Objectives This study compared the clinical and serological characteristics of seronegative and seropositive primary Sjogren syndrome (pSS) and examined whether current classification criteria for pSS cover seronegative pSS. Methods The study group comprised 375 patients (341 women and 34 men) diagnosed with pSS. A clinical diagnosis by an expert rheumatologist was considered the "gold standard" for the diagnosis of pSS. The clinical and serological characteristics of the patients were retrospectively collected from hospital medical files. Results Fifty-eight of the 375 pSS patients (15.5%) were seronegative for ANA, RF, anti-Ro, and anti-La autoantibodies. Seronegative pSS was diagnosed based on lymphocytic infiltrations in lip biopsy samples. There were no statistically significant differences in terms of patient age, age at diagnosis, sex distribution, clinical features, and laboratory findings between seronegative and seropositive pSS. The frequency of hypergammaglobulinemia was higher in seropositive pSS. The 2016 ACR/ULAR criteria best covered most seronegative pSS cases (84.5%). For seronegative pSS, the agreement between the 2002 AECG, 2012 ACR, and 2016 ACR/EULAR criteria was relatively low. Conclusions The clinical features of seronegative pSS (i.e., a lack of four autoantibodies in serum) were similar to those of seropositive pSS. The current classification criteria for pSS should not be used in the diagnosis of seronegative pSS, as the agreement between the different sets of criteria was low, and some patients fell outside the classification. Further clinical and laboratory studies are needed to identify the features that distinguish seronegative pSS.