Prognostic value of serum lactate kinetics in critically ill patients with cirrhosis and acute-on-chronic liver failure: a multicenter study

Lactate clearance (Δ24Lac) was reported to be inversely associated with mortality in critically ill patients. The aim of our study was to assess the value of Δ24Lac for the prognosis of critically ill patients with cirrhosis and acute-on-chronic liver failure (ACLF). We analysed 954 cirrhotic patients with hyperlactatemia admitted to intensive care units (ICUs) in the United States and eastern China. The patients were followed up for at least 1 year. In the unadjusted model, we observed a 15% decrease in hospital mortality with each 10% increase in Δ24Lac. In the fully adjusted model, the relationship between the risk of death and Δ24Lac remained statistically significant (hospital mortality: odds ratio [OR] 0.84, 95% confidence interval [CI]: 0.78- 0.90, p < 0.001; 90-day mortality: hazard ratio [HR] 0.94, 95%CI 0.92- 0.97, p < 0.001; for Δ24Lac per 10% increase). Similar results were found in patients with ACLF. We developed a Δ24Lac-adjusted score (LiFe-Δ24Lac), which performed significantly better in the area under the receiver operating characteristic curves (AUROCs) than the original LiFe score for predicting mortality. Lactate clearance is an independent predictor of death, and the LiFe-Δ24Lac score is a practical tool for stratifying the risk of death.


INTRODUCTION
Lactate can be measured in critically ill patients to evaluate the severity of disease [1][2][3]. Patients are considered to have higher lactate levels (hyperlactatemia) at concentrations of more than 2 mmol/L. Hyperlactatemia occurs when lactate production exceeds clearance [4]. Tissue hypoxia and subsequent anaerobic metabolism are considered to be the main mechanisms of hyperlactatemia.
Increased lactate production and reduced lactate clearance are common and associated with high mortality in critically ill patients. Studies show that dynamic lactate measures in the intensive care unit (ICU) are better than static lactate measurements for predicting deaths [5][6][7]. AGING Liver cirrhosis is considered an irreversible end result of chronic liver diseases [9]. Hospital mortality of cirrhotic patients admitted to the ICU ranges from 34 to 86% [10]. The combination of decompensated cirrhosis, organ failure(s) and high mortality rate marks the diagnosis of acute-on-chronic liver failure (ACLF). Hyperlactatemia upon admission to the ICU was strongly associated with adverse outcomes in cirrhotic patients [11]. The liver exhibits a higher net lactate clearance than any other organ, accounting for up to 70% of lactate clearance [12]. Lactate kinetics in cirrhotic patients are significantly different from those in patients without hepatic impairment [13]. Fulminant liver dysfunction has been shown to impair lactate clearance [14].
Several scoring systems incorporating lactate levels have been established to evaluate the prognosis of patients with cirrhosis and ACLF [15][16][17][18]. The liver injury and failure evaluation (LiFe) score, which is calculated using arterial lactate, serves as a useful tool for predicting the mortality of critically ill patients with cirrhosis and ACLF [17,18].
A number of studies have demonstrated that lactate is a reliable prognostic marker in the intensive care setting to identify cirrhotic patients at high risk of death. However, few studies focused on the prognostic value of lactate clearance in critically ill patients with cirrhosis and ACLF. Our study aimed to evaluate the prognostic value of lactate clearance in critically ill cirrhotic patients with hyperlactatemia admitted to the ICU. Moreover, we aimed to evaluate whether incorporation of lactate clearance instead of lactate into the LiFe score for patients with cirrhosis and ACLF may help to improve the prognostic performance.
There were two validation cohorts (eICU cohort and WMU cohort) in our study. Patients who came from the eICU cohort had similar baseline characteristics to patients in the MIMIC cohort. In the WMU cohort, all of the participants were Chinese. The main cause of liver cirrhosis was viral hepatitis (59.1%). All of the scores were lower in the WMU cohort, compared to the other two cohorts. Moreover, mortalities in the WMU cohort were also lower than in the other cohorts. There were four patients who underwent liver transplantation within 28 days, and there were nine patients who underwent liver transplantation within 90 days after ICU admission. More information about the baseline characteristics of the patients in the three cohorts is listed in Table 1.
Patients were stratified by Δ24Lac quartiles in the derivation cohort. As presented in Table 2, Δ24Lac levels were positively associated with MAP levels, body temperature levels, 24-hour urine output, and mechanical ventilation support and were inversely associated with bilirubin levels, vasopressor use, and mortality. Figure 1 shows Kaplan-Meier curves illustrating the cumulative survival rates stratified by Δ24Lac quartiles.
We repeated the lactate measurements at day 3-7 and calculated the lactate clearance (ΔLac3-7). We demonstrated that ΔLac3-7 was also an independent and significant predictor for poor prognosis in critically ill patients with cirrhosis and ACLF (Supplementary Table 4).

