases consist of subjects with both RA and ILD. We identified subjects based on ICD-1395-CM diagnostic codes and prior DMARD or glucocorticoid treatment (n=442). Validation of ICD-1395-CM codes for ILD relied upon medical records, as well as radiographic and pulmonary function results. All demographics, putative risk factors, and pharmaceutical data were abstracted using a detailed review of the medical record and defined abstraction algorithms. We performed unconditional multivariate logistic regression, controlling for multiple RA and ILD risk factors. Methotrexate was forced into the final model as the variable of interest. All analyses were performed using Stata software version 10.2. Confidence intervals were established at 95%.


In the multivariate regression adjusting for all covariates, patients with ILD were no more likely to have been exposed to MTX than patients without MTX exposure. Conversely, rheumatoid factor, C reactive protein and history of emphysema were all associated with presence of ILD (see table).


Though markers of RA disease severity and activity are associated with ILD in patients with RA, MTX exposure does not demonstrate this relationship in our observational study. There is a strong relationship between prior emphysema and subsequent ILD in RA patients.

Table 1. Factors associated with the odds of interstitial lung disease in rheumatoid arthritis patients

  Multivariate ModelFinal Model
Conceptual Block/DomainVariableOdds RatiopLower 95% CIUpper 95% CIOdds RatiopLower 95% CIUpper 95% CI
Generic ConfoundersAge at Cohort Entry, years0.960.2740.891.03    
 Gender, male0.960.9780.0712.91    
Rheumatoid Arthritis-Biologic Therapy0.500.3880.102.41    
  associated VariablesRheumatoid Factor level*1.030.0441.
C Reactive Protein, mg/dL**1.210.0011.081.361.19<0.0011.081.31 
ILD-associated VariablesHistory of Emphysema17.140.0092.06142.3114.370.0022.5680.62
 History of Previous Pneumonia0.500.4520.083.02    
 History of Recurrent Pneumonia15.780.1440.39639.37    
 History of COPD3.410.1430.6617.58    
 History of Bronchitis2.890.2000.5714.68    
 Currently smoking0.560.5260.093.38    
DMARD exposureMethotrexate Exposure0.480.3860.092.530.620.5630.123.11
Chi2 (df3) = 23.9; Pseudo R2= 0.24 for final model

To cite this abstract, please use the following information:
Caplan, Liron, Hines, Anne E., McDonald, Jay R., Zeringue, Angelique L., Mann, Alyse, Sufka, Paul H., et al; Observational Study To Evaluate the Relationship of Methotrexate to Interstitial Lung Disease in Rheumatoid Arthritis. [abstract]. Arthritis Rheum 2010;62 Suppl 10 :1395
DOI: 10.1002/art.29161

Abstract Supplement

Meeting Menu