Accepted Manuscript

Hepatitis C Virus Infection is Associated With Increased Cardiovascular Mortality: AMeta-analysis of Observational Studies

Salvatore Petta, Marcello Maida, Fabio Salvatore Macaluso, Marco Barbara, AnnaLicata, Antonio Craxì, Calogero Cammà

PII: S0016-5085(15)01322-0DOI: 10.1053/j.gastro.2015.09.007Reference: YGAST 60028

To appear in: GastroenterologyAccepted Date: 3 September 2015

Please cite this article as: Petta S, Maida M, Macaluso FS, Barbara M, Licata A, Craxì A, Cammà C,Hepatitis C Virus Infection is Associated With Increased Cardiovascular Mortality: A Meta-analysis ofObservational Studies, Gastroenterology (2015), doi: 10.1053/j.gastro.2015.09.007.

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service toour customers we are providing this early version of the manuscript. The manuscript will undergocopyediting, typesetting, and review of the resulting proof before it is published in its final form. Pleasenote that during the production process errors may be discovered which could affect the content, and alllegal disclaimers that apply to the journal pertain.

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPTHepatitis C Virus Infection is Associated With Increased Cardiovascular Mortality: A Meta-

analysis of Observational Studies

AUTHORS: Salvatore Petta1, Marcello Maida1, Fabio Salvatore Macaluso1, Marco Barbara1, Anna

Licata1, Antonio Craxì1, Calogero Cammà1.

INSTITUTIONS:

1Sezione di Gastroenterologia, Di.Bi.M.I.S, Università di Palermo, Italia

SHORT TITLE: Cardiovascular Risk and HCV

CORRESPONDING AUTHOR: Dr. Salvatore Petta, Sezione di Gastroenterologia, Dipartimento

Biomedico di Medicina Interna e Specialistica, Piazza delle Cliniche, 2, 90127 Palermo, Italy

Phone: +39 091 655 2145. Fax +39 091 655 2156. E-mail: salvatore.petta@unipa.it;

petsa@inwind.it

ABBREVIATIONS: HCV: hepatitis C virus; CHC: chronic hepatitis C ; CVD : cardiovascular

disease ; CVV cardio-cerebrovascular.

FINANCIAL SUPPORT: No conflict of interest exists.

WORD COUNT: 3001

Tables/Figures/Supplemental materials: 2/4/6

CONFLICT OF INTEREST STATEMENT: None.

Author contributions: S. Petta, M. Maida, F.S. Macaluso, M. Barbara, A. Licata, A. Craxì, C.

Cammà take full responsibility for the study design, data analysis and interpretation, and

preparation of the manuscript. All authors were involved in planning the analysis and drafting the

manuscript. All authors approved the final draft manuscript.

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPTABSTRACT

Background & Aims: There have been many studies of the effects of hepatitis C virus (HCV)

infection on cardiovascular risk, but these have produced ambiguous results. We performed a meta-

analysis of these studies to systematically assess the risk of HCV infection on cardiovascular

disease (CVD)-related morbidity and mortality.

Methods: We searched PubMed Central, Medline, Embase, and Cochrane Library, as well as

reference lists of articles, for studies published through July 2015 that compared the occurrence of

CVD between HCV-infected and uninfected subjects, or assessed the prevalence of HCV infection

among subjects with CVDs. In total, 22 studies were analyzed. Data on the patient populations and

outcomes were extracted from each study by 3 independent observers and combined by a random-

effects model.

Results: Compared to uninfected individuals (controls), HCV-infected patients had increased risks

of CVD-related mortality (odds ratio [OR], 1.65; 95% confidence interval [CI], 1.07–2.56; P=.02),

carotid plaques (OR, 2.27; 95% CI, 1.76–2.94; P<.001), and cerebro-cardiovascular events (OR,

1.30; 95% CI 1.10–1.55; P=.002). Significant heterogeneity was observed in the risk of

cerebrocardiovascular disease among individuals with HCV infection. The effect of HCV infection

on cerebrocardiovascular disease was stronger in populations with a higher prevalence of diabetes

(>10%) or hypertension (>20%) (OR, 1.71; 95% CI, 1.32–2.23; P<.001 for both).

