Lactate-albumin ratio as a predictor of mortality in patients with acute on chronic liver failure in a third-level care hospital in Mexico

Introduction and Objectives

Acute-on-chronic liver failure (ACLF) is an abrupt worsening of clinical conditions in patients with chronic liver disease. It has a higher mortality rate with respect to patients who do not develop this entity (33.9% vs. 4.7%). The lactate/albumin ratio is a statistically significant predictor (p<0.001) of mortality during hospitalization in these patients. This study aimed to determine whether the lactate-albumin ratio predicts mortality in patients with ACLF in Mexican population, identify the sociodemographic characteristics of this group of patients and to determine the related mortality at 7, 28, 90 and 180 days.

Materials and Patients

An observational, retrospective, single-center study was conducted where patients with diagnosis of ACLF according to the EASL-CLIF criteria who were hospitalized during the period from 2017 to 2022 in the Gastroenterology department at National Medical Centre "Siglo XXI" were included. Patients diagnosed with terminal chronic extrahepatic diseases, hepatocellular carcinoma and extrahepatic neoplasms were excluded.

Results

A total of 186 patients were enrolled, 51% were women, with an age range of 56-65 years, 29% were secondary to fatty liver disease associated with metabolic dysfunction, obtaining that the most frequent precipitant was the infectious origin in 111 patients (60%), with abdominal origin being the most prevalent (36%). Renal failure was present in 71%, followed by coagulopathy (50%) and neurological failure (49%). On admission, grade I ACLF was present in 37 patients (20%), grade II in 72 (39%), grade III in 77 (49%). At 7, 28, 28, 90 and 180 days 73 patients (39.5%), 146 patients (78.9%), 159 (85.9%) and 172 patients (93%) died respectively, with a lactate albumin ratio for each of these, with a cut-off point 1.24 (AUC 70.70%), 0.87(AUC 71.20%), 0.84 (AUC 73.5%) and 1.04(AUC 64.90%) respectively with statistically significant values p <0.05.

Conclusions

Lactate levels and its clearance have been shown to predict outcome of critically ill patients with liver cirrhosis, improving the prediction of mortality. The lactate albumin ratio is useful for predicting mortality in this group of patients at 7, 28, 90 and 180 days with adequate sensitivity and specificity. The values obtained were statistically significant as shown in the complementary tables.

Ethical statement

The protocol was registered and approved by the Ethics Committee. The identity of the patients is protected. Consentment was obtained.

Declaration of interests

None

Funding

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

Table 1. Lactate/albumin ratio.

Category		Mean Lactate/albumin ratio	Standard deviation	p
General population		1.74	1,542	<0.001*
ACLF grade at 1 day	i	1.31	1,055	0.026‡
	ii	1.54	1,214
	ii	2.15	1,891
ACLF grade at 3 days	i	1.40	1,146	0.048‡
	ii	1.48	1,108
	ii	2.20	1,942
Outcome at 7 days	Alive	1.28	0.961	< ¥0.001
	Death	2.46	1,954
Outcome at 28 days	Alive	1.08	0.901	< ¥0.001
	Death	1.93	1,633
Outcome at 90 days	Alive	0.93	0.696	< ¥0.001
	Death	1.88	1,605
Outcome at 180 days	Alive	1.11	0.86	¥0.072
	Death	1.80	1,576

*Kolmogorov-Smirnov test, ‡Kruskal-Wallis test, ¥Mann-Whitney U test.

Assessment	Area down the curve	Cutt off point	Sensitivity	Specificity	p
Lactate/albumin ratio mortality at 7 days	70.70%	1.24	64.40%	60.40%	<0.001
Lactate/albumin ratio mortality at 28 days	71.20%	0.87	71.00%	61.50%	<0.001
Lactate/albumin ratio mortality at 90 days	73.50%	0.84	72.20%	61.50%	<0.001
Lactate/albumin ratio mortality at 180 days	64.90%	1.04	60.20%	61.50%	0.034

Roc curves.