Development and validation of lactate clearanceadjusted LiFe (LiFe-Δ24Lac)
Lactate clearance was identified as a significant predictor of mortality, independent of the LiFe score (Tables 3 and 4). We hypothesized that the addition of lactate clearance instead of lactate to the LiFe score may improve the performance of the score for predicting mortality. The LiFe-Δ24Lac score was   significantly higher than those obtained for original LiFe score (0.69, 0.67, 0.66 and 0.61, all p < 0.01]). In predicting hospital and 1-year mortality, the addition of Δ24Lac to the LiFe score increased net reclassification improvements (NRIs) by 12.9% (p<0.01) and 16.4% (p<0.01), respectively, and increased integrated discrimination improvements (IDIs) by 8.1% (p=0.03) and 6.6% (p=0.04), respectively.

AGING
In patients with ACLF, the LiFe-Δ24Lac score also improved the prediction of mortality, as shown by the significant increase in AUROCs (in-hospital mortality: 0.75; 28- We also compared LiFe-Δ24Lac score with CLIF-C-ACLF score. LiFe-Δ24Lac score performed with a higher discrimination than CLIF-C ACLF score, especially for predicting short-term mortality (hospital and 28-day mortality, p<0.05). The prognostic performance of these scoring systems in the derivation and validation cohort are illustrated in Figure 2. The calibration plots showed good agreement between the LiFe-Δ24Lac score prediction and actual observation in both the primary and validation cohorts (Supplementary Figure 1).

