Table 3. Multivariate Cox regression analyses of open-angle glaucoma according to fasting plasma glucose level in stratified subgroups based on age, body mass index, and presence of hypertension.
| Glucose (mg/dL) | N | OAG | IR (per 1,000) | MODEL 1 | P for trend | MODEL 2 | P for trend | ||
|---|---|---|---|---|---|---|---|---|---|
| Age | 40–64 years | < 80 | 26,366 | 96 | 0.887 | 1 (ref.) | < 0.001 | 1 (ref.) | < 0.001 |
| 80–99 | 176,653 | 906 | 1.265 | 1.359 (1.101, 1.677) | 1.373 (1.112, 1.695) | ||||
| 100–125 | 81,609 | 494 | 1.485 | 1.408 (1.130, 1.753) | 1.37 (1.098, 1.709) | ||||
| 126–139 | 7,877 | 71 | 2.212 | 1.839 (1.351, 2.502) | 1.492 (1.083, 2.054) | ||||
| 140–159 | 5,534 | 58 | 2.571 | 2.073 (1.494, 2.877) | 1.530 (1.081, 2.164) | ||||
| ≥ 160 | 8,501 | 116 | 3.430 | 2.928 (2.23, 3.843) | 2.022 (1.494, 2.736) | ||||
| ≥ 65 years | < 80 | 4,670 | 84 | 4.215 | 1 (ref.) | 0.009 | 1 (ref.) | 0.555 | |
| 80–99 | 33,055 | 582 | 4.100 | 0.979 (0.779, 1.231) | 0.988 (0.786, 1.243) | ||||
| 100–125 | 21,980 | 428 | 4.635 | 1.101 (0.871, 1.391) | 1.058 (0.836, 1.338) | ||||
| 126–139 | 3,190 | 64 | 4.934 | 1.165 (0.842, 1.613) | 0.987 (0.706, 1.38) | ||||
| 140–159 | 2,206 | 55 | 6.051 | 1.442 (1.026, 2.025) | 1.150 (0.805, 1.641) | ||||
| ≥ 160 | 2,735 | 62 | 5.615 | 1.338 (0.964, 1.858) | 1.047 (0.740, 1.482) | ||||
| Body mass index | < 25 kg/m2 | < 80 | 23,146 | 128 | 1.345 | 1 (ref.) | < 0.001 | 1 (ref.) | < 0.001 |
| 80–99 | 144,867 | 945 | 1.599 | 1.145 (0.952, 1.377) | 1.153 (0.958, 1.387) | ||||
| 100–125 | 59,771 | 562 | 2.291 | 1.382 (1.140, 1.675) | 1.333 (1.098, 1.618) | ||||
| 126–139 | 5,618 | 72 | 3.147 | 1.585 (1.186, 2.117) | 1.295 (0.959, 1.749) | ||||
| 140–159 | 3,806 | 55 | 3.512 | 1.770 (1.289, 2.430) | 1.339 (0.960, 1.868) | ||||
| ≥ 160 | 6,003 | 111 | 4.636 | 2.504 (1.940, 3.232) | 1.817 (1.372, 2.407) | ||||
| ≥ 25 kg/m2 | < 80 | 7,890 | 52 | 1.577 | 1 (ref.) | 0.001 | 1 (ref.) | 0.280 | |
| 80–99 | 64,841 | 543 | 2.031 | 1.271 (0.956, 1.690) | 1.294 (0.974, 1.721) | ||||
| 100–125 | 43,818 | 360 | 2.003 | 1.145 (0.856, 1.531) | 1.119 (0.835, 1.499) | ||||
| 126–139 | 5,449 | 63 | 2.840 | 1.438 (0.995, 2.077) | 1.216 (0.834, 1.773) | ||||
| 140–159 | 3,934 | 58 | 3.628 | 1.825 (1.255, 2.656) | 1.425 (0.963, 2.107) | ||||
| ≥ 160 | 5,233 | 67 | 3.203 | 1.747 (1.216, 2.510) | 1.316 (0.896, 1.931) | ||||
| Presence of Hypertension | Without hypertension | < 80 | 23,154 | 96 | 1.014 | 1 (ref.) | < 0.001 | 1 (ref.) | < 0.001 |
| 80–99 | 150,843 | 809 | 1.326 | 1.270 (1.028, 1.569) | 1.277 (1.033, 1.579) | ||||
| 100–125 | 58,913 | 378 | 1.578 | 1.364 (1.090, 1.708) | 1.346 (1.074, 1.688) | ||||
| 126–139 | 4,554 | 52 | 2.822 | 2.113 (1.506, 2.964) | 1.805 (1.269, 2.569) | ||||
| 140–159 | 3,074 | 36 | 2.864 | 2.111 (1.438, 3.099) | 1.674 (1.114, 2.515) | ||||
| ≥ 160 | 4,727 | 55 | 2.922 | 2.281 (1.637, 3.181) | 1.706 (1.180, 2.467) | ||||
| Hypertension without prior exposure to anti-hypertensive medication | < 80 | 2,231 | 20 | 2.147 | 1 (ref.) | 0.130 | 1 (ref.) | 0.152 | |
| 80–99 | 16,231 | 113 | 1.684 | 0.790 (0.491,1.272) | 0.798 (0.496, 1.285) | ||||
| 100–125 | 11,259 | 207 | 45.423 | 1.125 (0.697, 1.815) | 1.135 (0.702, 1.834) | ||||
| 126–139 | 1,197 | 9 | 1.876 | 0.842 (0.383, 1.850) | 0.822 (0.370, 1.825) | ||||
| 140–159 | 675 | 4 | 1.484 | 0.710 (0.242, 2.079) | 0.662 (0.221, 1.988) | ||||
| ≥ 160 | 1,214 | 13 | 2.696 | 1.285 (0.638, 2.587) | 1.202 (0.570, 2.535) | ||||
| Hypertension with exposure to antihypertensive medication | < 80 | 5,651 | 64 | 2.686 | 1 (ref.) | < 0.001 | 1 (ref.) | 0.057 | |
| 80–99 | 42,634 | 566 | 3.130 | 1.175 (0.907, 1.522) | 1.205 (0.931, 1.561) | ||||
| 100–125 | 33,417 | 437 | 3.120 | 1.138 (0.875, 1.48) | 1.098 (0.844, 1.429) | ||||
| 126–139 | 5,316 | 74 | 3.388 | 1.205 (0.862, 1.684) | 0.995 (0.707, 1.402) | ||||
| 140–159 | 3,991 | 73 | 4.455 | 1.604 (1.146, 2.243) | 1.241 (0.876, 1.758) | ||||
| ≥ 160 | 5,295 | 110 | 5.185 | 1.960 (1.440, 2.668) | 1.454 (1.049, 2.014) |
Data are incidence rates (IR, per 1,000 person-years) adjusted for age and sex (model 1) and additionally adjusted for current smoking, drinking, and exercising habits, diastolic blood pressure, body mass index, total serum cholesterol level, and the presence of type 2 diabetes (model 2). Each fasting plasma glucose level indicates a comparison of the OAG risk between those with that FPG level or higher and those with a lower level.