LABORATORIO DE URGENCIAS

Procalcitonin in Patients with Cirrhosis Hospitalized for Acute Infection

                                                              Paolo Girardi, Raffaele Buono, Camilla Bisazza, Luisa Marchi, Paolo Angeli, Marco Di Pascoli
                                                                                  Department of  Informatics and Statistics, Ca’ Foscari, University of Venice, Italy
                                                                                                Unit of Internal Medicine and Hepatology - DIMED, University of Padova, Padova, Italy

                                                                                                                         Dig. LIver.Dis; 2024 May; 56 (5): 810-17

Patients with liver cirrhosis have an up to three-fold higher risk of developing infections than the general population. According to a recent hypothesis, systemic inflammation occurring during infec[1]tion can promote the transition from a compensated to a decom[1]pensated condition [1]. Moreover, in these patients, infections sig[1]nificantly increase the risk of short and long-term mortality [2]. Therefore, in patients with cirrhosis, an early diagnosis and prog[1]nostic evaluation during infection are crucial to reduce the risk of morbidity and mortality. Plasma concentrations of procalcitonin (PCT) in subjects with[1]out infection are very low, but in presence of infection, especially ✩ The authors whose names are listed above certify that they have NO affiliations with or involvement in any organization or entity with any financial interest in the subject matter or materials discussed in this manuscript of bacterial origin, its production significantly increases in several tissues [3,4].
 Although the specific biological functions of PCT have not been fully clarified, it seems to act as a modulator of inflam[1]matory responses [5]. PCT, besides being a good indicator of infec[1]tion, has been shown to be a useful marker for predicting mortal[1]ity and guiding antibiotic therapy in the general population with ongoing infections [6].
The liver plays a central role in the production of PCT during infection, therefore it could be expected that in liver dis[1]eases the synthesis and blood concentration of PCT are decreased. However, in patients with advanced liver disease, the basal levels of PCT are even increased, suggesting a complex relationship between liver and PCT levels. Studies conducted in patients with liver cirrhosis have led to conflicting results and the prog[1]nostic role of PCT in subjects with cirrhosis has not been clarified yet [8–10].
The aim of this study is to evaluate the prognostic role of PCT in patients with liver cirrhosis hospitalized for acute infection, and to compare it with other markers of infection.

Patients and methods
This is a retrospective study including patients hospitalized for acute infection in the Unit of Internal Medicine and Hepatology, Hospital of Padova, Italy, from 2018 to 2020.
Patients were divided into two groups based on the diagnosis of liver cirrhosis. Inclusion criteria were as follows: (a) age >18 years; (b) hospitalization due to an acute infection.
The diagnosis of infections was made according to international guidelines. The severity of the infection was stratified based on the presence or not of sep[1]sis, according to Sepsis-1 criteria (the American College of Chest Physicians and the Society of Critical Care Medicine), and on the qSOFA score. Based on the data available, it was not possible to calculate SOFA and CLIF score. Demographics, clinical, and laboratory data from patients in[1]cluded in the study were recorded. The laboratory data recorded were the first obtained within 48 h after the admission. In patients with cirrhosis, acute-on-chronic liver failure (ACLF) presence was assessed at admission and the grade was calculated according to the diagnostic criteria of the European Foundation for the study of chronic liver failure (EF-CLIF) [11]. ACLF grade 1 was characterized by the presence of kidney failure (serum creatinine ≥2 mg/dl) or by other single organ/system failure (liver: serum bilirubin ≥12 mg/dl; brain: grade III-IV hepatic encephalopathy based on West Haven criteria7; coagulation: international normalized ratio [INR] ≥2.5 or platelet count ≤20 × 109/L; circulation: treatment with vasoconstrictors to maintain arterial pressure or inotropes to improve cardiac output; lungs: PaO2/FiO2 ≤200 or SpO2/FiO2 ≤214), if associated with kidney dysfunction (serum creatinine ranging from 1.5 to 1.9 mg/dl) and/or with mild-to[1]moderate (grade I-II) HE. ACLF grade 2 and ACLF grade 3 were defined by the presence of 2 or ≥3 organ/system failures, respec[1]tively.
Patients were followed-up till the end of the hospitalization or death. The primary endpoints of the study were the duration of hospitalization and antibiotic therapy. The secondary endpoint was in-hospital mortality.
Data characteristics were summarized by frequency and percentage if categorical variables, and by the median and the in[1]terquartile range (IQR) due to the non-normal distribution of some continuous variables of interest. The comparison of the characteristics between two groups was performed using the Wilcoxon rank sum test and Pearson Chi-square test for qualitative and quantitative variables, respectively. A Kruskal-Wallis test was used if the comparison by quantitative variables involved more than two strata.
The database had some missing values of the lactate variable (83, 29.7%) and since there was a correlation with other covariates, we imputed the missing values using a MICE (Multiple Imputations by Chained Equation; Van Buuren and Groothuis-Oudshoorn, 2011) procedure based on 30 iterations using a CART (Classification and Regression Tree) algorithm.
The analyses to determine the prognostic value of lactate have been repeated also with only non-imputed data, confirming the results, just with slightly lower statistical significance, due to the reduction of the sample size. We used a binomial logistic regression model to evaluate which variables were associated with increased mortality. We considered the following starting variables: leukocyte, C-reactive protein (CRP), procalcitonin (PCT), lactate, diagnostic group, gender, age, presence of comorbidities (heart disease of different aetiology, di[1]abetes mellitus, chronic kidney disease (CKD), chronic obstructive pulmonary disease (COPD), and neoplasms), presence of sepsis, and identification of multidrug-resistant organism (MDRO). Results were reported by means of Odds Ratio (OR) and the relative 95% Interval Confidence (95% IC). The outcomes of the length of hospitalization and the duration of the antibiotic therapy were analyzed using the Cox regression model.

