Anti-inflammatory drugs and psychosis

Abstract

This thesis focuses on the disorder we know as schizophrenia. Although there is treatment for schizophrenia in the form of anti-psychotic drugs, not all patients respond well to this treatment. A large part of patients will have remaining symptoms for the rest of their lives. A number of hypotheses have been made, trying to explain the pathophysiology of schizophrenia. One of these hypotheses is that inflammations may in part be involved in the disease. Inflammations can be inhibited by non-steroidal anti-inflammatory drugs (NSAIDs) and with glucocorticosteroids (GCS). The aim of this thesis was to investigate the relation between these anti-inflammatory drugs and schizophrenia. We first examined the financial burden of psychotic and other psychiatric disorders in the Netherlands. In total there are an estimated 21.000 patients with a psychotic disorder between the ages of 18 and 65. The associated healthcare costs are estimated at 147 million euros yearly. This results in around 7000 euros healthcare costs per case of schizophrenia per year. Two retrospective case-control studies were performed. The first investigated whether the use of NSAIDs was associated with psychosis. We indeed found a decreased risk for incident anti-psychotic use in men, however not in women. The relative risk in men was 0.41 (95% confidence-limits (CI): 0.17–0.97), when results were adjusted for age and total prescription volume. The results suggest that the use of NSAIDs is associated with a decrease risk of subsequent psychosis. The second retrospective study investigated the association between GCS and psychosis. This time a psychosis was defined by a new diagnosis of a psychotic disorder as registered in a psychiatric case register. Again we found an association between the anti-inflammatory GCS and a new diagnosis of a psychotic disorder for men only. The crude relative risk for men using systemic or inhaled GCS was 0.52 (95% CI: 0.36–0.75). This association was present also in the youngest subgroup and a dose-response relation appeared to be present. The main study in this thesis was a double-blind placebo controlled randomized adjuvant trail investigating the effect of the NSAID aspirin. During three months of follow up participants with a schizophrenic disorder were randomized to either aspirin or placebo, together with pantoprazole. The severity of symptoms was estimated using the positive and negative symptom scale (PANSS). The change in total PANSS score was the primary outcome, with secondary outcomes being changes on its subscales and on a number of cognitive tests. We found a significant monthly treatment effect of -1.62 (95% CI -2.93;-0.30) points on the total PANSS and on the positive PANSS subscale (-0.52 (95% CI -1.02;-0.02). The effect was substantially larger in patients with a lower immune ratio, as indicated by TH1/TH2 immune balance (p=0.018). Trends for the negative and general PANSS sub-scores were similar to the trends in the total and positive scores, without reaching statistical significance. Aspirin did not affect cognitive functions. The results of this randomized trial indicate that aspirin may be a beneficial therapy in combination with anti-psychotic drugs in schizophrenia.