Conclusion: In a meta-analysis of published studies, individuals with HCV infections were found

to be at increased risk for CVD-related morbidity and mortality—especially those with diabetes and

hypertension.

KEY WORDS: chronic liver disease, cirrhosis, heart disease, blood vessel

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPTINTRODUCTION

Hepatitis C virus (HCV) infection is a leading cause of morbidity and mortality worldwide, with a

mortality rate that currently surpasses that of immunodeficiency virus infections [1-2]. As expected,

several longitudinal studies showed that the mortality rate of patients with chronic hepatitis C

(CHC) is higher compared to that of uninfected subjects [3-5]. These studies also showed that CHC

patients experience increases in both liver disease- and cardiovascular disease (CVD)-related

mortality, although with contrasting results [4-8]. Consistent with these findings, clinical studies

have shown a higher prevalence of metabolic disorders that are CVD risk factors, specifically type 2

diabetes mellitus (DM) [9], insulin resistance [10], and hepatic steatosis [11], in HCV patients

compared to uninfected controls. Moreover, recent data have identified HCV infection as a risk

factor for subclinical [12-26] and clinical cardiovascular alterations [27-45]. However, the results of

these published studies have been inconsistent, and the overall impact of HCV infection on CVD-

related morbidity and mortality is difficult to evaluate.

Considering the availability of new antiviral therapeutic strategies with increased

effectiveness and excellent tolerability profiles, a definitive estimate of the impact of HCV infection

on CVD-related risk is strongly needed [46]. Therefore, the aim of this meta-analysis was to

estimate whether patients infected with HCV, compared to uninfected controls, evidenced increased

rates of CVD-related mortality, carotid atherosclerosis, and cerebro-cardiovascular (CCV) events .

METHODS

Information sources and search strategies

The Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) statement

was followed [47]. Primary sources of the reviewed studies, including non-English sources, were

PubMed Central/Medline, Embase, and the Cochrane Library, which were systemically searched for

records up to July 2015. Searches included combinations of the keywords “hepatitis c,” “hepatitis c

virus,” “hepatitis non A non B,” and “HCV,” with one or more of the following in both the title

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPTand/or the abstract: “cardiovascular disease,” “cardiac outcomes,” “heart,” “heart disease,”

“cardiovascular mortality,” “cardiac mortality,” “atherosclerosis,” “carotid atherosclerosis,”

“intima-media thickness,” “myocardial infarction,” “myocardial injury,” “angina,” “coronary

disease,” “congestive heart failure,” “stroke,” “transitory ischemic attack,” “cerebrovascular

disease,” “cerebrovascular outcome,” and “cardiovascular outcomes”. Database searches were

supplemented with literature searches of reference lists from potentially eligible articles by all the

three reviewers (FSM, MM, and SP) to find additional studies.

Eligibility criteria

Studies were included in the meta-analysis if they compared the prevalence of CVD between HCV-

infected subjects and uninfected controls, or if they compared the prevalence of HCV infection

among subjects with different CVDs.

Searches

Among the 1,861 records identified through electronic search strategies, 596 duplicates and

1,227 irrelevant studies were excluded. Thirty-eight potentially relevant reports [4-8,12-16,18-45]

were identified and retrieved for detailed evaluation (Figure 1). Among these, seven were excluded

for inability to extract exact numbers of subjects and/or numbers of events from case and/or control

groups [4,13-15,37,42,45], eight were excluded for irrelevant data or absence of considered

outcomes [12,16,38-41,43,44], and one was excluded for duplication bias [6]. The other 22 studies,

all of which assessed the prevalence of CVDs relative to HCV status, met the inclusion criteria.

