Amongst the patients attending our combined oral medicine/rheumatology clinic, we have identified a subset presenting with xerostomia due to non-specific sialadenitis, who also suffer from generalized nodal osteoarthritis (NOA). We have called this combination SOX syndrome: sialadenitis, osteoarthritis and xerostomia. In this study, we have examined the characteristics of these patients clinically and histologically, and then determined the prevalence of SOX syndrome in patients with NOA compared to healthy age-matched controls. Patients were obtained from rheumatology clinics and a local old people's home. The series consisted of 35 patients with NOA and 18 age- and sex-matched controls without evidence of NOA or inflammatory rheumatic disease. There was no significant difference in age and sex between the two groups. None were on drugs known to induce xerostomia. The subjects were assessed for whole salivary, parotid saliva and lacrimal flow, autoantibodies, rheumatoid factor (RF) and erythrocyte sedimentation rate (ESR). The whole saliva flow (mean +/- 95% CI) was 0.32 +/- 0.07 ml/min for the NOA group and 0.54 +/- 0.17 ml/min for the control group. The difference is statistically significant (P < 0.05, two-tailed Student's t-test). No statistically significant difference was found in the parotid and lacrimal flow rates of NOA and controls. Nine of the 35 NOA patients had reduced whole salivary flow (normal range > 0.02 ml/min) compared with only one out of 18 in the control group (P > 0.05, chi 2 test). All NOA patients with xerostomia and reduced whole salivary flow were RF, anti-Ro and anti-La negative, and had a normal ESR. Thus, 25% of subjects with NOA had clinical and laboratory features of SOX syndrome, suggesting that this is a defined disease entity.