Objective To investigate the auditory sensory gating P50 in schizophrenia patients with homicide during three months of follow up. Methods Twenty-five patients with schizophrenia patients with homicide and 27 healthy adults matched with age and education level were recruited in the present study. P50 was recorded in twenty-seven normal controls and twenty-five schizophrenia patients at the baseline. P50 was repeatedly recorded in eleven patients who completed the follow-up at three months following antipsychotics treatment. The psychotic symptoms of patients were assessed with Positive and Negative Syndrome Scale (PANSS) at the same time. Results ①In comparison with normal controls, schizophrenia patients with homicide showed increased S2-P50 amplitude (P<0.01), increased S2/S1 ratio (P<0.01), decreased S1-S2(P<0.05), and 100(1-S2/S1) (P<0.01) at the baseline and after three months treatment. The differences in P50 amplitude, latency, and P50 suppression index between the baseline and after three-month treatment in schizophrenia patients with homicide were not significant (P>0.05). ② Schizophrenia patients with homicide after three-month treatment showed decreased PANSS total scale, positive scale, general psychopathology scale, and six symptoms scales such as lack of response, thought disorder, activation, paranoid, depression, and aggression (P<0.05). ③ No significant correlations were found between the S2/S1 ratio, S2-S1, and 100(1-S2/S1) with disease duration, PANSS scores, and six symptoms scales at either baseline or after three-month treatment (P>0.05). Conclusions Schizophrenia with homicide has sensory gating deficits and P50 suppression index might be a stable trait maker for schizophrenia with homicide.