Data from these studies were collected directly from the corresponding paper [5,7,8,19-21,23-

28,31,32,36,37] or obtained by contacting the authors [18,22,29,30,33,34] (Figure 1). More

specifically, after identifying 1,861 records in electronic searches, all three reviewers (FSM, MM,

and SP) independently evaluated the titles and abstracts, removed duplicates and irrelevant studies,

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPTselected 38 potentially eligible articles, and read each full report to identify the 22 studies included

in the meta-analysis.

Data collection process

The studies were first reviewed by three reviewers (FSM, MM, and SP) using a list of predefined,

pertinent issues that concerned the characteristics and outcomes of HCV-infected and uninfected

groups. In some studies, HCV infection was defined based on the presence of anti-HCV only

[5,19,21,25,27], whereas the presence of both anti-HCV and HCV-RNA dictated HCV infection in

other studies [7,18,20,22-24,26,28,29,31-36]. Two other studies defined HCV infection based on

the presence of anti-HCV or HCV-RNA [8, 30]. When the data were available, patients from the

HCV cohorts who either had a virological profile of an inactive infection (anti-HCV positive but

HCV-RNA negative) [18,35] or underwent antiviral therapy [29,34] were excluded from the

analysis.

Several study- and patient-level variables were extracted from all eligible studies and

entered into a structured database. Study-level variables included the last name of the first author,

publication year, country where the study was conducted, study design, number of subjects, and

outcomes measured. Patient-level variables included the mean age, sex, race, cholesterol level, and

body mass index (BMI) of patients, as well as the presence of hypertension, DM, obesity, and

smoking habit among patients. Study quality was evaluated by the 9-star Newcastle-Ottawa Scale

[48], a validated technique for assessing the quality of nonrandomized studies in meta-analyses.

Discrepancies among reviewers about qualitative and quantitative data collection were infrequent

(overall interobserver variation < 10%) and resolved by discussion.

Data synthesis

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPTSummary parameters were CVD-related mortality, carotid plaques, and CCV events, defined as

ischemic stroke, myocardial infarction, angina, congestive heart failure, or transient ischemic attack.

Crude rates of CVD-related morbidity and mortality were retrieved in HCV-positive and HCV-

negative individuals. The odds ratio (OR) was computed for each study by using the observed

numbers of CVD cases and deaths.

Overall ORs and the corresponding 95% confidence intervals (CIs) of the frequencies of

events in the HCV-positive and HCV-negative groups were calculated by the DerSimonian and

Laird random-effects model [49], which accounts for both within-study variance and between-study

heterogeneity. The random-effects model was used because we believe that the relevant variation in

effects was a consequence of several interstudy differences. All statistical analyses were performed

by Rev MAN [50].

Two different methods were used to explore and explain the diversity among the results of

different studies: subgroup analyses and meta-regression. A Chi-square (χ2) test for interaction [51]

was used to examine whether the cardiovascular risk associated with HCV infection varied

significantly among subgroups.

We examined the extent to which differences in study outcomes could be explained by

differences in characteristics at the patient level (i.e., age, sex, BMI, arterial hypertension, DM, high

cholesterol/dyslipidemia, and smoking habit) or study level (i.e., publication year, geographical

source of the study, design, study size, and quality). Several independent explanatory variables were

included in univariate meta-regression models [52]. The dependent variable in the regression

analyses was the logarithmic OR of cardiovascular risk in the HCV-positive vs. HCV-negative

groups. Statistical analyses were performed with weighted univariate linear regression models, with

the inverse of the variance of the cardiovascular risk being used for weight. Regression analyses

were performed in the R statistical package [53]. Publication bias was assessed by the Egger

regression symmetry test for publication bias. For all analyses, P < 0.05 was considered

statistically significant.

