Objective: To measure the level of serum Semaphorin 3A (Sema3A) and to analyze the relationship between serum Sema3A and systemic lupus erythematosus (SLE) with thrombocytopenia. Methods: The concentration of serum Sema3A was detected by enzyme-linked immuno sorbent assay (ELISA) in 170 SLE patients, 50 Sjögren’s syndrome (SS) patients, 19 hypersplenism (HS) patients and 150 healthy controls (HC). Based on the presence of thrombocytopenia and whether the thrombocytopenia was in remission, the SLE patients were divided into three groups: SLE with thrombocytopenia (41 cases), SLE with thrombocytopenia remission (28 cases), and SLE without thrombocytopenia (101 cases). According to whether there was thrombocytopenia, the SS patients were divided into SS with thrombocytopenia (18 cases) and SS without thrombocytopenia (32 cases). The 28 SLE patients who underwent bone marrow aspiration biopsy were divided into two groups from the aspect of whether the bone marrow hyperplasia was normal (19 cases) or low (9 cases), as well as from the aspect of whether the maturity disturbance of megakaryocyte was positive (8 cases) or negative (20 cases). The serum Sema3A levels in SLE, SS, HS with HC were compared, meanwhile, the correlation between serum Sema3A level and platelet (PLT) in the patients with different diseases analyzed. Results: (1) Serum Sema3A levels in SLE were significantly lower than in HC [(3.84±2.76) μg/L vs. (6.96±2.62) μg/L, P<0.001], serum Sema3A levels in SS were also obviously lower than in HC [(4.35±3.57) μg/L vs. (6.96±2.62) μg/L, P<0.001], and in HS it was lower than HC at a certain extant [(5.67±2.26) μg/L vs. (6.96±2.62) μg/L, P=0.041]. (2) Serum Sema3A levels in SLE were slightly lower than in SS, but there was no significant difference [(3.84±2.76) μg/L vs. (4.35±3.57) μg/L, P=0.282]. However, when compared with HS, serum Sema3A levels in SLE were significantly lower [(3.84±2.76) μg/L vs. (5.67±2.26) μg/L, P=0.006]. (3) Serum Sema3A concentration in SLE with thrombocytopenia was significantly lower than in SLE with thrombocytopenia remission [(1.28±1.06) μg/L vs. (3.83±2.65) μg/L, P<0.001], and in SLE patients without thrombocytopenia [(1.28±1.06) μg/L vs. (4.87±2.60) μg/L, P <0.001]. There was no significant difference between SLE with thrombocytopenia remission and SLE without thrombocytopenia [(3.83±2.65) μg/L vs. (4.87±2.600 μg/L, P=0.123]. Serum Sema3A concentration in SLE with thrombocytopenia was slightly lower than in SS with thrombocytopenia, but there was no significant difference [(1.28±1.06) μg/L vs. (1.68±1.11) μg/L, P=0.189]. (4) Strong positive correlations were found between serum Sema3A and PLT in SLE (r=0.600, P<0.001). Positive correlations were also found between serum Sema3A and PLT in SS (r=0.573, P<0.001). However, there was no such correlation showed in HS patients (P=0.393). (5) There was no significant difference of serum Sema3A concentration in SLE whether the bone marrow hyperplasia was normal or low. And the same situation appeared in the patients whether the maturity disturbance of megakaryocyte was positive or negative (P>0.05). Conclusion: Serum Sema3A was significantly reduced in SLE patients, and it was highly correlated with the blood damage. Similar conclusions could be drawn in patients with SS. The serum level of Sema3A was generally decreasing in desmosis which merged thrombocytopenia, and was obviously positive correlated with platelet counts.