Molecular Vision: Fang, Mol Vis 2017; 23:306-317. Table 1

Table 1. Demographic data for corneal ulcer patients with suspected microbial keratitis.

	Bacterial  keratitis (n = 16)	Non-bacterial  keratitis (n = 45)	P-value^c
Sex, F:M	6:10	16:29	1.000
Age, mean ± S.D.	54.4 ± 14.6	54.7 ± 19.3	0.755
Laterality, OD:OS	11:5	29:16	1.000
Ocular risk factors
Wear contact lens, N (%)	3 (18.8%)	8 (18.8%)	1.000
Trauma, N (%)	5 (31.3%)	13 (28.9%)	1.000
Others^a, N (%)	4 (25.0%)	18 (40.0%)	0.370
Untraceable, N (%)	4 (25.0%)	6 (13.3%)	0.431
Systemic risk factors
Diabetes mellitus, N (%)	4 (25.0%)	7 (15.6%)	0.457
Cancer, N (%)	1 ( 6.3%)	0 ( 0.0%)	0.262
Untraceable, N (%)	11 (68.8%)	38 (84.4%)	0.270
Complications
Hypopyon, N (%)	7 (43.8%)	5 (11.1%)	0.028
Surgery^b, N (%)	2 (12.5%)	7 (15.6%)	1.000

^aThis item included swimming, farming work, neovascular glaucoma, pseudophakic bullous keratopathy, phthisis cornea, instillation of topical corticosteroids or anti-glaucoma agents. ^bSurgery was indicated by refractory to medical therapy and severe corneal melting to perforation. The surgical modalities included amniotic membrane transplantation, anterior lamellar keratoplasty, penetrating keratoplasty, and evisceration. ^cMann–Whitney U test was used to compare the continuous variable (age), while Fisher’s exact test was used to compare the categorical variables (i.e., variables other than age). P < 0.05 was statistically significant.