Platelet-to-Monocyte Ratio as a Novel Promising Agent for the Prognosis of Hepatitis B Virus-Associated Decompensated Cirrhosis

Aim The present study aimed at investigating associations of the platelet-to-monocyte ratio (PMR), a novel hematological indicator of inflammatory responses with 30-day outcomes in patients with HBV-associated decompensated cirrhosis (HBV-DeCi). Methods We recruited 329 patients with HBV-DeCi for this retrospective study and extracted baseline clinical data and laboratory characteristics from medical records. Univariate and multivariate analyses were performed to determine major factors influencing 30-day mortality. Receiver operating characteristic curve analysis was performed to compare the predictive values of prognostic markers. Results During the 30-day follow-up period, 21 (6.4%) patients died. The PMR was significantly different between nonsurvivors and survivors. Lower PMR was found to be associated with an increased risk of 30-day mortality, and PMR (odds ratio: 1.011; 95% CI: 1.003–1.019; P=0.005) was found to be an independent predictor of 30-day mortality in patients with HBV-DeCi with a significant predictive value (AUC = 0.826, 95% CI: 0.781–0.865). The combination of PMR and MELD score could improve prognostic accuracy in these patients (AUC = 0.911, 95% CI: 0.876–0.940). Conclusions Our results demonstrate that low PMR may be an independent predictor of 30-day mortality in patients with HBV-DeCi, and combined with the MELD score, it may be useful to complement other conventional measures to enable effective management of these patients.


Introduction
Hepatitis B virus (HBV) infection remains a major problem that endangers national health seriously in China. Many patients develop liver cirrhosis without receiving efective antiviral therapy [1]. Liver cirrhosis is a chronic, progressive, difuse liver fbrosis with high morbidity and mortality. It is the 11 th most common cause of death in the world, causing approximately 1.03 million deaths each year [2,3]. In China, nearly 3% of patients with compensated liver cirrhosis gradually progress to the decompensated stage characterized by overt clinical signs, such as ascites, rupture of esophageal and gastric varices, hepatic encephalopathy, and hepatorenal syndrome, which can lead to progressive multisystem organ failure. Te survival rate of patients with HBV-related decompensated liver cirrhosis (HBV-DeCi) is not optimistic. Te median survival is about 2 years in decompensated cirrhosis patients compared to more than 12 years in compensated cirrhosis patients, and the 5-year mortality rate is as high as 85% [4,5]. Patients need to be hospitalized for multiple times and sufer heavy medical expenses. Liver transplantation is the most efective treatment method currently, which can signifcantly improve the survival rate. However, the shortage of liver sources, cost, and technical defciencies limit its wide clinical application. Terefore, in order to accurately assess the severity of the patient's condition, it is urgent to fnd simple, objective, and efective biomarkers for the prognosis and disease monitoring of HBV-DeCi to improve the clinical management and survival rate of patients and reduce the economic burden.
Infection and increased systemic infammation cause organ dysfunction and death in patients with HBV-DeCi. Previous studies provide support for the diagnostic and prognostic roles of a series of serological indicators in HBVassociated liver diseases, such as C-reactive protein (CRP) [6], CRP to albumin ratio (CAR) [7,8], albumin-bilirubin score (ALBI) [9], aspartate aminotransferase to platelet ratio index (APRI) [10], fbrosis index based on four factors (FIB-4) [11], gamma-glutamyl transpeptidase-to-platelet ratio (GPR) [12], and gamma-glutamyl transpeptidase-toalbumin ratio (GAR) [13]. Monocytes, neutrophils, and lymphocytes are innate immune cells that can detect tissue damage, invade microorganisms, coordinate tissue healing, and eliminate infections. Infammation can cause changes in the level of these blood cells. Terefore, the combination of these infammatory indicators may be potential prognostic factors for predicting patients with decompensated liver cirrhosis.
In recent years, there is increasing evidence that neutrophil-to-lymphocyte ratio (NLR), lymphocyte-tomonocyte ratio (LMR), and platelet-to-lymphocyte ratio (PLR) are reliable infammatory markers and prognostic indices in assessing the severity and mortality of various diseases such as cancers [14,15], liver cirrhosis [16,17], and myocardial infarction [18]. Due to their low cost and easy to obtain and interpret nature, these hematological ratios have been widely used in the laboratory testing. However, there are few related literature studies exploring the clinical signifcance of platelet-to-monocyte ratio (PMR) in HBV-DeCi. Terefore, the present study aimed at determining the role of PMR in predicting the 30-day mortality of HBV-DeCi patients to provide help for the future clinical management and prognosis.

