The primary SS group accounted for 40% (100/252) of the patients (mean age 61.9 ± 14.5), the secondary SS group for 40% (102/252; mean age 58.5 ± 15.2), and the non-SS group for 20% (50/252; mean age 58.4 ± 16.2; Table 1).

RA accounted for more than half of all autoimmune diseases other than SS in both the secondary SS and non-SS groups (secondary SS: 71.6%, non-SS: 62%), followed by SLE (secondary SS: 19.6%, non-SS: 24%; Table 1). It should also be noted that some patients had multiple autoimmune diseases.

The mean corneal scores in the primary SS, secondary SS, and non-SS groups were 0.77 ± 0.84, 0.92 ± 0.85, and 0.56 ± 0.76, respectively. There was a significant difference in corneal score between the secondary SS and non-SS groups (p = 0.041; Fig. 1a). The mean conjunctival scores in the primary SS, secondary SS, and non-SS groups were 1.52 ± 1.65, 1.88 ± 1.90, and 0.69 ± 1.19, respectively. There were significant differences in conjunctival scores between the primary SS and non-SS groups, and between the secondary SS and non-SS groups (p = 0.003, < 0.001, respectively; Fig. 1b). The mean ST1 values of the primary SS, secondary SS, and non-SS groups were 8.9 ± 9.0, 7.0 ± 6.7, and 10.0 ± 8.7 mm, respectively. There were significant differences in ST1 between the primary SS and non-SS groups, and between the secondary SS and non-SS groups (p = 0.01, 0.004, respectively; Fig. 1c). The FBUT scores of the primary SS, secondary SS, and non-SS groups were 2.7 ± 1.3, 2.5 ± 1.7, and 3.0 ± 2.0 seconds, respectively, with no significant differences between the three groups (Fig. 1d).

Comparison of objective findings between the three patient groups. a corneal score, b conjunctival score, c Schirmer 1 test, d. breakup time. a There was a significant difference in corneal score between secondary SS and non-SS. b There were significant differences in conjunctival score between primary SS, secondary SS, and non-SS groups. c There were significant differences in ST1 between primary SS, secondary SS, and non-SS. d. There were no significant differences between the three groups. Data are expressed as mean ± SD, *p < 0.05 (Mann–Whitney U test). SS, Sjogren’s syndrome

Observation of FBUPs was performed in 81 cases. In the primary SS group, the FBUPs included 12 cases of dimple break (41%, 12/29), 10 of line break (34%, 10/29), 3 of area break (10%, 3/29), 3 of random break (10%, 3/29), and 1 of spot break (3%, 1/29). In the secondary SS group, the FBUPs included 12 cases of line break (36%, 12/33), 8 of dimple break (24%, 8/33), 6 of area break (18%, 6/33), 4 of random break (12%, 4/33), and 3 of spot break (9%, 3/33). In the non-SS group, the FBUPs included 6 cases of random break (32%, 6/19), 5 of dimple break (26%, 5/19), 4 of line break (21%, 4/19), and 4 of spot break (21%, 4/19). No patient showed an area break (Fig. 2). Overall, area and line breaks in the aqueous-deficient type showed frequencies of 11% (9/81) and 32% (26/81), respectively, dimple and spot breaks representing the decreased wettability type showed frequencies of 31% (25/81) and 10% (8/81), respectively, and random breaks indicating the increased evaporation type showed a frequency of 16% (13/81) (Fig. 2).

Classification of fluorescein breakup patterns in patients with autoimmune disease. No area break was observed in the non-SS group. Area break and line break, which reflect tear-deficiency-type dry eye, represented more than half of breaks in the secondary SS group, but less than half in total. More than half of the patients’ breakup patterns were classified as short breakup-time-type dry eye. SS Sjogren’s syndrome

The mean anti-SS-A antibody values of the primary SS, secondary SS, and non-SS groups were 626.5 ± 547.7, 586.3 ± 536.6, and 197.6 ± 399.1 U/mL, respectively. There were significant differences between the primary SS and non-SS groups, and between the secondary SS and non-SS groups (p = 0.024, < 0.001, respectively; Fig. 3a). The mean anti-SS-B antibody values of the primary SS, secondary SS, and non-SS groups were 121.0 ± 311.7, 165.9 ± 346.2, and 20.1 ± 98.6 U/mL, respectively. There were significant differences between the primary SS and non-SS groups, and between the secondary SS and non-SS groups (p = 0.006, 0.001, respectively; Fig. 3b). The ANA-positive rates of the primary SS, secondary SS, and non-SS groups were 92.4%, 91.5%, and 81.3%, respectively. There were no significant differences between the three groups (Fig. 3c). The RFs of the primary SS, secondary SS, and non-SS groups were 44.6 ± 63.2, 119.5 ± 403.5, and 35.6 ± 49.1 IU/mL, respectively. There were no significant differences between the three groups (Fig. 3d).

Comparisons of various autoantibodies between the three patient groups. a anti-SS-A antibody, b anti-SS-B antibody, c positive rate of antinuclear antibody, d. rheumatoid factor. Anti-SS-A and anti-SS-B antibodies showed significant differences between primary SS, secondary SS, and non-SS. The positive rates of antinuclear antibody and rheumatoid factor showed no significant differences between the three groups. Data are expressed as mean ± SD, *p < 0.05 (Mann–Whitney U test). SS Sjogren’s syndrome

There was no significant correlation between ST1 and any of the autoimmune antibodies, nor between ST1 and the conjunctival scores (primary SS: r = 0.183, p = 0.077; secondary SS: r = 0.21, p = 0.051; non-SS group: r = 0.09, p = 0.635). In the primary SS group, there were significant positive correlations between anti-SS-A antibody and corneal scores (r = 0.281, p = 0.005), and anti-SS-A and anti-SS-B antibodies titers and conjunctival scores (SS-A: r = 0.391, p < 0.001; SS-B: r = 0.250, p = 0.013). In the secondary SS group, there were significant positive correlations between all antibodies and conjunctival scores (SS-A: r = 0.352, p < 0.001; SS-B: r = 0.314, p = 0.002; RF: r = 0.280, p = 0.005). In the non-SS group, conjunctival scores showed significant positive correlations with anti-SS-A antibody and RF (SS-A: r = 0.293, p = 0.040; RF: r = 0.271, p = 0.029) (Table 2).