Proteinuria can predict prognosis after liver transplantation

BMC Surgery

Table 7 Prediction of subsequent 90-day mortality

Predictive factors	Cutoff point	Youden index	Sensitivity (%)	Specificity (%)	Overall correctness (%)
Proteinuria
On admission	positive	0.16	28	88	58
Postoperative day 1	positive	0.22	64	58	61
Postoperative day 7	positive	0.51	85	66	76
Postoperative day 14	positive	0.55	90	65	77
Child-Pugh points
On admission	12	0.17	95	22	59
Postoperative day 1	10	0.26	90	37	64
Postoperative day 7	9	0.42	60	82	71
Postoperative day 14	8	0.42	50	92	71
MELD score
On admission	15	0.19	65	56	61
Postoperative day 1	22	0.27	60	71	66
Postoperative day 7	20	0.58	69	87	78
Postoperative day 14	23	0.62	75	87	81
RIFLE
On admission	I category	0.04	7	97	52
Postoperative day 1	I category	0.20	57	64	61
Postoperative day 7	I category	0.56	72	82	77
Postoperative day 14	R category	0.60	74	82	78
SOFA
On admission	5	0.23	52	71	62
Postoperative day 1	11	0.41	77	64	71
Postoperative day 7	7	0.70	83	87	84
Postoperative day 14	7	0.69	85	84	84
Proteinuria plus SOFA
On admission	5	0.26	57	70	63
Postoperative day 1	12	0.54	79	75	77
Postoperative day 7	8	0.72	85	85	85
Postoperative day 14	8	0.72	89	83	86
Proteinuria plus nonrenal SOFA
On admission	5	0.26	54	76	63
Postoperative day 1	12	0.54	54	100	77
Postoperative day 7	8	0.68	86	82	84
Postoperative day 14	7	0.67	100	67	84

Optimal cutoff points for predicting 3-month mortality were derived from receiver operator characteristic analysis. On admission (pre-transplant), post-transplant day 1, 7, and 14, the Youden index and overall correctness for predicting 3-month mortality were higher for the proteinuria plus SOFA score than those for the proteinuria, Child-Pugh points, MELD score, RIFLE criteria, SOFA, and proteinuria plus nonrenal SOFA scores
Abbreviation: MELD model for end-stage liver disease, SOFA sequential organ failure assessment

ISSN: 1471-2482