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

RESULTS

Characteristics of the studies

Main features of the studies included in the meta-analysis are shown in Table 1 and Supplemental

Tables 1 and 2. Three cohort studies [5,7,8] including 68,365 patients considered CVD-related

mortality as an outcome. The sample size of each cohort ranged from 19,636 to 28,546 subjects.

These studies included a prospective Asiatic community-based cohort study [7], a retrospective

cohort study of HCV-infected blood donors in the United States [5], and an Australian cohort study

of opioid-dependent people [8].

When considering carotid atherosclerosis as an outcome, we included nine studies [18-26]

that assessed the presence of carotid plaques. Seven of these studies [18-24] also examined intima-

media thickness (IMT). The nine studies collectively enrolled 9,083 patients, and the sample size of

each cohort ranged from 72 to 4,784 subjects. The prevalence of HCV infection varied greatly (1.2–

10.5%). There were four studies performed in European cohorts, two in Asiatic cohorts, two in

African cohorts, and one in a U.S. cohort.

The search for CCV events as an outcome led to the inclusion of eight studies [27-33,35]

(six cohort and two case-control studies), which together enrolled 390,602 patients. The sample size

of each cohort ranged from 582 to 171,665 subjects. The prevalence of HCV infection among the

studies varied from 6.2% to 9.6%. Six studies were performed on U.S. or European cohorts,

whereas two studies were performed on Asiatic populations. Two studies [34,36] were not

considered in the overall analysis of CCV events due to duplication bias [29,30], but were included

in subanalyses of cardiovascular or cerebrovascular events alone.

Cardiovascular-related mortality and HCV infection

Three cohort studies (68,365 patients, 735 deaths) examined the effect of HCV infection on CVD-

related mortality (Figure 2). Two studies [5,7] reported a significantly higher mortality rate in HCV-

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPTinfected patients compared to uninfected controls. The third study [8] did not show any effect. The

pooled estimate of the effect of HCV infection was significant (OR 1.65, 95% CI 1.07–2.56, P =

0.02) with a significant heterogeneity (I2 = 76%, P = 0.02). Univariate regression analyses showed

that CVD-related mortality was not significantly affected by the geographical source of the study

(West vs. non-West) or study quality (both variables were available in all three studies; Table 2).

Other potential patient-level covariates (i.e., well-known cardiovascular risk factors) were not

available in all three studies.

Carotid atherosclerosis and HCV infection

Nine case-control studies (9,083 patients in total, 1,979 patients with carotid plaques) examined the

effect of HCV infection on carotid plaques (Figure 3A). HCV infection favored the presence of

carotid plaques in eight of the nine studies, but this difference was significant in only five studies.

The pooled estimate of the effect of HCV infection was significant (OR 2.27, 95% CI 1.76–2.94, P

< 0.001), without significant heterogeneity (I2 = 31%, P = 0.17).

Univariate regression analyses of seven studies showed that the impact of HCV infection on

the presence of carotid plaques was directly affected by smoking status (P = 0.02; Table 2). The

pooled OR for this relationship was 2.66 (95% CI 1.96–3.61, P < 0.001) in populations with a high

prevalence of smoking (>20%) [54] compared to 1.07 (95% CI 0.57–2.01, P = 0.84) in populations

with a low prevalence of smoking (<20%) [54], with a significant interaction (χ2 = 6.45; P = 0.01,

Figure 3B). Seven case-control studies (3,749 patients) were considered in terms of the impact of

HCV infection on IMT (Supplemental Material 1). We observed a significant pooled estimate

(mean difference 0.09, 95% CI 0.03–0.16, P < 0.001), albeit in the presence of heterogeneity (I2 =

90%; P = 0.007; Supplemental Figure 1), similar to the results of the analysis of carotid plaques.

Cerebro-cardiovascular events and HCV infection

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPTThe effect of HCV infection on CCV events was evaluated in eight studies (390,602 patients in

total, 18,388 events; Figure 4A). Although HCV infection had a significant impact on CVD-related

morbidity in five of the eight studies, three studies showed a decrease in the number of CCV events.