Patients.
Tree hundred twenty-nine patients with HBV-DeCi who underwent treatment in the First Afliated Hospital of Nanjing Medical University between January 5, 2017, and June 5, 2020, were retrospectively recruited. Te inclusion criteria included an age of at least 18 years. Te defnition of HBV-related decompensated cirrhosis diagnosed standard: (1) positive for hepatitis B surface antigen (HBsAg) ≥ 6 months; (2) liver histology or ultrasonography and other imaging methods suggested cirrhosis; and (3) current or past ascites, rupture of esophageal and gastric varices, and hepatic encephalopathy complications [19]. Patients who met the following criteria were excluded: (1) malignancy, (2) hematological diseases, (3) history of other chronic liver disease (e.g., infection with hepatitis C virus and autoimmune hepatitis), (4) undergoing platelet transfusion or immunosuppressive therapy in the 3 months before the study period, and (5) missing data. Te study and all its protocols were approved by the Institutional Ethics Committee of the First Afliated Hospital of Nanjing Medical University.

Data Collection and Follow-Up.
Relevant demographic and clinical data of the patients were obtained from medical records. Total protein, albumin, alanine aminotransferase (ALT), aspartate aminotransferase (AST), total bilirubin, creatinine, and blood urea nitrogen were measured in a Beckman Coulter AU5800 analyzer (Beckman Coulter, United States). Routine blood tests including hemoglobin levels, white blood cell (WBC), monocyte, and platelet (PLT) count were measured in a Sysmex XN series automated hematology analyzer (Sysmex, Japan). A Sysmex CS5100 automated blood coagulation analyzer (Sysmex, Japan) was used to determine the coagulation indices that included the international normalized ratio (INR). Te PMR was defned as PLT divided by monocyte. Severity of liver disease was evaluated at the time of admission by the Model for End-Stage Liver Disease (MELD) score as previously described [20]. Patients were followed for 30 days to evaluate survival. Data on mortality were obtained from medical records or by telephone.

Statistical Analysis.
All quantitative data are presented as the mean and standard deviation or median and interquartile range. Qualitative data are reported as count. Differences between quantitative variables were analyzed by the independent sample t test or Mann-Whitney U tests. Te chi-squared test was used for qualitative variables. Te correlation between the PMR and the MELD score was evaluated using Spearman's correlation test. In order to determine the risk factors of death in HBV-DeCi patients, univariate regression analysis was used. Ten, covariates with P < 0.1 were included as candidate variables in the multivariate stepwise logistic regression analysis to identify independent predictors for the prognosis of HBV-DeCi patients. Te receiver operating characteristic (ROC) curve was performed, and the area under the curve (AUC) was calculated to assess the prognostic value. DeLong's test was performed to compare the AUCs. Statistical analyses were performed using IBM SPSS 21.0 (SPSS Inc., Chicago, IL, USA) and MedCalc version 16.8.4 (MedCalc, Ostend, Belgium). Diferences with bilateral P < 0.05 were considered statistically signifcant.

