Mean Platelet Volume in Patients with Liver Cirrhosis and Chronic Liver Disease: A Case-Control Study

Introduction. Liver cirrhosis (LC) is characterized by advanced damage to the hepatic architecture. Hemostasis is rebalanced in patients with LC because of the copresence of pro-thrombotic and pro-hemorrhagic pathomechanisms. With regards to primary hemostasis, thrombocytopenia, platelet function defects and low thrombopoietin (TPO) synthesis are counterbalanced by increased von Willebrand factor and low levels of ADAMTS13 in LC. Fewer data, however, are available regarding the platelet size, which is measured as mean platelet volume (MPV), in LC. MPV correlates with the degree of fibrosis and inflammation in patients with chronic liver disease (CLD). Moreover, MPV predicts the risk of death and thrombosis in healthy subjects and cardiovascular patients. We designed this study to investigate the MPV in patients with LC.

Methods. For this case-control study we included adult patients with LC or CLD and control subjects. All patients (LC and CLD) were consecutively enrolled among the outpatients followed up at the Hepatology outpatient clinic of Ospedale San Paolo (Milan). Control subjects were enrolled among healthcare workers of Ospedale San Paolo. All subjects had at least one complete blood count (CBC) performed at the laboratory of Ospedale San Paolo between 2015 and 2020 (normal range for MPV: 9.3-12.6 fL). All CBCs were performed with the same instrument during this time. We retrieved the necessary data to calculate the prognostic scores Child-Pugh and MELD-Na for all LC patients. The primary outcome was the comparison of the MPV (median and proportion of subjects with a value above the upper limit of the normal range) in the three groups. Additionally, the MPV was evaluated in LC patients classified according to their Child-Pugh score at the time of blood withdrawal and history of previous decompensation (defined by any among ascites, portosystemic encephalopathy, variceal bleeding, spontaneous bacterial peritonitis and portal vein thrombosis). Finally, in patients with LC, we evaluated the correlations between platelet count and the MELD-Na score at the time of blood withdrawal with the MPV.

Results. We enrolled 38 controls (21 females, median age 46 years, IQR 35-53), 120 patients with CLD (65 females, 60 years, 53-73) and 114 patients with LC (52 females, 67 years, 56-75). LC patients had significantly lower median platelet counts (124x10⁹/L, IQR 80-184) compared to CLD patients (220x10⁹/L, IQR 183-260) and controls (257x10⁹/L, IQR 216-287) (p<0,0001). LC patients’ Child-Pugh score was A for 59 subjects, B for 36 and C for 19 (with B and C scores indicating a greater severity of LC). Fifty-five LC patients (48%) had a history of previous decompensation. The median MPV was significantly higher both in LC patients (11.3 fL, IQR 10.4-12.3) and CLD patients (11.0 fL, IQR 10.2-11.8) compared to controls (10.7, IQR 10.0-11.1; p=0.0008 and p=0.005, respectively). The proportion of LC patients with a MPV higher than the upper limit of the reference range was 14.0% (95% CI, 8.2-21.8), significantly higher compared to controls (0%, 95% CI 0-9.3, p=0,0327), but non-significantly higher compared to CLD patients (8.3%, 95% CI 4.1-14.8%, p=0.2383). Neither the median MPV nor the proportion of patients with a MPV higher than the reference range significantly differed across Child-Pugh A, B and C LC patients. The proportion of LC patients with a MPV higher than the reference range was significantly higher in patients with previous decompensation (21.8%, 95% CI 11.8-35.0 vs 6.8%, 95% CI 1.9-16.5; p=0.04). There was an inverse correlation between the platelet count and the MPV (r=-0.37, p<0.0001) and a mild direct correlation between the MELD-Na score and the MPV (r=0.21, p=0.02) in LC patients.

Conclusions. We found that 14.0% of LC patients have a MPV higher than the reference range and that such proportion increases to 21.8% in patients with previous decompensation, but not in patients with a higher Child-Pugh score. MPV inversely correlated with the platelet count, which could suggest that MPV is increased in LC because of compensatory platelet production in response to platelet consumption and/or sequestration. The fact that MPV does not increase with increasing disease severity could be explained by the copresence of pathomechanisms with opposite effects on platelet size (inflammation, hypersplenism, decline in TPO synthesis) in patients with more advanced disease.