Skip to main content
. 2023 Jul 12;14(9):e00614. doi: 10.14309/ctg.0000000000000614

Table 5.

Univariable and multivariable Cox proportional hazards regression models for the association of spontaneous bacterial peritonitis development risk with PMN percent and count, among those with PMN < 250 cells/mm3 at first paracentesis

HR (95% CI), P
Multivariablea
Characteristic at paracentesis Univariable Adjusted for baseline characteristics Adjusted for PMN-C ≥ 115 cells/mm3 Adjusted for MELD-Nab
PMN-%
 <10% 1.00 1.00 1.00 1.00
 10%–29% 1.91 (1.13–3.22), 0.02 1.86 (1.10–3.14), 0.02 1.86 (1.10–3.14), 0.02 1.68 (0.97–2.91), 0.07
 ≥30% 4.11 (2.20–7.66), <0.001 3.83 (2.06–7.13), <0.001 3.86 (1.88–7.92), <0.001 3.48 (1.80–6.72), <0.001
Creatinine (per 1 mg/dL) 0.78 (0.57–1.06), 0.11 0.78 (0.57–1.07), 0.13 0.78 (0.57–1.07), 0.13
PMN-C ≥ 115 (vs <115 cells/mm3) 2.33 (1.11–4.86), 0.02 0.98 (0.41–2.36), 0.97
MELD-Na (per 1 unit) 1.03 (0.99–1.06), 0.10 1.02 (0.99–1.06), 0.18

CI, confidence interval; HR, hazard ratio; MELD-Na, Model for End-Stage Liver Disease-Sodium; PMN, polymorphonuclear leukocyte; PMN-%, PMN percentage; PMN-C, PMN count.

a

Multivariable models adjusted for serum creatinine at paracentesis; N = 377 due to missing serum creatinine values.

b

Adjusted for MELD-Na at paracentesis; N = 346 due to missing MELD-Na values.