The second method was defining nuclear and cytoplasmic staining as positive separately in IHC examination, which was used only in 3 studies. We made an effort to contact all primary authors of studies by e-mail to standardize their data according to the meta-analysis definitions whenever possible. In the present study, only nuclear staining was regarded as positive
[18–20]. All data were extracted independently by 2 reviewers (Wang XT and Kong FB) according to the prespecified selection criteria. The following data were extracted: the year of publication, first author’s surname, number of cases and controls, and numbers of different clinical and pathologic parameters. Statistical analysis VX-770 molecular weight Results were expressed with risk ratio (RR) for dichotomous data, and 95% selleck screening library confidence intervals (CI) were counted [21]. P<0.05 was required for the overall RR to be statistically Ferrostatin-1 significant. The between-study heterogeneity was assessed using I2 and χ2 measures. The pooled statistical analysis was calculated using the fixed effects model, but a random-effect
model was performed when the P value of heterogeneity test was <0.1. The data on the predictive ability of Cdx2 overexpression for 5-year survival rate were combined across studies using fixed and random effect models for the synthesis of hazard ratio (HR). The HR of 5-year survival rate was calculated from the reported data directly by number of events within 5 years after surgery was used, or data reading from Kaplan-Meier survival curve. The funnel plot was examined to explore the possibility of publication bias [21–23]. Kaplan-Meier curves were read by Engauge Digitizer version 2.11 (free software downloaded from http://sourceforge.net). Casein kinase 1 The data analysis
was performed using the meta-analysis software Review Manager (RevMan) v5.0.17 (Copenhagen: The Nordic Cochrane Centre, The Cochrane Collaboration, 2008; http://cc-ims.net/revman/download). Results Eligible studies As shown in Figure 1, our initial search yielded 412 studies. According to the inclusion and exclusion criteria, 13 papers [9, 11, 13–16, 24–30] were recruited into our meta-analysis. Only four studies reported the association between the Cdx2 and 5-year survival rate [9, 15, 16, 26]. Studies were carried out in Japan, China, Korea, Turkey and Germany. Table 1 presents the study characteristics for the included trials. Figure 1 Flow chart for our meta-analysis. Table 1 Study characteristics for the included studies Autor (year-country) Total number of patients Median age (range) Male: Female Adequacy of antibody methods Blinding of Cdx2 evaluation Cdx2 positive Cdx2 negative Cdx2 positive Cdx2 negative Ge [34] 59 107 52.2 37:22 51:56 Yes Yes (2008-china) (32–72) Okayama [14] 55 80 63.4 46:9 45:35 Yes Yes (2009-Japan) (31–87) Kim [5] 150 109 57.8 114:36 61:48 Yes Yes (2006- Korea) Roessler [15] 109 81 61.