Calpain (CANP) is a calcium-activated protease present in two isoforms in the neuronal cells, CANP I and CANP II, with different sensitivities to calcium ions. Only CANP I is present in human erythrocytes. Its activity is naturally inhibited by calpastatin (CAST). It has been suggested that CANP and CAST are involved in the regulation of long term potentiation (LTP), the long-lasting enhancement of synaptic transmission related to learning and memory processes. Obsessive-compulsive (OC) patients have been investigated for memory impairments with mixed results, some suggesting deficits in logic memory and in memory for actions .The aim of this study was to explore whether CANP and CAST activities could be peripheral biochemical markers for OCD.Fifteen informed, drug-free, DSM-III-R OC patients and fifteen normal controls matched for age and sex were recruited, excluding subjects with clinically severe memory disfunction and severe medical illness. Severity of OC symptoms in patients was rated by the Yale-Brown Obsessive-Compulsive Scale (Y-BOCS) CANP and CAST were extracted from erythrocytes and their activities measured. Mean age, age at onset, mean duration of illness and the Y-BOCS scores for the sample of patients as a whole and by sex were calculated and compared using non-parametric tests because of the small size of the samples. Mean CANP and mean CAST values from patients and controls were compared by the Student t-test. Correlations (Pearson's) between CANP and CAST values and age, age at onset, duration of illness, Y-BOCS total and subtotal scores were calculated for patients only. CANP activity was significantly higher in patients (t = -2.442, df = 28, p = 02), while there were no differences in CAST activity. There was no significant correlation between CANP values and age, age at onset of OCD, duration of illness, severity of OC symptoms. The same results were found for CAST values. According to the main result from this preliminary study, OC patients show higher CANP values in comparison to matched healthy controls.We might hypothesize that OC patients have some subclinical impairment of memory and that the increase in CANP activity is compensatory, as in elderly controls. Alternatively, higher CANP activity may be an epiphenomenon of the overstimulation of memory processes in OC patients. Meanwhile, the lack of differences in CAST activity between patients and controls and the lack of correlation with severity of OC symptoms remain to be explained.Our findings need further investigation. Replication with larger groups and re-testing after adequate antiobsessional treatment should be useful to clarify the role of CANP and CAST activities in OCD.