DISCUSSION
In our study cohort of critically ill patients with cirrhosis and ACLF, we demonstrated that lactate clearance was independently associated with mortality rates after correction for other confounders. In addition, we found that incorporation of lactate clearance instead of lactate into the LiFe score improves the performance of the score for predicting outcome.
Despite aggressive medical interventions, critically ill patients with cirrhosis and ACLF have poor outcomes. Consistent with the findings from our study, the mortality rates of cirrhotic patients with hyperlactatemia admitted to the ICU ranged from 13 to 40% during hospital and ranged from 35 to 56% after 1-year followup. Most patients had infection, sepsis, and even multiorgan failures at ICU admission. A recent multicenter study revealed that ICU and hospital mortality rates in critically ill patients with ACLF were 39.2% and 54.6%, respectively [15]. The hospital and 1-year mortality of patients with ACLF in our study were 49.1% and 63.9%. Therefore, quick and accurate assessment of the severity of disease, which can help with timely initiation of organ support to improve prognosis, is urgently needed.
In clinical practice, blood lactate level is a useful marker for predicting the outcomes of critically ill patients, such as patients with sepsis and shock [19]. In different studies, optimal cutoffs of single static arterial lactate measurements vary considerably. In this context, fluctuations of blood lactate levels have attracted the interest of clinicians and researchers in recent years.  Previous studies demonstrated that early changes in lactate levels could be a practical tool for risk assessment [8,[20][21][22]. They also found that higher lactate clearance is associated with lower mortality rates. A large multicenter randomized controlled trial showed that lactate clearance could guide therapeutic measures in septic patients [23]. However, few studies focused on the prognostic value of lactate clearance in critically ill patients with cirrhosis and ACLF.
Recently, Drolz and colleagues conducted a retrospective study that included 816 critically ill patients with cirrhosis from 3 university hospitals in Europe [16]. They reported for the first time that lactate and lactate clearance were independent predictors of outcome in critically ill patients with cirrhosis and ACLF. Overall, our finding of decreased risk for death with increased lactate clearance is consistent with their results. In our study, risk reductions for hospital mortality were 16% (95%CI 10%-22%) in patients with cirrhosis and 18% (95%CI 10%-25%) in patients with ACLF for each 10% increase in Δ24Lac. In predicting 90-day all-cause mortality, risk reductions were 6% (95%CI 3%-8%) in patients with cirrhosis and 7% (95%CI 4%-9%) in patients with ACLF for each 10% increase in Δ24Lac.
There are, however, several differences between the two studies. ACLF is a dynamic syndrome. Gustot et al. found that evaluating ACLF patients at 3-7 days after ICU admission provided a tool for predicting the prognosis [24]. Therefore, we repeated the lactate measurements at day 3-7 and calculated the lactate clearance to assure our prediction was right. Through multivariate regression, we demonstrated that higher lactate clearance in patients at day 3-7 was independently associated with lower mortality rates.
We evaluated for the first time the potential contribution of lactate clearance to current scoring systems in a large multinational cohort of critically ill patients with cirrhosis and ACLF. We developed and externally validated a new score containing lactate clearance, named LiFe-Δ24Lac, for critically ill patients with cirrhosis and ACLF. The laboratory-based LiFe-Δ24Lac score is simple and objective and can be calculated quickly at the bedside. The superiority of the new score for predicting mortality rates was supported by a number of tests, such as the AUROC, NRI, IDI, and calibration curves. To improve its feasibility, we divided the score into three levels of risk (relatively low risk, intermediate risk, and high risk). The patients stratified into different groups had significantly different prognoses.

AGING
Our study has several limitations. First, we excluded patients who stayed in the ICU for no more than 48 hours and patients without hyperlactatemia. Our findings only can be generalized to patients with hyperlactatemia who stayed in the ICU for more than 48 hours. Second, this is a retrospective study. Limited by its retrospective nature, 289 patients were excluded because of the absence of repeated lactate measurements, which may have increased the selection bias. However, we demonstrated that baseline characteristics of these patients were comparable to those with repeated lactate measurements (Supplementary Table 1). We also externally validated the results in the other two cohorts. These lost cases might have a limited effect on the evaluation of the independent effect of Δ24Lac on the prognosis of cirrhosis. Second, the hospital mortality of the patients in the WMU cohort was significantly lower than in the other two cohorts. This phenomenon could be explained by the fact that most Chinese critically ill patients wish to die at home based on Chinese cultural perspectives. A study conducted in China revealed that 26-44% of adult ICU patients were transported home to die [25]. Third, there was no information about liver transplantation within the 1-year follow-up in the MIMIC database, so we could not conduct competing risk model in this cohort. However, we utilized a competing risk approach in the WMU cohort considering liver transplantation as a competing risk.
In conclusion, lactate clearance is a good and independent predictor of death in critically ill patients with cirrhosis and ACLF. Incorporation of lactate clearance into the current scoring system improved the performance of the score for predicting outcome. LiFe-Δ24Lac is a quick and easy risk stratification score and may help with the timely identification of patients at high-risk of death. More large-scale prospective multicenter studies are needed to evaluate and verify the applicability of this approach.

Data source
We extracted the patients' dataset from the Medical Information Mart for Intensive Care (MIMIC)-III and the eICU Collaborative Research Database (eICU) [26,27].

