This article was submitted to Gastrointestinal and Hepatic Pharmacology, a section of the journal Frontiers in Pharmacology
This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.
Liver cirrhosis is a frequent end stage of liver disease, which itself results from a long-term process of fibrosis and sustained inflammation and leads to chronic liver disease (
Anemia is now identified as an important predictor of adverse outcomes in liver cirrhosis patients, such as the development of acute-on-chronic liver failure (ACLF) in outpatients with cirrhosis and hepatocellular carcinoma mortality rates (
Previous a study showed that the relationship between MCV and MELD (
This is a retrospective study from the Big Data Platform of the First affiliated hospital of Dalian Medical university from May 2011 to April 2018, our data consists of 1732 patients with decompensated HBV associated decompensated cirrhosis. Our research used the International Classification of Diseases codes to identify decompensation cirrhosis with HBV hospitalized patients. Decompensated Cirrhosis in Patients with hepatitis B according to the China’s Guidelines for the Prevention and Treatment of Chronic Hepatitis B (
Flow chart for the selection of patients.
The research protocol was reviewed and approved with a waiver of written informed consent by the Ethics Committee of the First affiliated hospital of Dalian Medical university, informed consent by telephone was obtained from each participant. All the methods were performed in accordance with relevant guidelines and regulations.
Demographic characteristics were obtained from face-to-face communication with patients or their families when the patient was admitted to our hospital. Blood samples were taken from the patients on an empty stomach for more than 10 h after the whole night and fast sent to the laboratory assessments. Having more than one cigarette per day is considered as smoking and alcohol intaking more than 20 g per day for at least a year is considered as drinking (
Using the following formula to calculate the MELD score: 9.57 × loge (creatinine mg/dl) + 3.78 × loge (bilirubin mg/dl) + 11.2 × loge (INR) + 6.43, where INR is the international normalized ratio and 6.43 is the constant of the etiology of liver disease (
Categorical variables were described in counts (percentages) and continuous variables as means ± standard deviation (SD). Patients were distributed into 3 groups by mean corpuscular volume (MCV) classification. The variables were followed normal distribution and homogeneous in variance. The levels within 3 groups of the continuous variables were analyzed using one-way ANOVA. categorical variables were analyzed using Chi-square test. To evaluate the relationship between the MELD score and macrocytic anemia were analyzed using univariate and multivariate linear regression analyses. Only variables with a
All data analysis and form generation were produced using the statistical package R (
The baseline characteristics of subjects with anemia were divided into three groups (
Baseline Characteristics of participants (
Variable | Macrocytic anemia | Normocytic anemia | Microcytic anemia |
|
---|---|---|---|---|
No. of participants | 72 | 330 | 55 | |
Mean corpuscular volume, fl | 105.40 ± 4.49 | 91.14 ± 5.09 | 72.92 ± 5.16 | <0.001 |
Age, years | 66.29 ± 13.79 | 65.32 ± 12.65 | 65.78 ± 13.72 | 0.617 |
Sex | 0.120 | |||
