Skip to main content
. 2017 Mar 10;31(7):1051–1059. doi: 10.1038/eye.2017.33

Table 3. Logistic regression analysis with the dependent variable being the development of parafoveal scotoma in early myopic normal tension glaucoma.

Variables Univariate
Multivariatea
  Odds ratio (95% CI) P-value Odds ratio (95% CI) P-value
Vertical CDR 16.976 (0.624–461.532) 0.093    
Tilt ratio 27.268 (2.913–255.279) 0.004 16.060 (1.022–252.287) 0.047
Significant optic disc rotation 0.261 (0.112–0.610) 0.002 0.259 (0.091–0.734) 0.011
Inferior rotation 3.607 (1.520–8.560) 0.004    
β-zone PPA 3.243 (1.168–9.005) 0.024    
Average mGCIPL thickness,μm 0.928 (0.879–0.980) 0.007    
Minimum mGCIPL thickness, μm 0.912 (0.867–0.959) <0.001 0.920 (0.863–0.981) 0.010
Baseline MD, dB 0.745 (0.581–0.955) 0.020 0.699 (0.504–0.969) 0.032
VF defect on superior hemifield 4.687 (1.707–12.874) 0.003 8.518 (2.309–31.421) 0.001

Abbreviations: CI, confidence interval; CDR, cup-to-disc ratio; MD, mean deviation; mGCIPL, macular ganglion cell-inner plexiform layer; PPA, parapapillary atrophy; VF, visual field. Variables with P>0.10 in the univariate model are not shown in this table. Variables with P<0.10 in the univariate model were entered in a multivariate model.

a

Variables were entered in the model if P<0.05 and removed if P>0.10 in the saturated multivariate model.