Patient Characteristics.
After the exclusion criteria were applied, a total of 329 patients hospitalized with HBV-DeCi from January 2017 to June 2020 were enrolled in this retrospective study. Among the patients, 245 (74.5%) were male, and the median age was 51 years (range: 25-86 years). A signifcant negative correlation was found between the PMR and the MELD score (r � −0.384, P < 0.001) (Figure 1).
During the follow-up period, 308 patients survived and 21 died, giving a 30-day mortality rate of 6.4%. Te demographic and laboratory parameters are compared between nonsurvivors and survivors in Table 1. Te PMR was observed to be signifcantly lower in the nonsurvivors than that in survivors (median 82.43, IQR 45.66-112.01 vs. 194.04, 127.27-311.61, P < 0.001). Furthermore, compared to the survivors, the nonsurvivors also had a lower albumin level (P � 0.02) and a much higher WBC, monocyte, neutrophil, total bilirubin level, MELD score, and INR (all P < 0.001). Te other laboratory characteristics (i.e., gender ratio, median age, lymphocyte, hemoglobin, PLT, ALT, AST, and serum creatinine levels) were not signifcantly diferent between the two groups.

Independent Predictors of 30-Day Mortality in Patients
with HBV-DeCi. Following the univariate analysis, variables less than 0.1 were entered into multivariate logistic regression. As shown in Table 2, the multivariate analysis revealed the MELD score and PMR to be independent predictors of poor outcomes after adjustments (P < 0.001, P � 0.005). Te ROC curves of the MELD score and PMR to predict mortality are illustrated in Figure 2. Te cutof value of the MELD score was found to be 12.21, which had a sensitivity of 95.24% and specifcity of 70.45%. For PMR, the cutof value was 118.62 with a sensitivity of 80.95% and specifcity of 77.27%. Te predictive powers of the MELD score and PMR for mortality were not signifcantly diferent, as indicated by the similar AUC values (0.874 for the MELD score vs. 0.826 for PMR; Z � 0.949, P � 0.343). When the MELD score and PMR were analyzed in combination, the AUC (0.911) was slightly higher than that of the MELD score (Z � 1.741, P � 0.082) and PMR (Z � 2.641, P � 0.008), as were the specifcity (95.24%) and the sensitivity (78.57%). Te formula we used to combine the PMR and the MELD score was Logit P � 0.145 × MELD score − 0.011 × PMR − 3.360. Te ROC curves and comparison of prognostic scores are shown in Table 3, respectively.