Study design
We conducted an international multicenter retrospective cohort study. Adult cirrhotic patients with hyperlactatemia on admission were eligible for inclusion in this study. We excluded patients who stayed in the ICU for no more than 48 hours. The primary endpoint of the study was hospital mortality, and the secondary endpoint was determined considering the all-cause 28-day, 90-day and 1-year mortality rates. The diagnosis of cirrhosis was based on histopathology, ultrasonography or computed tomography findings, clinical evidence of liver dysfunction or portal hypertension. We diagnosed and graded ACLF according to the criteria established by the EASL-CLIF consortium [28]. Hyperlactatemia was defined by arterial lactate levels above 2.0 mmol/L.
In the derivation cohort, we included 429 cirrhotic patients (7,189 measurements) with hyperlactatemia from the MIMIC database. The flow chart is presented in Figure 3. Additionally, we analysed a validation AGING cohort of 303 patients collected from the eICU database. Another validation cohort (WMU cohort) of 222 cirrhotic patients collected from ICUs of the First Affiliated Hospital of Wenzhou Medical University (Zhejiang, China) between January 2012 and January 2018 was also analysed.
The MIMIC and eICU cohorts' mortality information came from the social security database in the United States. The mortality information of the WMU cohort was collected by review of medical records or by contacting the patients. The ethics committee of the First Affiliated Hospital of Wenzhou Medical University approved this study. All study procedures followed the principles of the Declaration of Helsinki. No informed consent was required because all the data were anonymized.

Lactate-clearance and scoring systems calculation
The main focus of the study was arterial lactate kinetics over the first 48 hours of admission. To obtain Δ24Lac, we recorded the maximum lactate levels of day 1 and day 2. We calculated Δ24Lac based on the formula: Δ24Lac (%) = (Lactate day1 max -Lactate day2 max)/ Lactate day1 max [8]. Time-based protocols of lactate concentration determination were not prespecified. This reflected the real-world scenario regarding the timing and frequency of lactate measurements. Although there was no timebased protocol regarding lactate measurements, the lactate values were measured on a relatively regular basis in enrolled patients (at least one time during each 6-h shift on admission day, and at least one time per day during the ICU stay).
We also repeated the measurement on day 3-7 to assure our prediction was right. We calculated the lactate clearance at day 3-7 according to the formula: ΔLac % = Lac -Lac / Lac .

Statistical analysis
We categorized lactate clearance as ≤ 0, >0-0.3, >0.3-0.55, and >0.55. Continuous variables were expressed as median with interquartile range (IQR) and compared by Kruskal-Wallis test. Categorical variables were expressed as percentages (%) and compared by Chi-Square test. To maximize data availability, we used multiple imputations, based on 5 replications and a chained equation approach method in the R MI procedure, to account for missing albumin, INR, bilirubin, creatine, and urine output date. We also performed sensitivity analyses using a complete-case analysis. Univariate and multivariate logistic regressions were used to identify the association between lactate clearance and hospital mortality. Odds ratios (OR) were reported with a 95% confidence interval (CI). The association between 28-day and 90day mortality and lactate clearance was analysed by univariable and multivariable Cox proportional hazards models, and hazard ratios (HR) were calculated. We performed tests for linear trend by entering the median value of the variable. We conducted a multivariate competing risk regression (Fine-Gray model) in the WMU cohort to assess the effect of lactate clearance on mortality. Liver transplantation was taken into account as a competing risk.
We plotted cumulative survival curves by the Kaplan-Meier method and compared the survival curves using the log-rank test. The performances of the scoring systems were assessed by calculation of the area under the receiver-operating characteristic curve (AUROC) and assessed using the DeLong test [33]. Furthermore, we used the net reclassification index (NRI) and integrated discrimination improvement (IDI) to evaluate the additive predictive value of lactate clearance instead of lactate over the LiFe score in assessing the improvement of prognostic value [34]. A calibration curve was used to compare the predicted probability of survival versus actual, using 500 bootstraps resamples to reduce overfit bias. All of the tests were two sided, and a p value of <0.05 was considered statistically significant.