Objective: To estimate the incidence of cancer (especially lymphoproli
ferative malignancies) in patients with the Felty syndrome. Design: A
retrospective cohort study. Setting: A computerized database of all di
scharge records for 1969 to 1990 from a Veterans Affairs hospital. Pat
ients: 906 men with a discharge diagnosis of the Felty syndrome. Measu
rements: Standardized incidence ratios (SIR) (ratios of observed-to-ex
pected events) estimated the risk for specific cancers. Hospital recor
ds confirmed the diagnoses of the Felty syndrome and cancer. Results:
We observed a twofold increase in total cancer incidence (137 patients
; SIR = 2.09; 95% CI, 1.8 to 2.5). The risk for non-Hodgkin lymphoma (
19 patients; SIR = 12.8, CI, 7.7 to 20.0) was much greater than the tw
ofold increase in risk for lymphoma generally reported for rheumatoid
arthritis. The risk for leukemia was increased but only within 5 years
of the first hospitalization for the Felty syndrome, (13 patients; SI
R = 7.67; CI, 4.1 to 13.1). Conclusion: The increased risk for non-Hod
gkin lymphoma after the Felty syndrome in our study is similar to the
risk associated with the Sjogren syndrome and may reflect similar immu
nostimulatory mechanisms.