Discussion
Tis study frstly identifed that the PMR was an independent predictor for 30-day mortality in HBV-DeCi patients, and the combination of the PMR and the MELD score could improve prognostic accuracy in these patients. Furthermore, the results also displayed that there was a signifcant inverse correlation between these two indicators.
Currently, the MELD score has been an ideal scoring system for assessing the severity of end-stage liver disease and widely used for the evaluation of organ allocation in liver transplant patients [21]. Compared with the Child-Pugh scoring system, the MELD model can assess the severity more objectively and accurately due to containing no subjective indicators such as ascites and hepatic encephalopathy, which may vary according to the physicians' judgement. However, the MELD score does not consider all factors that may afect the prognosis, such as infammation. Te creatinine level in the model is easily infuenced by hemodynamics and diuretics. Besides, many factors such as starvation and systemic infections can afect serum bilirubin. Accounting for these limitations, approximately 15%-20% of liver transplantation candidates are not well served by the MELD score [22]. In this study, we fnd that nonsurvivors have lower PMR levels than survivors. In addition, PMR is independently related to the adverse outcome of HBV-DeCi patients and has similar predictive power to the MELD score. Te unique advantages of the laboratory-based PMR are its objectivity, time-saving nature, and noninvasive interpretation, which are much easier to obtain and cheaper than the MELD score. Of note, the combination of PMR and MELD score further improves the prognostic accuracy of adverse outcomes compared with PMR or MELD alone. Our fndings demonstrate that PMR can be used as an independent predictor of 30-day mortality in patients with HBV-DeCi, and when it is combined with the MELD score, it may perform better to enable efective management of these patients.
Te underlying mechanism between PMR and prognosis in patients with HBV-DeCi needs to be illustrated. Our results indicate that the decrease of the PMR level in nonsurvivors is mainly due to the elevated number of monocytes and slightly increased number of PLT compared with surviving patients. It is generally known that infammation plays a key role in the progression of liver cirrhosis and is linked to adverse outcomes [23,24]. Monocytes can diferentiate into diferent cell groups and serve an important function in the innate and acquired immune response. Te infammatory response can trigger the release of monocytes from bone marrow into the peripheral blood [25]. Accumulated evidence shows that patients with liver cirrhosis have mononucleosis, which is positively correlated with disease progression [26]. Terefore, the markedly increase in monocytes that we observed among nonsurvivors may indicate persistent infammation, which lead to the poor prognosis. However, the relationship between changes in the number, phenotype, and function of monocytes and their subpopulations and disease progression during pathogen infection, especially in severe chronic HBV infection, remains unclear.
Among the 329 patients in the present study, 246 (74.8%) have thrombocytopenia (PLT < 100 × 10 9 /L). Trombocytopenia, common in liver cirrhosis, is a multifactorial condition involving both impaired thrombopoiesis and increased platelet clearance [27,28]. Te pathophysiology of thrombocytopenia in patients with cirrhosis has long been associated with the hypothesis of hypersplenism, in which splenic congestion and portal hypertension cause the mononuclearmacrophage system to phagocytose platelets, resulting Canadian Journal of Gastroenterology and Hepatology in thrombocytopenia and functional defects [29]. Te discovery of thrombopoietin (TPO) illustrates another central mechanism. TPO is mainly produced by the liver and is decreased when hepatocytes are seriously damaged. Tis can lead to decreased platelet production in the bone marrow, resulting in thrombocytopenia in patients with cirrhosis [30]. In addition, immune complexes are produced that destroy bone marrow megakaryocytes, leading to a block in platelet production. According to the already performed surveys, PLT is reported to be signifcantly associated with survival in patients with liver cirrhosis and plays a vital role in coordinating infammatory and immune responses [31]. However, according to the multivariate analysis in our study, it was found that neither monocytes nor PLT were independent predictors of mortality. Tis diference may have arisen because PLT and monocytes are individual parameters, which can be altered by several variables such as dehydration, overhydration, or blood specimen handling. In contrast, the PMR is much more stable as a ratio. It can be quickly calculated at the patient's bedside using a simple formula and easy to be applied for daily clinical practice. Furthermore, we also found a signifcant association between the low PMR and the high MELD score as well as 30-day mortality, suggesting that the lower PMR may be predictive of the severity and progression of liver injury and afect the short-term prognosis of patients with HBV-DeCi, while the exact mechanism of PMR to explain its association with prognosis and correlation with the MELD score remains elusive.
A few limitations in our study need to be considered. First, our single-center study is designed retrospectively in nature, which may have led to a selection bias. Te fndings need to be verifed by prospective and multicenter studies with larger sample numbers. Second, only 21 people died in the entire cohort, which made the study underpowered. Tird, we fail to give virological data and infammatory biomarkers, such as viral loads or CRP, in our study, which might be useful in elucidating the association between PMR and adverse outcome. We focused on DeCi mainly caused by HBV infection. Whether these fndings are applicable to other etiologies of DeCi needs further study.

Conclusions
In summary, this study concludes that the PMR can be utilized as an independent predictor for 30-day mortality in HBV-DeCi patients. A combination of PMR and MELD score further augmented the predicting power. Due to its time-saving and noninvasive nature, the PMR may be a useful supplement to standard approaches to enable effective management of these HBV-DeCi patients.

Data Availability
Te datasets are available from the corresponding author upon reasonable request.

Ethical Approval
Te study was performed to conform with the Declaration of Helsinki and was approved by the local ethics committee of the hospital.