Associations between estrogen receptor-beta polymorphisms and endometriosis risk: a meta-analysis

Diagnostic Pathology

Table 2 Summary ORs and 95% CIs of the association between ER-β rs4986938 polymorphism and endometriosis

Comparison	No. of studies	A vs. G^b		AA vs. GG^a,b		GA vs. GG^b		AA + GA vs. GG		AA vs. GA + GG^a,b
Comparison	No. of studies	OR (95% CI)	P _H	OR (95% CI)	P _H	OR (95% CI)	P _H	OR (95% CI)	P _H	OR (95% CI)	P _H
Overall	8	1.16 (0.82-1.64)^R	<0.001	0.95 (0.46-1.97)^F	0.735	1.24 (0.80-1.91)^R	<0.001	1.21 (0.83-1.77)^R	<0.001	0.86 (0.41-1.77)^F	0.790
ethnicity
Asia	5	0.82 (0.65-1.04)^F	0.552	0.75 (0.27-2.07)^F	0.335	0.81 (0.62-1.05)^F	0.456	0.85 (0.67-1.08)^F	0.496	0.91 (0.26-3.13)^F	0.331
Mixed	3	1.72 (1.37-2.17) ^{c ,F}	0.204	1.23 (0.43-3.49)^F	0.795	2.08 (1.59-2.71) ^{c ,F}	0.159	2.03 (1.56-2.64) ^{c ,F}	0.148	0.97 (0.34-2.74)^F	0.880
Sample size
>300	5	1.26 (0.82-1.93)^R	<0.001	0.83 (0.34-1.99)^F	0.611	1.41 (0.82-2.40)^R	<0.001	1.38 (0.81-2.34)^R	<0.001	0.71 (0.30-1.73)^F	0.717
-300	3	0.94 (0.62-1.41)^F	0.412	1.31 (0.34-5.07)	-	0.86 (0.53-1.39)^F	0.498	0.97 (0.67-1.39)^F	0.624	1.32 (0.35-5.06)	-

OR, odds ratio; CI, confidence interval; No., number; P_H, p value of Q-test for heterogeneity test; R, random-effect model; F, fixed-effect model.
^aThe study by Gu et al. was not included since they presented 0 frequency of AA genotype in cases and controls.
^bThe study by Wu et al. was not included since they just presented the data on GG and GA + AA genotypes in cases and controls.
^cStatistically significant results (in bold).

ISSN: 1746-1596