ObjectiveCitalopram, a highly potent SSRI, is effective in the treatment of depressive disorders and obsessive-compulsive disorder (OCD); however, very few studies have reported concentration-effect relationships for SSRIs. The aim of this study was to investigate the relationship between citalopram concentrations and clinical response in patients with OCD.Methods and study designFifteen patients (aged 18−65 years) with a DSM-IV diagnosis of OCD were included in this open-label, single-blind study. Citalopram was started at a dosage of 20 mg/day; the dosage was increased to a maximum of 60 mg/day by the third week, on the basis of clinical need and tolerability. The dosage then remained unchanged until the end of the 10-week study. Clinical assessments were made at baseline, weekly for the first four weeks and then at weeks 6, 8 and 10. The assessment scales used were the Yale-Brown Obsessive-Compulsive Scale (Y-BOCS), the Clinical Global Impression Scale (CGI) and the Hamilton Depression Rating Scale (HDRS). Plasma citalopram concentrations were determined using a high performance liquid chromatography method after solid phase extraction.ResultsOne patient was withdrawn from the study because of poor compliance. Of the 14 patients who completed the study, nine did not meet the treatment response criterion of an improvement of >25% from the baseline total Y-BOCS score and a score of ≤3 for the global improvement item of the CGI (these patients were termed non-responders), while five did (responders). There were no differences in the main demographic and baseline clinical variables between responders and non-responders. Steady-state citalopram concentrations were similar in the two groups, suggesting that the anti-obsessional effects of citalopram were not related to pharmacokinetic differences between responders and non-responders. There was no linear relationship between steady-state citalopram concentrations and response. The citalopram concentrations and Y-BOCS scores of individual responders obtained at baseline and various study timepoints showed a sigmoid relationship when analysed using the Emax(maximum change in Y-BOCS score) model, with a mean EC50value (drug concentration that elicits 50% of the Emax) of 152 μg/L, whereas a similar analysis of the non-responders generated a flat line.ConclusionThe results of this preliminary study suggest that plasma citalopram concentrations may be related to the clinical response in responders, but do not seem to account for the lack of clinical effect in non-responders. These data, as well as the usefulness of the model in relating plasma concentrations to response, even after repeated administration, need to be validated by further investigations.