METHODS: Literature was searched in multiple databases including PubMed, Web of Science, EMBASE (Ovid SP), Airiti Library, Medline Complete, and ProQuest up to July 2015. Allelic frequency for TCF7L2 rs7903146 polymorphism in GDM and control subjects was extracted and statistical analysis was performed using Comprehensive Meta-Analysis (CMA) 2.0 statistical software. The association between TCF7L2 rs7903146 polymorphism and GDM risk was assessed by pooled odd ratios (ORs) using five gene models (dominant, recessive, homozygote, heterozygote, and allele). Stratified analysis based on race/ethnicity was also conducted. The between-study heterogeneity and contribution of each single study to the final result was tested by Cochran Q test and sensitivity analyses, respectively. Publication bias was evaluated using Egger's linear regression test.

RESULTS: A total of 16 studies involving 4,853 cases and 10,631 controls were included in this meta-analysis. Significant association between the T-allele of rs7903146 and GDM risk was observed under all genetic models, dominant model (OR = 1.44, 95% CI = 1.19-1.74), recessive model (OR = 1.35, 95% CI = 1.08-1.70), heterozygous model (OR = 1.31, 95% CI = 1.12-1.53), homozygous model (OR = 1.67, 95% CI = 1.31-2.12), and allele model (OR = 1.31, 95% CI = 1.12-1.53). Stratified analysis by race/ethnicity showed a statistically significant association between rs7903146 polymorphism and susceptibility to GDM under homozygous genetic model (TT versus CC) among whites, Hispanics/Latinos and Asians. Sensitivity analysis showed that the overall findings were robust to potentially influential decisions of the 16 studies included. No significant evidence for publication bias was observed in this meta-analysis for overall studies and subgroup studies.

METHODS: Computed tomography (CT) scans of 100 normal adult knees, aged 18 years and above, were analysed using a 3-dimensional (3D) analysis software. All tibiae were first aligned to a standard frame of reference and then rotationally aligned to the tibial centroid axis (TCAx) and the transmalleolar axis (tmAx). MPTA was measured from best-fit planes on the surface of the proximal tibia for each rotational alignment. Diaphyseal bowing was assessed by dividing the shaft to three equal portions and establishing the angle between the proximal and distal segments.

RESULTS: The mean MPTA was 87.0° ± 2.2° (mean ± SD) when rotationally aligned to TCAx and 91.6° ± 2.7° when aligned to tmAx. The mean diaphyseal bowing was 0.1° ± 1.9° varus when rotationally aligned to TCAx and 0.3° ± 1.6° valgus when aligned to tmAx. The mean difference when the MPTA was measured with two different rotational alignments (TCAx and tmAx) was 4.6° ± 2.3°. No statistically significant differences were observed between males and females. Post hoc tests revealed statistically significant difference in MPTA between different ethnic sub-groups.

METHODS: One hundred computed tomography scans of disease-free knees were analyzed. A 3-dimensional reconstructed image of the tibia was generated and aligned to its anatomic axis in the coronal and sagittal planes. The tibia was then rotationally aligned to the tibial plateau (tibial centroid axis) and PTS was measured from best-fit planes on the surface of the proximal tibia and individually for the medial and lateral plateaus. This was then repeated with the tibia rotationally aligned to the ankle (transmalleolar axis).

RESULTS: When rotationally aligned to the tibial plateau, the mean PTS, medial PTS, and lateral PTS were 11.2° ± 3.0 (range, 4.7°-17.7°), 11.3° ± 3.2 (range, 2.7°-19.7°), and 10.9° ± 3.7 (range, 3.5°-19.4°), respectively. When rotationally aligned to the ankle, the mean PTS, medial PTS, and lateral PTS were 11.4° ± 3.0 (range, 5.3°-19.3°), 13.9° ± 3.7 (range, 3.1°-24.4°), and 9.7° ± 3.6 (range, 0.8°-17.7°), respectively.