Discussion
Some of the markers of infection commonly used in the gen[1]eral population are not equally reliable diagnostic and prognostic indicators in patients with cirrhosis. In these patients, the use of leukocytes is limited because their levels are often reduced due to the common presence of hypersplenism. CRP is a protein pro[1]duced only in the liver and its production decreases proportionally with the reduction in liver function [12]. In recent years, the inter[1]est in PCT has gained prominence. A decrease in PCT levels during an infection is associated with improved clinical conditions, while increasing values are associated with worse outcomes, including mortality. This is not observed for other biomarkers: CRP shows stable or intermittent serum levels that do not correlate with the severity of the infection [13]. On the other hand, PCT prognostic value in patients with cirrhosis has not been clarified. In our study, both the length of hospitalization and the du[1]ration of antibiotic therapy were longer in the cirrhotic group; this could be a consequence of the condition of immunosuppres[1]sion that characterizes patients with cirrhosis, with a longer pe[1]riod of treatment needed for recovery [14]. Both durations were also longer in presence of infection by a multidrug-resistant or[1]ganism: this may be explained by the necessity of further antibi[1]otic treatment, following the first-line empirical therapy, after the identification of the microorganism and the definition of the an[1]tibiogram. In-hospital mortality was 10%, without differences be[1]tween patients with and without cirrhosis. As expected, in patients with cirrhosis, lower levels of leukocytes and CRP were observed. On the contrary, in these patients, the levels of lactate were higher than in patients without cirrhosis, probably as a consequence of a lower hepatic clearance [15]. PCT levels at admission were sim[1]ilar between the two groups, which is in line with a previous study [8]. This could be explained by the fact that hepatocytes maintain the ability to synthesize PCT even in case of cirrhosis and that other organs contribute to the production of PCT during infection.
All the infection markers positively correlate with increased mortality: for values included in the second and third tertile of leukocytes, CRP, PCT, and lactate, the estimated ORs were higher than one. However, because of the limited sample size to analyse this outcome, statistical significance was found only for high serum lactate levels.
Surprisingly, CKD was found to be a protective factor for mortality in patients admitted for infection. With regard to the duration of hospitalization and antibiotic therapy, a positive correlation was observed with high levels of PCT. High levels of lactate and leucocytes had a statistically sig[1]nificant correlation only with the hospitalization duration, prob[1]ably because the analysis on antibiotic therapy was restricted to the population with a confirmed bacterial infection, reducing the sample size. These data confirm the prognostic value of PCT, lac[1]tate and leucocytes in relation to the course of an infection. Subse[1]quently, we evaluated whether there was a difference in the prog[1]nostic value of the markers of infection between the two groups. From the descriptive analysis, only in the cirrhotic population, higher PCT values are associated with an increase in the duration of hospitalization and antibiotic therapy, and higher CRP values with an increased duration of antibiotic therapy. This result sug[1]gests even greater predictive value of PCT for the cirrhotic popula[1]tion. On the other hand, the ANOVA analysis did not show statisti[1]cal significance, indicating that PCT is a good prognostic indicator in both cirrhotic and non-cirrhotic populations. Overall, these data indicate that in the cirrhotic population PCT is a good prognos[1]tic indicator for the course of the infection, especially if compared to the other markers of inflammation, and this can be explained by the reasons explained above regarding the role of the differ[1]ent markers of infection in patients with cirrhosis. In addition, we showed that, in patients with cirrhosis, PCT was the only marker associated with the grade of ACLF secondary to infection. One of the limitations of our study is primarily its retrospec[1]tive design.
Furthermore, the decision to include patients present[1]ing non bacterial infection is questionable considering that during viral infections the production of PCT is generally inhibited by IFN[1]γ ; however, the percentage of non bacterial infections was very low and a selection of patients would not be suitable with clin[1]ical practice, since generally the aetiology of the infection is not known at admission, but only later during the hospital stay [16]. Moreover, in addition to the single value of the infection markers at admission, it would have been useful to observe their variations over time in relation to the outcomes, but the data were not ho[1]mogeneous and often missing. Finally, a problem of reverse causal[1]ity may be present, since the length of antibiotic therapy and hos[1]pitalization could have been influenced by the levels of PCT and other markers; on the other hand, more than their levels at admis[1]sion, their variation during hospitalization and other clinical out[1]comes are more often considered by clinicians to take decisions about duration of antibiotic therapy and hospitalization. In conclusion, our study indicates that in patients with cirrho[1]sis and acute infection, PCT is a good prognostic indicator for the course of hospitalization.
Studies with a larger number of patients are needed to confirm the potential role of PCT in cirrhosis during hospitalization because of infection.