Male, n (%) | 60 (83.33) | 237 (71.82) | 42 (76.36) | |
Female, n (%) | 12 (16.67) | 93 (28.18) | 13 (23.64) | |
Smoke, n (%) | 28 (41.18) | 98 (31.11) | 23 (43.40) | 0.091 |
Alcohol, n (%) | 25 (37.88) | 88 (27.76) | 19 (35.85) | 0.171 |
Diabetes, n (%) | 8 (11.11) | 58 (17.58) | 11 (20.00) | 0.332 |
Hypertension, n (%) | 9 (12.50) | 55 (16.67) | 12 (21.82) | 0.376 |
Hemoglobin, g/L | 102.48 ± 21.07 | 102.22 ± 18.98 | 73.15 ± 18.14 | <0.001 |
Hemoglobin, categorical recoded, n (%) | <0.001 | |||
>90 | 57 (79.17) | 245 (74.24) | 9 (16.36) | |
60–90 | 11 (15.28) | 78 (23.64) | 36 (65.45) | |
<60 | 4 (5.56) | 7 (2.12) | 10 (18.18) | |
Blood glucose, mmol/L | 5.58 ± 2.00 | 6.19 ± 3.22 | 6.21 ± 2.57 | 0.588 |
SBP, mmHg | 128.36 ± 17.88 | 128.86 ± 19.29 | 127.64 ± 19.71 | 0.715 |
DBP, mmHg | 76.41 ± 10.40 | 77.56 ± 11.88 | 79.62 ± 12.71 | 0.378 |
Bilirubin,μmol/L | 56.3 (25.6–120.2) | 40.4 (25.1–75.0) | 34.1 (19.2–71.6) | <0.001 |
Creatinine,μmol/L | 63.5 (50.5–93.2) | 62.0 (49.0–86.0) | 66.0 (55.5–90.5) | 0.209 |
INR | 1.50 ± 0.69 | 1.27 ± 0.27 | 1.30 ± 0.22 | <0.001 |
eGFR | ||||
mL/min/1.73 m2 | 110.0 (74.2–761.2) | 112.3 (75.4–2,362.1) | 121.5 (83.0–377.2) | 0.091 |
ALB | 31.31 ± 6.82 | 31.04 ± 6.06 | 33.04 ± 5.87 | 0.090 |
AST | 60.5 (33.2–117.8) | 59.0 (34.0–105.8) | 47.0 (31.0–93.5) | 0.518 |
ALT | 37.5 (23.0–88.0) | 42.0 (24.0–74.0) | 37.0 (23.0–67.0) | 0.864 |
ALP | 121.0 (81.8–204.5) | 123.5 (89.0–186.8) | 138.0 (81.5–244.0) | 0.924 |
GGT | 80.5 (36.8–190.5) | 94.0 (45.0–217.5) | 107.0 (43.5–364.5) | 0.663 |
MELD | 17.02 ± 6.94 | 14.82 ± 4.20 | 15.16 ± 4.25 | <0.001 |
Complications, n(%) | ||||
UGB | 10 (13.89) | 50 (15.15) | 2 (3.64) | 0.069 |
SBP* | 1 (1.39) | 5 (1.52) | 0 (0.00) | 1.000 |
HE | 0 (0.00) | 12 (3.64) | 0 (0.00) | 0.093 |
Abbreviations: SBP, systolic blood pressure; DBP, diastolic blood pressure; MELD, model for end stage liver disease; UGB, upper gastrointestinal bleeding; SBP*, spontaneous bacterial peritonitis; HE, hepatic encephalopathy; INR, international normalized ratio; eGFR, estimated GFR; ALB, albumin; AST, aspartate aminotransferase; ALT, alanine aminotransferase; GGT, gamma-glutamyl transferase; ALP, alkaline phosphatase.
In univariate regression analysis, we found that a significant correlation was present both macrocytic anemia and the MELD score (
Univariate analysis for MELD score.
Statistics | β (95%CI) |
|
|
---|---|---|---|
Sex | |||
Female | 118 (25.82%) | Ref | |
Male | 339 (74.18%) | −0.57 (−1.58, 0.43) | 0.2627 |
Age | 65.5 ± 12.9 | −0.0 (−0.0, 0.0) | 0.786 |
Smoking | 149 (32.6%) | 0.5 (−0.5, 1.4) | 0.315 |
Drinking | 132 (28.9%) | 0.5 (−0.5, 1.4) | 0.354 |
Diabetes | 77 (16.8%) | −0.7 (−1.8, 0.5) | 0.271 |
Hypertension | 76 (16.6%) | −0.8 (−2.0, 0.4) | 0.179 |
Hemoglobin, g/L | |||
>90 | 311 (68.1%) | Ref | |
60–90 | 125 (27.4%) | 0.6 (−0.4, 1.6) | 0.248 |
<60 | 21 (4.6%) | 1.9 (−0.2, 4.0) | 0.074 |
Blood glucose | 6.1 ± 3.0 | −0.0 (−0.2, 0.1) | 0.827 |
SBP | 128.6 ± 19.1 | −0.0 (−0.0, 0.0) | 0.520 |
DBP | 77.6 ± 11.8 | 0.0 (−0.0, 0.0) | 0.988 |
MCV, fl | 91.20 ± 9.86 | 0.04 (0.00, 0.09) | 0.0485 |
Anemia classification | |||
Normocytic anemia | 330 (72.21%) | Ref | |
Macrocytic anemia | 72 (15.75%) | 2.20 (0.99, 3.41) | 0.0004 |
Microcytic anemia | 55 (12.04%) | 0.34 (−1.01, 1.70) | 0.6195 |
Abbreviations: MELD, model for end stage liver disease;
Relationship between MCV and MELD in different models.
