Elias Zintzaras1, Nikolaos Sakellaridis. 1. Department of Biomathematics, University of Thessaly School of Medicine, Papakyriazi 22, 41222 Larissa, Greece. zintza@med.uth.gr
Abstract
OBJECTIVES: Case-control studies relating 472G/A catechol-O-methyl-transferase polymorphism with the risk of developing panic disorder showed inconclusive or contradictory results. To shed some light on these results a meta-analysis of all available case-control studies was conducted. METHODS: We searched PubMed database for English-languages case-control studies using the key words: Catechol-O-methyl-transferase and panic. Case-control studies that determined the distribution of 472G/A genotypes in cases with primary and predominant panic disorder, and in controls free of psychiatric disorders were eligible for inclusion in the meta-analysis. The pooled risk effect (odds ratio, OR) was estimated using fixed effects and random effects (RE) models. The heterogeneity between studies was tested using the Q-statistic, heterogeneity was also quantified with the I metric. Possible sources of bias were also explored. RESULTS: The main analysis for investigating the association of the allele G and the risk of developing panic disorder relative to the allele A, showed significant heterogeneity (PQ<0.01, I=79%) between studies, then the RE OR was nonsignificant, heterogeneity OR=1.04 with 95% confidence interval (0.71-1.53). Although for both sexes the OR was not significant, there is evidence that in females the risk of PD [RE OR=1.07 (0.54-2.11)] was greater than in males [RE OR=0.86 (0.53-1.39)]. Inconsistency in genetic effects between East Asians [RE OR=0.73 (0.41-1.30) and (PQ=0.03, I=73%)] and Caucasians [OR=1.26 (0.93-1.69) and (PQ=0.24, I=28%)] existed. The genotype differences for the homozygotes, the recessive and dominant models for allele G produced the same overall pattern like the allele contrast in terms of association and heterogeneity. No differential magnitude of effect in large versus small studies for each polymorphism investigating was found. The cumulative meta-analysis showed an increase in OR as evidence accumulated. CONCLUSIONS: No conclusive evidence showing that 472G/A polymorphism is a reliable marker for panic disorder was found; moreover, large heterogeneity between studies existed. Large and more rigorous association studies investigating also the interaction with other genetic/environmental factors might provide more conclusive evidence.
OBJECTIVES: Case-control studies relating 472G/A catechol-O-methyl-transferase polymorphism with the risk of developing panic disorder showed inconclusive or contradictory results. To shed some light on these results a meta-analysis of all available case-control studies was conducted. METHODS: We searched PubMed database for English-languages case-control studies using the key words: Catechol-O-methyl-transferase and panic. Case-control studies that determined the distribution of 472G/A genotypes in cases with primary and predominant panic disorder, and in controls free of psychiatric disorders were eligible for inclusion in the meta-analysis. The pooled risk effect (odds ratio, OR) was estimated using fixed effects and random effects (RE) models. The heterogeneity between studies was tested using the Q-statistic, heterogeneity was also quantified with the I metric. Possible sources of bias were also explored. RESULTS: The main analysis for investigating the association of the allele G and the risk of developing panic disorder relative to the allele A, showed significant heterogeneity (PQ<0.01, I=79%) between studies, then the RE OR was nonsignificant, heterogeneity OR=1.04 with 95% confidence interval (0.71-1.53). Although for both sexes the OR was not significant, there is evidence that in females the risk of PD [RE OR=1.07 (0.54-2.11)] was greater than in males [RE OR=0.86 (0.53-1.39)]. Inconsistency in genetic effects between East Asians [RE OR=0.73 (0.41-1.30) and (PQ=0.03, I=73%)] and Caucasians [OR=1.26 (0.93-1.69) and (PQ=0.24, I=28%)] existed. The genotype differences for the homozygotes, the recessive and dominant models for allele G produced the same overall pattern like the allele contrast in terms of association and heterogeneity. No differential magnitude of effect in large versus small studies for each polymorphism investigating was found. The cumulative meta-analysis showed an increase in OR as evidence accumulated. CONCLUSIONS: No conclusive evidence showing that 472G/A polymorphism is a reliable marker for panic disorder was found; moreover, large heterogeneity between studies existed. Large and more rigorous association studies investigating also the interaction with other genetic/environmental factors might provide more conclusive evidence.