Skip to main content
. 2022 Jun 7;14:79. doi: 10.1186/s13098-022-00849-2

Table 3.

Meta-analysis of the associations of TLR4Asp299Gly polymorphism with DMI risk

Genetic model Comparison Test of association Test of heterogeneity
OR (95% CI) P-value P I2 (%)
Dominant model G/G + A/G vs A/A 1.52 (1.10–2.09) 0.01 0.008 58.0
Allelic model G vs A 1.42 (1.02–1.96) 0.04 0.003 63.0
Recessive model G/G vs A/G + A/A 1.87 (0.78–4.46) 0.16 0.94 0.0
Additive model G/G vs A/A 1.94 (0.81–4.61) 0.14 0.94 0.0
Subgroups
Ethnicity
 Caucasians G/G + A/G vs A/A 1.69 (1.22–2.35) 0.002 0.03 54.0
G vs A 1.58 (1.10–2.21) 0.01 0.007 62.0
 Asians G/G + A/G vs A/A 0.85 (0.52–1.39) 0.52 0.46 0.0
G vs A 0.89 (0.56–1.39) 0.60 0.47 0.0
Difference of DMI types
 Diabetic retinopathy G/G + A/G vs A/A 1.81 (1.04–3.14) 0.03 0.004 74.0
G vs A 1.77 (1.05–2.98) 0.03 0.004 74.0
 Diabetic nephropathy G/G + A/G vs A/A 1.19 (0.81–1.74) 0.38 0.58 0.0
G vs A 0.91 (0.50–1.64) 0.75 0.24 31.0
 Diabetic neuropathy G/G + A/G vs A/A 1.42 (0.76–2.67) 0.27 0.19 41.0
G vs A 1.40 (0.80–2.44) 0.24 0.22 34.0

Bold values indicate P-value < 0.05, that is, the combined effect size is statistically significant