The aim of this study was to determine the frequency of anti-prolactin autoantibodies and the relationship among anti-prolactin autoantibodies, serum prolactin (PRL) levels and lupus activity in paediatric patients with systemic lupus erythematosus (SLE) using a transversal study. One-hundred and three consecutive paediatric SLE patients were tested for serum anti-PRL autoantibodies and PRL levels. Clinical disease activity was scored using the SLEDAI index. Anti-PRL autoantibodies were measured by means of gel filtration. The frequency of anti-PRL autoantibodies was 6.7% (7/103), on the basis of the amount of immunoreactive PRL eluted in molecular weight fraction corresponding to IgG (150 kDa). No anti-PRL autoantibodies were found in normoprolactinaemic patients. By contrast, 21.8% (7/32) hyperprolactinaemic patients (hPRL) had anti-PRL autoantibodies. There was a correlation between anti-PRL autoantibody and serum levels of PRL (r(s) = 0.98, P = 0.0001). Lupus activity was present in 64/103 (62.1%) patients, without a significant difference in the frequency of anti-PRL autoantibodies when compared to inactive lupus (7.8 vs 5.1%, P > 0.05). Higher levels of serum PRL were associated with lupus activity regardless of other variables (39.6% vs 17.9%, P = 0.05). Patients with anti-PRL autoantibodies had higher levels of serum PRL than those without anti-PRL autoantibody (41.85 vs 17.77 ng/ml, P = 0.01) and significantly different frequency of hPRL (100 vs 26%, r = 0.4531, P < 0.001). We have identified a subset of paediatric SLE patients with hPRL and anti-PRL autoantibodies. Anti-PRL autoantibodies were associated with hPRL state and antibody titres correlated positively with serum PRL levels. These data suggest that anti-PRL autoantibodies could be responsible for hPRL in a subset of SLE patients. An increase in serum PRL levels proved to be related to lupus activity, but there was no statistical relationship between anti-PRL autoantibodies and lupus activity.