OBJECTIVES Prospectively collected computer database information was previously assessed on a cohort of 300 patients who fulfilled the Copenhagen classification criteria for primary Sjögren's syndrome. Analysis of the clinical data showed that patients who smoked had a decreased lower lip salivary gland focus score (p<0.05). The aim of this original report is to describe the tobacco habits in patients with primary Sjögren's syndrome or stomatitis sicca only and to determine if there is a correlation between smoking habits and focus score in lower lip biopsies as well as ciculating autoantibodies and IgG.

RESULTS The questionnaire was answered by 98% (n=355) of the cohort and the percentage of current smokers, former smokers and historical non-smokers at the time of lower lip biopsy was not statistically different from that of the control group. Cigarette smoking at the time of lower lip biopsy is associated with lower risk of abnormal focus score (p<0.001; odds ratio 0.29, 95%CI 0.16 to 0.50). The odds ratio for having focal sialadenitis (focus score > 1) compared with having a non-focal sialadenitis or normal biopsy (focus score ⩽ 1) was decreased in all three age groups (10–45: odds ratio 0.27, 95%CI 0.11 to 0.71; 46–60: odds ratio 0.22, 95%CI 0.08 to 0.59; and ⩾ 61: odds ratio 0.36, 95%CI 0.10 to 1.43) although there was only statistical significance in the two younger age groups. Moreover, among current smokers at the time of the lower lip biopsy there was a decreasing odds ratio for an abnormal lip focus score with increasing number of cigarettes smoked per week (p trend 0.00). In the group of former smokers, which included patients that had stopped smoking up to 30 years ago, the results were in between those of the smokers and the historical non-smokers (odds ratio 0.57, 95%CI 0.34 to 0.97, compared with never smokers). Present or past smoking did not correlate with the function of the salivary glands as judged by unstimulated whole sialometry, stimulated whole sialometry or salivary gland scintigraphy. Among former smokers, the median time lapse between the first symptom of primary Sjögren's syndrome and the performance of the lower lip biopsy was approximately half as long as the median time lapse between smoking cessation and biopsy (8 versus 15 years). Hence, symptoms of Sjögren's syndrome are unlikely to have had a significant influence on smoking habits at the time of the biopsy. Among the seroimmunological results only anti-SSA/Ro and anti-SSB/La antibodies reached statistical significance in a manner similar to the way smoking influenced the focus score in lower lip biopsies. On the other hand the level of significance was consistently more pronounced for the influence of smoking on the focus score than for the influence on anti-SSA/Ro and anti-SSB/La autoantibodies.

CONCLUSION This is believed to be the first report showing that cigarette smoking is negatively associated with focal sialadenitis—focus score >1—in lower lip biopsy in patients with primary Sjögren's syndrome. Furthermore, tobacco seems to decrease the focus score in a dose dependent manner. Smoking may also negatively influence the presence of anti-SSA/Ro and/or anti-SSB/La antibodies in circulating blood. Thus, smoking habits of patients might invalidate the use of both lower lip salivary gland focus score and of anti-SSA/anti-SSB antibodies. It is suggested that the simultaneous performance of other objective tests is required to avoid misdiagnosis of oral involvement in smoking and former smoking patients. Therefore, classification criteria for Sjögren's syndrome that more or less rely on an abnormal focus score and/or presence of anti-SSA/anti-SSB antibodies should be used with great caution.