This study was performed according to PRISMA guidelines[15]. PubMed, EMBASE and Web of Science databases were searched comprehensively for relevant publications before January 1, 2015. We used the following terminologies in all possible combinations without restrictions: “hepatitis B virus”, “hepatitis C virus”, “HBV”, “HCV”, “intrahepatic cholangiocarcinoma”, “ICC”, “prognosis”, “prognostic”, “survival”, “clinical”, and “clinicopathological.” We scanned the reference lists of relevant publications for additional available researches.

Inclusion criteria were: (1) articles were written in English; (2) patients were diagnosed with ICC by pathology; (3) HBV or HCV infection was detected; and (4) sufficient information about HBV or HCV infection, overall survival (OS) or disease-free survival (DFS) and other clinicopathological parameters was given directly or could be calculated indirectly.

Exclusion criteria were: (1) non-English articles; (2) reviews, letters, case studies and conference records; and (3) duplicated data of previous research.

Two reviewers assessed the quality of the retrieved studies independently in accordance with the Newcastle-Ottawa scale (NOS)[16]. The NOS is composed of three dimensions: selection, comparability, and outcome. The publication was not included in our review if its quality score was too low.

Two reviewers independently scanned the eligible publications and extracted the original information from the included papers. All publications were double-checked and disagreements were settled by discussion. For each research, the following data were recorded: (1) the first author name and year of publication; (2) the number of cases; (3) the study design; (4) the clinical parameters, including age, gender, clinical stage, treatment and other clinicopathological features; (5) the markers detected for HBV and HCV infections; and (6) the hazard ratio (HR) and 95% confidence interval (CI) of HBC or HCV infection for OS and DFS. If the HR was unavailable, we used the total number of deaths and the total number of cases to calculate the HR. If the information was only available in Kaplan-Meier curves, we obtained the HR with Engauge Digitizer according to Parmar et al[17].

The statistical software Stata version 12.0 was used to synthesize the outcomes of the enrolled studies. The HR and 95%CI from each paper were applied for calculating pooled HR. The χ² and I² tests were used to evaluate heterogeneity between the studies, and P < 0.05 was defined as statistical significance. If there was no heterogeneity (P≥ 0.05), a fixed-effects model was used, and a random-effects model was applied if there was heterogeneity (P < 0.05). Egger’s test and Begg’s funnel plot were applied to assess publication bias.

A flowchart demonstrating the publication search and selection process is shown in Figure 1. Four hundred and six potentially eligible papers were retrieved in the initial search. According to the titles and abstracts, 344 publications (25 reviews, 14 conference records, 10 case reports, 3 letters, 1 animal study, 4 cell line studies, 8 non-English publications and 279 papers that had no relationship with our study) were excluded. Sixty-two full articles were captured, among which 49 were finally excluded due to the paucity of sufficient information on HBV/HCV and OS, DFS, or clinicopathological features. One additional paper was identified by manual search. Ultimately, our review enrolled a total of 14 studies. Of these publications chosen for further assessment, 13 investigated the correlation between HBV and specific parameters, and 5 studied the relationship between HCV and survival and clinicopathological features.

Open in New Tab   Full Size Figure   Download Figure

Figure 1 Flowchart of the literature search strategy. HBV: Hepatitis B virus; HCV: Hepatitis C virus; OS: Overall survival; DFS: Disease-free survival.

The main characteristics of each enrolled publication are shown in Table 1. In total, 14 studies involving 2842 cases were enrolled in the present research. The papers included in our study were published between 2002 and 2014. Eight papers enrolled less than 100 cases, and six papers enrolled more than 100 cases. The mean age of the enrolled patients ranged from 51.0 to 62.0 years in these studies. One study was prospective, and the other 13 were well-designed, retrospective studies.

Each of the included studies in our review was evaluated in accordance with the NOS standard. A study with a quality value of 6 stars or more was of high quality. According to the NOS, all publications enrolled in our study were high-quality studies (Table 1).

The meta-analysis evaluating the relationship between HBV infection and OS was performed on 11 studies. The pooled HR was 0.76 (95%CI: 0.70-0.83, Z = 6.14, P = 0.000) (Figure 2), and no heterogeneity existed (I² = 35.3%, P = 0.116). Two studies assessed the association of HBV infection with DFS; the pooled HR was 0.78 (95%CI: 0.66-0.94, Z = 2.67, P = 0.008) (Figure 2) without heterogeneity (I² = 0%, P = 0.780). These results indicated that patients with HBV infection had longer OS and DFS.

Open in New Tab   Full Size Figure   Download Figure

Figure 2 Forest plot of the hazard ratio for the association of hepatitis B virus infection and survival. The survival data are reported as overall survival and disease-free survival.

In the five studies evaluating the correlation between HCV infection and OS, no significant heterogeneity was found (I² = 6.8%, P = 0.368). The pooled HR was 2.64 (95%CI: 1.77-3.93, Z = 4.76, P = 0.000), suggesting that patients with HCV infection had a poorer prognosis (Figure 3).

Open in New Tab   Full Size Figure   Download Figure

Figure 3 Forest plot of the hazard ratio for the association between hepatitis C virus infection and survival. The prognostic information is reported as overall survival.

