Pruritus in patients with chronic liver disease and serum autotaxin levels in patients with primary biliary cholangitis

A total of 2477 patients with chronic liver disease who visited Hiroshima University Hospital from 2016 to 2017 were retrospectively investigated for itch frequency and severity. Characteristics of the 2477 patients with chronic liver disease at the time of evaluation of pruritus are shown in Table 1. The study included 1239 males and 1238 females, aged 20 to 96 years (median, 67 years). Seven hundred eighteen patients (29.0%) were positive for hepatitis B surface antigen (HBsAg) (Chemiluminescent Enzyme Immunoassay; Abbott Laboratories, Tokyo, Japan). 1114 (45%) were positive for hepatitis C virus (HCV) antibody (third generation enzyme immunoassay; Chiron, Emeryville, CA, USA). Seven hundred eighty-five out of 1114 HCV antibody-positive patients had received anti-HCV therapies before this survey, out of which 775 had achieved sustained viral response. One hundred two patients (4.1%) had autoimmune hepatitis (AIH), diagnosed according to the criteria established by the International Autoimmune Hepatitis Group (IAIHG). One hundred nineteen patients (4.8%) had PBC, diagnosed according to the criteria established by the Intractable Hepato-Biliary Diseases Study Group of Japan. Two hundred eighty-nine patients (11.7%) had non-alcoholic fatty liver disease (NAFLD), diagnosed based on existence of hepatic steatosis, either by imaging or histology and exclusion of secondary causes of liver steatosis as alcohol, viral hepatitis, autoimmune disorder, and hereditary liver disease. One hundred thirty-five patients (5.5%) had alcohol liver damage, diagnosed based on chronic alcohol use presenting with liver disease and excluding other secondary causes of liver disease. Patients who had complications due to allergic or skin diseases assessed by medical interviews were excluded from this study. Of all patients, 384 patients (15.5%) had been treated with topical cream and/or oral antipruritic agents such as hydrocortisone, antihistamines, kappa-opioid receptor agonists, and moisturizers. Itch frequency was also investigated in 130 control subjects, including healthy subjects (subjects without liver disease) and patients without liver disease. FIB-4 index was calculated as a surrogate marker of liver fibrosis: age [years] × aspartate aminotransferase (AST) [U/L] / platelet counts [10⁹/L] × alanine aminotransferase [U/L]^1/2 [17]. All participants provided written informed consent to participate in the study according to the process approved by the ethical committee of Hiroshima University and conforming to the ethical guidelines of the Declaration of Helsinki.

Itch severity was self-assessed using pruritus scores based on a scale ranging from 0 (no itch) to 10 (maximum itch) using the numerical rating scale (NRS) [18]. All 2477 patients with chronic liver disease were evaluated by their pruritus scores.

Serum autotaxin levels were measured in 114 out of 119 PBC patients. Five PBC patients were removed due to lack of sufficient serum samples. Serum autotaxin antigen concentration was determined in frozen serum samples with a specific two-site enzyme immunoassay using AIA-360 system (Tosoh, Tokyo, Japan), as described previously [19].

Liver stiffness evaluation was performed using FibroScan (Echosens, Paris, France) with a standard probe (M probe). FibroScan measurements were carried out by three independent and experienced observers, blinded with respect to patient data. The objective was to obtain at least ten acceptable measurements. Results were expressed as the median and the IQR (kPa) of all valid measurements. According to the usual definition, liver stiffness was considered reliable when it included ≥10 valid measurements with a success rate ≥ 60% and IQR/M ≤ 0.30. Only reliable results were used for analysis.

The Chi-squared and Mann–Whitney U-tests were applied to detect significant associations. All statistical tests were two-sided, and P < 0.05 was considered significant. Correlation analysis was performed using Spearman’s rank-order correlation. Variables that achieved statistical significance (P < 0.05) on univariate analysis were entered into multiple logistic regression analysis to identify independent predictors of severe pruritus. All statistical analysis was performed using BellCurve for Excel ver.2.15 (Social Survey Research Information Co., Ltd.).

Seven hundred thirty-eight patients (29.8%) were evaluated as having ≥1 point pruritus severity by NRS (Fig. 1). The frequencies of pruritus in patients with chronic liver disease were significantly higher than those in subjects without liver disease (16.2%) (P < 0.001). Figure 2 shows the distribution of pruritus scores according to the etiology of chronic liver disease. The average NRS scores for HBV, HCV, AIH, PBC, NAFLD and alcohol were 1.15, 1.23, 1.38, 2.23, 1.43 and 1.51, respectively. NRS was high in patients with chronic liver disease, especially in those with PBC, as is generally expected.

We analyzed factors associated with severe pruritus, which we define as NRS ≥5. Univariate analysis showed that hemoglobin (P = 0.003), eosinophil count (P = 0.003), γ-glutamyl transpeptidase (P = 0.003), albumin (P < 0.001), estimated glomerular filtration rate (P = 0.022), PBC (P = 0.006), and hepatic encephalopathy or esophageal varices (P = 0.002) were each significantly associated with pruritus (Table 1). Multivariable analysis identified higher eosinophil count (OR = 1.001; P = 0.002), lower albumin levels (OR = 0.637; P < 0.001), and PBC (OR = 2.065 for others; P = 0.004) as independent factors associated with severe pruritus.

Severe pruritus (NRS ≥5) was reported more frequently in PBC patients. Therefore, we primarily focused on PBC patients in this study. A recent study demonstrated that serum autotaxin was correlated with the degree of liver fibrosis in PBC patients [20], and serum autotaxin levels are known to be high in cholestasis patients with pruritus [16]. Therefore, we analyzed the relationship between serum autotaxin level and liver fibrosis and pruritus in PBC patients. Serum autotaxin levels were measured in 114 out of 119 PBC patients (91 patients were NRS < 5 and 23 patients were NRS ≥ 5). Significant correlations between serum autotaxin values and liver fibrosis markers such as platelet count and liver stiffness were observed (Fig. 3a). Serum autotaxin values were also significantly associated with levels of hepatobiliary enzymes such as total bilirubin, AST, and alkaline phosphatase (Fig. 3b). The relationship between autotaxin levels and pruritus was analyzed; however, autotaxin levels were not correlated with severity of pruritus. In patients with PBC, no factor was significantly correlated with severe pruritus (Table 2).