Table 3.
The association of TMAO with DR in T2DM patients
| Quartiles of plasma TMAO concentrations, µmol/L | P-trend a | Per-SD of ln (TMAO) | P b | ||||
|---|---|---|---|---|---|---|---|
| Q1 (< 0.346) | Q2 (0.346–0.610) | Q3 (0.610–1.0) | Q4 (> 1.0) | ||||
| N (DR/DWR) | 14/16 | 12/19 | 22/9 | 24/6 | 72/50 | ||
| OR (95% CI) | |||||||
| Crude | 1 (reference) | 0.72 (0.26–1.99) | 2.79 (0.97–8.03) | 4.57 (1.45–14.40) | 0.001 | 1.99 (1.29–3.07) | 0.002 |
| Model 1 | 1 (reference) | 0.87 (0.29–2.67) | 5.21 (1.52–17.87) | 9.62 (2.54–36.48) | < 0.001 | 2.64 (1.54–4.53) | < 0.001 |
| Model 2 | 1 (reference) | 0.72 (0.22–2.37) | 5.06 (1.39–18.46) | 8.76 (2.21–34.76) | < 0.001 | 2.62 (1.51–4.55) | 0.001 |
| Model 3 | 1 (reference) | 0.73 (0.22–2.41) | 4.60 (1.24–17.17) | 7.42 (1.72–32.00) | 0.002 | 2.31 (1.30–4.12) | 0.005 |
TMAO, trimethylamine-N-oxide; T2DM, type 2 diabetes mellitus; DR, diabetic retinopathy; DWR, diabetes without diabetic retinopathy; BMI, body mass index; Q, quartile; SD, standard deviation; ln, logarithmic; OR, odds ratio; CI, confidence interval
aP values for linear trend across quartiles of TMAO were calculated by treating the median of each quartile as a continuous variable in binary logistic regression
bP values were obtained by binary logistic regression analyses. Model 1 was adjusted for age and gender. Model 2 was adjusted for age, gender, BMI, diabetes duration and hypertension. Model 3 was adjusted for covariates included in Model 2 plus estimated glomerular filtration rate