The subgroup meta-analysis was carried out for the relationship between HBV infection and OS (Table 2). When stratified by sample size, the pooled HRs were 0.72 (95%CI: 0.65-0.81) for studies with more than 100 subjects and 0.84 (95%CI: 0.72-0.98) for studies with less than 100 subjects (Figure 4A). This finding indicated that HBV infection was a favorable prognostic marker regardless of sample size. When stratified by tumor type, HBV infection was a favorable prognostic marker for mass-forming ICC (HR = 0.75, 95%CI: 0.62-0.90) and ICC without tumor-type restriction (HR = 0.74, 95%CI: 0.64-0.85) (Figure 4B). When stratified by study region, HBV infection was a favorable predictor for Chinese patients (HR = 0.75, 95%CI: 0.69-0.82), but not for patients in other countries (HR = 1.10, 95%CI: 0.67-1.80) (Figure 4C). In the subgroup analysis by mean age, the pooled HRs were 0.71 (95%CI: 0.58-0.87) for patients with a mean age more than 55 and 0.80 (95%CI: 0.67-0.95) for patients with a mean age less than 55 (Figure 4D). These results suggest that HBV infection indicates a favorable prognosis regardless of patient age.

Open in New Tab   Full Size Figure   Download Figure

Figure 4 Forest plots of the hazard ratio for the association of hepatitis B virus infection with overall survival by subgroup analysis. A: By sample size; B: By tumor type; C: By study region; D: By mean age.

An analysis of the pooled data revealed that HBV infection in ICC patients was associated with specific clinicopathological features (Table 3). Four of the studies assessed the association of HBV infection with gender and the pooled OR was 1.91 (95%CI: 1.06-3.44) (Figure 5A). This finding suggests that HBV infection occurs more commonly in male patients. Three studies evaluated the correlation between HBV infection and aspartate transaminase (AST) levels. The pooled OR (1.93, 95%CI: 1.11-3.35) showed that HBV infection was associated with elevated AST (Figure 5B). Four of the studies investigated the relationship between HBV infection and tumor markers, including alpha-fetoprotein (AFP) and carbohydrate antigen 19-9 (CA19-9). The results demonstrate that HBV-infected cases had a higher level of AFP (OR = 3.86, 95%CI: 2.58-5.78) and a lower incidence of CA19-9 (OR = 0.47, 95%CI: 0.34-0.65) compared to the control group (Figure 5C, D). Additionally, four publications identified an association between HBV infection and cirrhosis; the pooled OR was 6.44 (95%CI: 4.33-9.56), which indicates that HBV infection was correlated with cirrhosis (Figure 5E). When performing a meta-analysis of HBV infection and tumor capsule formation, we found that HBV infection was correlated with a higher proportion of capsule formation (OR = 6.04, 95%CI: 3.56-10.26) (Figure 5F). Moreover, it was found that lymph node metastasis occurred less in patients (OR = 0.39, 95%CI: 0.25-0.58) (Figure 5G).

Open in New Tab   Full Size Figure   Download Figure

Figure 5 Forest plots of odds ratios for the association of hepatitis B virus infection with clinicopathological features. A: Meta-analysis for gender; B: Meta-analysis for AST levels; C: Meta-analysis for AFP levels; D: Meta-analysis for CA19-9 levels; E: Meta-analysis for cirrhosis; F: Meta-analysis for capsule formation; G: Meta-analysis for lymph node metastasis.

We also found that HBV infection had no relation to ALT level, TBIL level, γ-GT level, tumor differentiation, tumor location, tumor number, tumor size or vascular invasion. The pooled ORs were 1.23 (95%CI: 0.64-2.35), 0.91 (95%CI: 0.62-1.33), 0.77 (95%CI: 0.43-1.38), 0.86 (95%CI: 0.41-1.80), 0.76 (95%CI: 0.31-1.87), 0.91 (95%CI: 0.57-1.46), 0.72 (95%CI: 0.46-1.14) and 1.10 (95%CI: 0.49-2.43), respectively (Table 3).

There was no publication bias for the meta-analysis of the impact of HBV infection on patient survival (Begg’s test, P = 0.938; Egger’s test, P = 0.923) (Figure 6A). Additionally, no publication bias existed for studies regarding HCV infection and overall survival in this research (Begg’s test, P = 0.142; Egger’s test, P = 0.157) (Figure 6B).

Open in New Tab   Full Size Figure   Download Figure

Figure 6 Funnel plots of the studies included in the meta-analysis. Hepatitis B virus and overall survival (OS); B: Hepatitis C virus and OS.

The incidence and mortality rates of intrahepatic cholangiocarcinoma (ICC) have been increasing globally in recent decades. Hepatitis B virus (HBV) and hepatitis C virus (HCV) infections are correlated with the increasing incidence rate of ICC; however, the prognostic significance of HBV and HCV infections in ICC patients remains unknown.

In this research, a meta-analysis was performed to identify the prognostic value of HBV and HCV infections in patients with intrahepatic cholangiocarcinoma.

This research is the first comprehensive meta-analysis to identify the correlation of HBV and HCV infections with survival and clinical parameters in intrahepatic cholangiocarcinoma.

This study shows that HBV infection is related with a better prognosis and certain clinical features in ICC patients, while HCV infection is correlated with shortened overall survival in ICC patients. Thus, HBV and HCV infections could be useful prognostic markers for ICC.

Intrahepatic cholangiocarcinoma is one of the common hepatic malignancies. Pathologically, it consists of altered epithelial cells which originate from the biliary system in the liver.

This is a well-performed meta-analysis of currently available studies on the prognostic role of HBV and HCV infections in patients with intrahepatic cholangiocarcinoma. The results are interesting and suggest a favorable prognosis of ICC associated with HBV infection in comparison with HCV infection.