Nota:results, tables, graphs in the original article cited at the beginning

References
[1] Foreman MG, Mannino DM, Moss M. Cirrhosis as a risk factor for sep[1]sis and death: analysis of the national hospital discharge survey. Chest 2003;124:1016–20.
[2] Arvaniti V, et al. Infections in patients with cirrhosis increase mortality four-fold and should be used in determining prognosis. Gastroenterology 2010;139:1246–56.
[3] Russwurm S, et al. Procalcitonin and CGRP-I mRNA expression in var[1]ious human tissues. Shock: Injury, Inflamm Sepsis: Lab Clin Approach 2001;16:109–12 2001.
[4] Jin M, Khan AI. P.rocalcitonin: uses in the clinical laboratory for the diagnosis of sepsis. Lab Med 2010;41:173–7.
[5] Meisner M. Pathobiochemistry and clinical use of procalcitonin. Clin Chim Acta 2002;323:17–29.
[6] Schuetz P, et al. Procalcitonin (PCT)-guided antibiotic stewardship: an inter[1]national experts consensus on optimized clinical use. Clin Chem Lab Med 2019;57:1308–18.
[7] Dong R, et al. Procalcitonin and liver disease: a literature review. J Clin Transl Hepatol 2019;7:51–5.
[8] Bota DP, van Nuffelen M, Zakariah AN, Vincent JL. S.erum levels of C-reactive protein and procalcitonin in critically ill patients with cirrhosis of the liver. J Lab Clin Med 2005;146:347–51. [9] Cekin Y, et al. The role of serum procalcitonin levels in predicting ascitic fluid infection in hospitalized cirrhotic and non-cirrhotic patients. Int J Med Sci 2013;10:1367–74.
[10] Lin KH, et al. Serum procalcitonin and C-reactive protein levels as markers of bacterial infection in patients with liver cirrhosis: a systematic review and meta-analysis. Diagn Microbiol Infect Dis 2014;80:72–8.
[11] Moreau R, Jalan R, Gines P, Pavesi M, Angeli P, Cordoba J, Durand F, Gus[1]tot T, Saliba F, Domenicali M, Gerbes A, Wendon J, Alessandria C, Laleman W, Zeuzem S, Trebicka J, Bernardi M, Arroyo VCANONIC Study Investigators of the EASL–CLIF Consortium. Acute-on-chronic liver failure is a distinct syndrome that develops in patients with acute decompensation of cirrhosis. Gastroen[1]terology 2013;144(7):1426–37 1437.e1-9.
[12] Pieri G, Agarwal B, Burroughs AK. C.-reactive protein and bacterial infection in cirrhosis. Ann Gastroenterol : Q Public Hellenic Soc Gastroenterol 2014;27:113.
[13] Gregoriano C, Heilmann E, Molitor A, Schuetz P. Role of procalcitonin use in the management of sepsis. J Thorac Dis 2020;2:S5–S15.
[14] Albillos A, Lario M, Álvarez-Mon M. Cirrhosis-associated immune dysfunction: distinctive features and clinical relevance. J Hepatol 2014;61:1385–96.
[15] Drolz A, et al. Lactate improves prediction of short-term mortality in critically Ill patients with cirrhosis: a multinational study. Hepatology 2019;69:258–69.
[16] Mayr FB, Yende S, Angus DC. Epidemiology of severe sepsis. Virulence 2014;5:4