of T2D. An example search strategy for PubMed is provided as
follows (seeNote 3):
(“miRNA,” “diabetes,” and “expression” in the title/abstract)
or (“miRNA,” “diabetes,” and “profile” in the title/abstract) or
(“miRNA,” “diabetes,” and “profiling” in the title/abstract) or
(“microRNA,” “diabetes,” and “expression” in the title/abstract)
or (“microRNA,” “diabetes,” and “profile” in the title/abstract) or
(“microRNA,” “diabetes,” and “profiling” in the title/abstract).
3.2 Study Selection
(Screening, Eligibility,
and Inclusion)
After identification of relevant studies, study selection criteria are
used to include or exclude studies. Eligible studies have to meet
the inclusion criteria: (1) they are miRNA expression profiling stud-
ies on human patients or on animal models of T2D; (2) they com-
pare diabetic and nondiabetic samples; (3) they use miRNA
expression arrays; (4) they report cutoff criteria of differentially
expressed miRNAs; and (5) they report sample sizes. In our exclu-
sion criteria, miRNA profiling studies using saliva or urine samples
are excluded because we focus on the miRNAs in blood. However, if
we were studying miRNAs in oral cancer [39] and urinary tract
cancer [40–42], saliva and urine samples would be important. In
any case, a meta-analysis report should provide a flow diagram of
study selection, which includes identification, screening, eligibility
extraction, and inclusion steps. The PRISMA guideline [8] provides
a four-stage flow diagram, which is a template downloadable for use.
Figure4 shows our example study selection process according to our
selection criteria (seeNote 4).
Fig. 3A workflow for meta-analysis of dysregulated miRNAs
478 Hongmei Zhu and Siu-wai Leung