| Schizophrenia is one of the most common and serious mental diseases, the pathophysiological damage of which mainly occurs at early stage, therefore if patients take the suitable antipsychotics earlier, the symptoms might be largely alleviated. At the moment, the atypical antipsychotic clozapine and Risperidone have been applied as the first line clinical drugs for schizophrenia treatment. However, the quality of research reports on efficacy of Clozapine and Risperidone is still unsatisfied because of the low number of the cases included. Besides, the research data of large- sample randomized controlled trial are rarely found. ObjectiveTo compare the efficacy and safety of Clozapine and Risperidone for the first-case schizophrenia and assess the differences on efficacy and safety between the two drugs by using Meta-analysis, the core method of evidenced-based medicine, with a larger sample and less differences caused by random error. MethodsA thorough literature search was performed among Ovid-medline, Pubmed, CBMdisc, CNKI, Wanfang database through internet and professional magazines of psychiatry by hand, aiming at the efficacy and safety of Clozapine and resperidone for the first-case schizophrenia by randomized controlled trial or clinical controlled trial from1998 to 2008 in the openly published journals both in China and abroad. All the literatures are further selected by rigid criteria formed in advance to include or exclude samples. The included literatures were evaluated by Jadad Quality Scale and the data was extracted by self-designed data-extracting tables. All the data extracted was put in RevMan 5.0 software and analyzed by it. The continuous variable was analysized by WMD, while the binary variable was analyzed by OR. The research heterogeneity was judged by the results of heterogeneity test-P value, which manifested in the forest plots. The combined statistic was calculated by M-H or IV solid-effect model for the index with homogeneity, and by M-H or D-L random-effect model for the index with heterogeneity. Statistic significance is judged by the Z value or P value resulting from the combined statistic manifested in the forest plots. The sensitivity analysis was performed in solid-effect model and random-effect model at the same time, either by OR and RR, or by WMD and SMD. The publication bias was analyzed by the degree of asymmetry manifested in the funnel plots.ResultsAccording to the inclusion and exclusion criteria, 23 papers were finally included, 22 of which in Chinese and 1 in English. All the papers included do not mention blind method and 11 gave a detailed description of follow-up failure and withdrawal. Only 3 papers were high quality research scored from 3 to 5, while 20 papers were low quality scored from 0 to 2 by Jadad Quality Scale.It is manifested that from the combined analysis of the therapeutic effect, total effective rate, effective rate, cure rate, total point of BPRS table and scores in PANSS all have the homogeneity(P>0.05) and have no significant difference between clozapine and Risperidone groups (Z=0.36,P=0.72,Z=0.20,P=0.84,Z=0.30,P=0.77). The scores in PANSS, positive symptoms, negative symptoms had no significant difference between clozapine and Risperidone groups (Z=0.72,P=0.47;Z=0.20,P=0.84;Z=0.93,P=0.35). However, the scores in BPRS table had a significant difference (Z=2.21,P=0.03). The results of sensitivity analysis shown that all the results were stable and had no significant difference except the scores in BPRS table. The publication bias analysis shown that the curative rate, the scores of negative symptoms and positive symptoms in PANSS table were asymmetry in forest plots, suggesting the publication bias. While other index shown no publication bias.The side effect analysis shown that lethargy, drowsiness, constipation, sialorrhea, blurred vision, pernicious vomiting, eeg abnormal, tachycardia, ecg abnormal, dizziness, abnormal hemogram, blood pressure change (decrease), body weight change (increase), were statistic significant(P<0.05), those in Risperidone were lower than those in Clozapine group. however other side effects including insomnia, tremor, muscle tension, had a higher rate in Risperidone than in Clozapine (P<0.05). The CHOL-combined analysis by the combined- WMD shown the significant difference between the two treatment groups [95%CI= -0.81to-0.13]. In the self-control analysis, i.e., comparison of data before and after the treatment, both Risperidone and Clozapine could cause the CHOL level-up, and the combined-WMD had a significant difference[95%CI =-0.49to-0.04; 95%CI=-1.21to-0.16]. For the TG-combined analysis between the two treatment groups, the combined-WMD had a significant difference [95%CI =-0.46to-0.06], while for the comparison of data before and after the treatment the combined-WMD had a significant difference [95%CI= -0.16to-0.10; 95%CI=-0.55to-0.14]. For the FPG-combined analysis between the two treatment groups, the combined-WMD had no significant difference [95%CI=-0.50to 0.03], while for the comparison of data before and after the treatment, the combined-WMD in Risperidone had no significant difference [95%CI= -0.44to0.02],while the combined-WMD in Clozapine had a significant difference [95%CI=-0.90to-0.15]. ConclusionBy Meta-analysis, it can be easily seen that Risperidone had no difference from Clozapine on efficacy for schizophrenia based on the conclusions that the therapeutic effect, total effective rate, effective rate, cure rate, total point of BPRS table and scores in PANSS all have the homogeneity and had no significant difference. While the effective therapy of Risperidone lasts longer than that of Clozapine as a result that the BPRS scores in Risperidone is lower than in Clozapine. In the safety analysis of Clozapine and Risperidone for the first-case schizophrenia, the incidence of insomnia, tremor, muscle tension in Risperidone group is higher than in Clozapine, but the incidence of other adverse effect is lower. Both of the drugs can cause the patients'blood glucose and fat to go up, but Clozapine affects more than Risperidone. Since the including criteria is in a low quality, and there could be publication bias existing in some results, the findings are still need to be confirmed by high-quality clinical controlled trial and Meta-analysis.
|
PDF Full Text Request