An increase in red blood cell distribution width from baseline predicts mortality in patients with severe sepsis or septic shock

Table 2 Cox proportional hazards analyses for 28-day and 90-day mortality according to baseline RDW value and ΔRDW _72hr-adm

Groups based on baseline RDW value and ΔRDW_72hr-adm	Unadjusted model		Adjusted model 1		Adjusted model 2
	HR (95% CI)	P value	HR (95% CI)	P value	HR (95% CI)	P value
28-day mortality
Group 1	1.0	-	1.0	-	1.0	-
Group 2	10.09 (2.24 to 45.52)	0.003	4.57 (0.93 to 22.47)	0.062	4.86 (0.93 to 25.29)	0.060
Group 3	3.64 (0.77 to 17.14)	0.102	1.31 (0.26 to 6.49)	0.743	1.72 (0.34 to 8.69)	0.513
Group 4	21.78 (4.85 to 96.95)	<0.001	7.85 (1.63 to 37.76)	0.010	9.97 (1.99 to 49.91)	0.005
90-day mortality
Group 1	1.0	-	1.0	-	1.0	-
Group 2	12.41 (2.80 to 54.98)	0.001	4.93 (1.06 to 22.94)	0.042	4.72 (1.00 to 22.27)	0.050
Group 3	7.44 (1.71 to 32.34)	0.007	3.85 (0.87 to 17.04)	0.076	4.78 (1.07 to 21.31)	0.040
Group 4	33.41 (7.71 to 144.70)	<0.001	11.66 (2.57 to 52.97)	0.001	13.74 (2.95 to 64.10)	0.001

Unadjusted Model: crude relative risk. Adjusted Model 1: adjusted for age, sex, SOFA score, Charlson Comorbidity Index, renal replacement therapy, albumin, hemoglobin, lactate, and C-reactive protein. Adjusted Model 2: Model 1 plus adjustment for infection site. RDW, red cell distribution width.

ISSN: 1364-8535