Skip to main content
. 2017 Feb 25;8(15):25261–25269. doi: 10.18632/oncotarget.15705

Table 1. Relationships between IDO markers as well as other variables and new bacterial infection as determined by multivariate Cox regression analysis.

Model 1 including IDO MFI
Variables Hazard ratio (95% CI) P
IDO MFI (Quartile 4 vs. Quartile 1) 9.257 (2.723-31.475) <0.001
Age (for every 10-year increase) 1.017 (0.807-1.281) 0.888
GNRI (for every 10-point increase) 1.284 (0.773-2.132) 0.335
Months on dialysis (for every 1-month increase) 1.001 (1.000-1.002) 0.045
Model 2 including the Kyn/Trp ratio
Variables Hazard ratio (95% CI) P
Kyn/Trp ratio (Quartile 4 vs. Quartile 1) 0.634 (0.228-1.761) 0.382
Age (for every 10-year increase) 1.075 (0.842-1.371) 0.563
GNRI (for every 10-point increase) 1.056 (0.637-1.749) 0.834
Months on dialysis (for every 1-month increase) 1.001 (1.000-1.002) 0.059

Multivariate Cox regression analysis was performed by adjusting for 4 different variables. Models 1 and 2 were applied to investigate the quartiles of the two IDO markers, the IDO MFI and the Kyn/Trp ratio. In Model 1, high IDO MFI quartile (Quartile 4 vs. Quartile 1) was correlated with new bacterial infection, whereas in Model 2, high Kyn/Trp ratio quartile was not correlated with new bacterial infection. IDO: indoleamine 2,3-dioxygenase; MFI: median fluorescence index; GNRI: Geriatric Nutritional Risk Index.