Niacin Flushing as Marker of Cannabis Effects on Arachidonic Acid Pathways in Schizophrenia

Subjects Niacin skin tests are performed on acutely ill consecutively admitted schizophrenia patients suffering not more than two psychotic episodes. All meet DSM-IV criteria for paranoide schizophrenia. Diagnosis is made by two independent experienced psychiatrists and further supported by structured clinical interview (SCID IV) (Wittchen et al 1997). Majority of patients is treated mostly with atypical neuroleptic drugs. The patient population is subdivided in one group having used cannabis on a regular basis (≥ 0,5 g/d, at least 3 month) prior to admission, and another group having never used cannabis apart from unique trials. Cannabis consuming patients do not use any other drug or alcohol on a regular basis. Psychiatric symptoms are assessed using Brief Psychiatric Rating Scale (BPRS) (Overall and Gorham 1962), Scale of Assessment of Positive Symptoms (SAPS) (Andreasen 1984), Scale of Assessment of Negative Symptoms (SANS) (Andreasen 1983), and Symptom Check List 1990 Revised (SCL 90-R) (Kaplan et al 1998).

Patients are compared to healthy volunteers recruited by newspaper advertisement including again one group of cannabis users (duration and dose of cannabis use as in patients) and one group without any cannabis experience. Controls are interviewed in-depth to rule out a current psychiatric diagnosis or psychiatric personal or family history. As in patients, SCL 90-R is also applied in controls.

All cannabis-using participants are tested positive for cannabinoids in urine at the time of niacin testing. Subjects with any current or history of skin disorders (eczema, atopical dermatitis, psoriasis) or recent treatment with steroids or non-steroidal antiinflammatory drugs (e.g. acetylsalicylic acid) are excluded from the study before niacin testing. The study is approved by the Ethics Committee of Friedrich-Schiller-University Jena. All participants give written informed consent to participate in the study.

Niacin skin test protocol Methylnicotinate (C7H7NO2, 99%, Sigma-ALDRICH Chemie GmbH, Germany) is applied simultaneously in three dilutions (0.001 M, 0.01 M, 0.1 M) of 50 µl each to the skin at the inner side of the forearm using chambered plaster for epicutaneous testing. After 90 sec the plaster is removed. Skin flushing is quantified before and up to 15 minutes after methylnicotinate exposure in 3-min intervals, starting 90 sec after removal of the methylnicotinate patches. Methylnicotinate solutions are freshly prepared for each test to prevent any influence of sunlight.

Reflection spectroscopy Optical reflection spectroscopy (ORS) is applied as described in more detail by Smesny et al 2001. Skin content of oxygenated blood is assessed with a handheld optical reflection spectrometer (spectral range: 400 nm to 700 nm, area of measurement: diameter 5 mm), using the oxyhemoglobin (HbO2) absorption double peak at 542 nm and 577 nm. Each measurement is repeated three times (within 10 sec) and then averaged.

Spectroscopic data are processed automatically creating difference spectra by subtraction of pre-stimulation reflection intensities (also measured three times) from test intensities. Two Gaussian curves are fitted to the HbO2-absorption double peak. The area under the resulting sum curve is taken as measure of current skin flushing (measured in arbitrary units [a.u.]).

Data analysis We plan to conduct a repeated measure analysis of variance (ANOVA) with TIME (3, 6, 9, 12, 15 min) and methylnicotinate CONCENTRATION (0.001 M, 0.01 M, 0.1 M) as within-subject factors and GROUP (patients, controls) and CANNABIS (cannabis user, cannabis non-user) as between subjects factors. GENDER, and AGE are treated as co-variates. For post-hoc comparison of single values, Mann-Whitney-U-Tests are calculated. Psychopathological ratings of the SCL 90-R are compared between groups using an univariate ANOVA with the same between group factors (GROUP, CANNABIS) and co-variates (AGE, GENDER) as above. Furthermore, effects of cannabis use on SAPS, SANS, and BPRS are investigated within the patient group. In order to explore a possible association between age or psychopathological ratings and skin flushing, Spearman correlation coefficients will be calculated. Due to the high number of calculated coefficients the significance level will be set on p \< 0.01.