A case–control study of pre-operative levels of serum neutrophil gelatinase-associated lipocalin and other potential inflammatory markers in colorectal cancer

Table 3 Multivariate associations between serum levels of NGAL, sTNFR-1, sTNFR-2 and risk of colorectal cancer

		Multivariate logistic model^a			Multivariate logistic model adjusted for CRP^b
Serum marker (ng/ml)	Number of case/controls	OR	95% CI	P value^c	OR	95% CI	P value^c
NGAL				<0.001			<0.001
< 75	40/82	1			1
75- 104	52/85	1.42	0.81-2.50		1.43	0.81-2.50
≥ 104	127/83	2.76	1.59-4.78		2.81	1.61-4.89
NGAL/MMP-9				0.75			0.74
< 18.5	60/83	1			1
18.5-42.3	61/84	0.95	0.56-1.62		0.95	0.56-1.62
≥ 42.3	98/84	1.16	0.65-2.06		1.17	0.66-2.09
sTNFR-1				0.015			0.013
< 1.30	38/83	1			1
1.30-1.77	72/84	2.00	1.14-3.50		2.03	1.16-3.55
≥ 1.77	109/83	2.44	1.34-4.45		2.51	1.36-4.61
sTNFR-2				0.016			0.015
< 2.23	47/83	1			1
2.23-3.09	63/85	am 1.34	0.79-2.28		am 1.34	0.79-2.29
≥ 3.09	109/82	1.93	1.12-3.31		1.95	1.13-3.37

^aModels were stratified on centre and adjusted for age, gender, body mass index, waist/hip ratio, family history of colorectal cancer, alcohol intake, pre-albumin, recent weight-loss and regular statin intake in the last month.
^bC reactive protein (CRP) was forced in the multivariate models (p values for association with the risk of colorectal cancer were respectively of 0.58, 0.87, 0.60, 0.69 in the NGAL, NGAL/MMP-9, sTNFR-1 and sTNFR-2 models).
^cp value for trend.

ISSN: 1471-2407