ABSTRACT
INTRODUCTION: We studied the efficacy of St. John's Wort compared with placebo in patients with minor depressive symptoms or dysthymia, with the main focus on which diagnostic entities are optimally amenable to treatment with two different doses of Hypericum, and which are not. METHODS: One hundred and fifty patients, 25-70 years old, meeting ICD-10 criteria for mild or moderately severe depressed episodes or with dysthymia, and having a 17-item Hamilton Depression Scale for Depression (HAM-D) total score between 7 and 17, were randomly assigned to an extract. The extract, PM235, manufactured by Cederroth International AB, Sweden, was given t.i.d. in a lower (0.12% hypericine) or a higher (0.18% hypericine) formulation, based on 270mg extractions or identical placebo. Clinical response was defined by HAM-D as a 50% reduction and/or a score 7. The Beck Depression Inventory (BDI) and Visual Analog Scales (VAS) were used as secondary efficacy parameters. Measures were conducted at screening, baseline, and after 3 and 6 weeks of treatment. RESULTS: We found a large discrepancy in response between dysthymic and non-dysthymics, the latter seemingly more sensitive to Hypericum. HAM-D showed tendency but no significance toward a more frequent improvement of the non-dysthymics treated with Hypericum (p=0.057). BDI criteria showed significance (p=0.045) for both doses of Hypericum compared to placebo. Pooling high- and low-dose groups together, a significant reduction for HAM-D7 and BDI criteria was found among non-dysthymic patients (p=0.03). Significant improvement in response to Hypericum was found in symptoms reflected by VAS - again only in non-dysthymic patients (p=0.041). DISCUSSION: We observed, a tendency toward a more frequent significant improvement of the non-dysthymic patient treated with PM235, though this did not reach the level of statistical significance. In a secondary analysis, pooling both hypericine-treated groups concluded that Hypericum has a clinical significant effect in minor depressed patients with HAM-D up to 17. This finding was significant only in non-dysthymic patients.