Variable | Crude model | Model I | Model II | |||
---|---|---|---|---|---|---|
β (95%CI) |
|
β (95%CI) |
|
β (95%CI) |
|
|
MCV, fl | 0.04 (0.00, 0.09) | 0.0485 | 0.05 (0.00, 0.09) | 0.0428 | 0.05 (0.00, 0.10) | 0.0367 |
Anemia classification | ||||||
Normocytic anemia | References | References | References | |||
Macrocytic anemia | 2.20 (0.99, 3.41) | <0.001 | 2.31 (1.09, 3.52) | <0.001 | 2.40 (1.06, 3.74) | <0.001 |
Microcytic anemia | 0.34 (−1.01, 1.70) | 0.6195 | 0.39 (−0.97, 1.74) | 0.5762 | 0.29 (−1.17, 1.76) | 0.6958 |
Abbreviations: CI, confidence interval.
Model I adjusted for Sex and Age. Model II adjusted for Sex, Age, Smoking, Drinking, SBP, DBP.
The two-piece wise smooth curve for MCV-MELD association in decompensated HBV associated cirrhosis. MCV negatively correlated with MELD when the MCV was smaller or equal than 98.2 fl (regression coefficient = 0.793,95% CI—0.1, 0.1). There was a strong positive correlation if MCV greater than 98.2 fl (regression coefficient = 0.008,95% CI 0.1, 0.4) (
Two-piece piecewise regression and smooth curve-fitting for association between MCV and MELD stratified.
Threshold Effect Analysis of MCV and MELD using Piece-wise Linear Regression.
Inflection point of MCV | Effect size( |
95%CI |
|
---|---|---|---|
<98.2 | 0.0 | −0.1 to 0.1 | 0.793 |
≥98.2 | 0.2 | 0.1 to 0.4 | 0.008 |
Effect: MELD Cause: MCV
adjusted :Sex, Age, Smoking, Drinking, SBP, DBP.
In this study the analysis was done retrospectively, we suggested that macrocytic anemia (MCV >100 fl) related to the degree of liver damage in decompensated HBV associated cirrhosis patients. This variability persists even after adjusting for age, gender, smoking, drinking, SBP and DBP.
Positive association between MCV and MELD was found among HBV-associated decompensated cirrhosis. Our findings by the two-piece piece-wise regression model to display the relationship between MCV and MELD as non-linear relationship. Positive correlation was observed when the MCV was higher than 98.2 fl, while negative correlation occurred when the MCV was lower than 98.2 fl.
Macrocytosis, which is also known as MCV >100 fl, is not necessarily corelated with anemia. Moreover, in most of the cases it is unattached to anemia (
The significance of macrocytosis remains an underestimated issue in the past. Only a small number of studies had relevant reports (
Macrocytosis is considered a structural and functional abnormality of the erythrocyte membrane. Several potential pathological mechanisms may explain our observations. First, irrespective of the etiology vitamin deficits is common in patients with cirrhosis, such as vitamin B12 and folate deficiency (
The MELD score is approved for assessing the degree of liver diseases. These variables include prothrombin time, INR, serum bilirubin and creatinine level. MELD score changes with variations in these variables. Higher MELD scores associate with increased risks of death and hepatic events in cirrhosis. In our study, among the parameters of MELD score, bilirubin and INR showed an increase on patients with macrocytic anemia. However, there was no remarkable difference with creatinine and eGFR. Therefore, macrocytic anemia may not be relevant to renal injury in patients with decompensated HBV associated cirrhosis.
Several study limitations are noted. First, the main limitation of this study lies in its retrospective observational nature, the cross-sectional nature of our study does not permit the determination of causality between MCV and MELD. Second, this study included only Chinese participants, and therefore these findings may not be generalizable to other biogeographic ethnic groups. Third, we did not perform an analysis on the data of folate, serum vitamin B12 and reticulocyte count, which could provide a better understanding of macrocytic anemia in cirrhotic patients.
Macrocytic anemia was highly correlated with the degree of hepatic dysfunction and may be a reliable predictor for mortality in patients with decompensated HBV associated cirrhosis. We found a non-linear relationship between MCV and MELD. Moreover, further large-scale, well-designed and multicenter studies need to be conducted to confirm our conclusions, it is important to evaluate and investigate this association and to gain insight the underlying mechanisms.
The raw data supporting the conclusion of this article will be made available by the authors, without undue reservation.
The studies involving human participants were reviewed and approved by the research protocol with a waiver of written informed consent by the Ethics Committee of the First affiliated hospital of Dalian Medical university. Written informed consent for participation was not required for this study in accordance with the national legislation and the institutional requirements.
T-YZ designed the study. Q-WC and FL interpreted the data. T-YZ and L-YY drafted the paper. YZ designed the experiments, improved the manuscript. All the authors have read and approved the final manuscript.
This study was funded by the National Natural Science Foundation of China, grant number 81673728.
The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.
All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.
We thank everyone who participated in the study.