Abstract
OBJECTIVES
This research was applied to case-control studies of the association between pancreatitis and SPINK1 gene to assess the joint evidence for the association, the influence of individual studies, and evidence for publication bias.
METHODS
MEDLINE and Embase were searched to identify longitudinal studies evaluating pancreatitis and SPINK1. Odds ratios (ORs) and 95% confidence interval (CI) were pooled using random-effect models and calculated using Carlin method. Publication bias was assessed using Egger et al's approach (A famous statistic method by Egger et al). Sensitivity, heterogeneity, and trim and fill analyses were conducted.
RESULTS
Based on the results, we found that (1) the results support for the association between pancreatitis and SPINK1, when analyzed totally and by subdivision (total [OR, 7.771; 95% CI, 5.232-11.543; P < 0.000]; European [OR,6.400; 95% CI, 4.346-9.426; P < 0.000]; Asian [OR, 11.823; 95% CI, 4.612-30.310; P < 0.000]; American [OR, 3.777; 95% CI, 1.596-8.939; P = 0.002]; mixed: [OR, 13.566; 95% CI, 2.322-79.252, P = 0.004]); (2) no evidence indicates that this association is accounted for by any one study, and no evidence indicates any publication bias exists.
CONCLUSIONS
The results indicated that SPINK1 gene, particularly the N34S mutation, has a genetic association with the development of pancreatitis.
Collapse