The pooled estimate of the effect of HCV infection on CCV events was significant (OR 1.30, 95%

CI 1.10–1.55, P = 0.002) but had significant heterogeneity (I2 = 91%, P < 0.001). Upon separating

CCV into cerebrovascular or cardiovascular events, we confirmed that HCV infection was

associated with both cardiovascular (OR 1.20, 95% CI 1.03–1.40; P = 0.02; Supplemental Figure

2A) and cerebrovascular (OR 1.35, 95% CI 1.00–1.82, P = 0.05; Supplemental Figure 2B) events.

Univariate regression analyses showed that the impact of HCV infection on CCV risk was

higher in cohort studies (P = 0.01), studies of older patients (P < 0.001), and studies with a higher

prevalence of hypertension (P = 0.008) or DM (P < 0.001; Table 2). We performed subgroup

analyses to evaluate whether there were differential effects of HCV infection in subgroups of

patients defined on the basis of study design (cohort vs. case-control), mean cohort age (threshold

of 50 years), prevalence of DM (threshold of 10%) [55], or prevalence of hypertension (threshold of

>20%) [56]. Among studies that reported the prevalence of DM, the pooled OR was 1.71 (95% CI

1.3–2.23, P < 0.001) for studies with a prevalence > 10% and 0.83 (95% CI 0.72–0.96, P = 0.01)

for studies with a prevalence < 10%, with a significant interaction (χ2 = 22.48, P < 0.001; Figure

4B). Among studies that reported the prevalence of hypertension, the pooled OR was significant for

studies with a prevalence > 20% (OR 1.71, 95% CI 1.32–2.23, P < 0.001) but not for those with a

prevalence < 20% (OR 1.00, 95% CI 0.72–1.40; P = 1.00) with a significant interaction (χ2 = 6.21,

P = 0.01; Figure 4C). The pooled OR was 1.21 (95% CI 1.05–1.39, P = 0.008) for cohort studies

and 2.01 (95% CI 0.31–13.04, P = 0.46) for case-control studies, without a significant interaction

(χ2 = 0.28, P = 0.60; Figure 4D). The pooled OR was 2.46 (95% CI 0.60–10.16; P = 0.21) for

studies with a mean population age > 50 years and 1.35 (95% CI 0.97–1.88, P = 0.07) for those

with a mean population age < 50 years, without a significant interaction (χ2 = 0.65, P = 0.42; Figure

4E).

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPTFinally, the Egger test for publication bias showed no significant risk of having missed or

overlooked studies in terms of overall mortality (P = 0.385), CCV events (P = 0.682), carotid

atherosclerosis (P = 0.831), or IMT (P = 0.357).

DISCUSSION

This meta-analysis of aggregate data from 22 studies shows that compared to uninfected controls,

HCV-infected individuals have increased risks of CVD-related mortality and subclinical carotid

atherosclerosis. We observed a slightly significant increase in CCV events among HCV-infected

patients, despite the high heterogeneity among studies that was mostly related to the prevalence of

DM and hypertension.

Historically, HCV infection has been considered to affect only the liver, via the development

of cirrhosis and its complications. HCV infection has also been shown to increase the risk of overall

mortality and the occurrence of extrahepatic complications, such as lymphoproliferative disorders

and metabolic alterations (insulin resistance and DM). However, recent studies have suggested that

HCV-infected patients have an increased risk of developing CVDs [12-45]. To our knowledge, our

meta-analysis clearly highlights, for the first time, that HCV infection increases the risk of CVD-

related mortality.

We found a two-fold higher risk of subclinical carotid plaques among HCV-infected

individuals compared to uninfected controls, without significant heterogeneity among studies, as

well as an increased risk of carotid thickening in HCV-infected subjects. These findings, which are

in agreement with our now reported increased risk of CVD-related mortality among HCV-infected

individuals, add relevant data to the debated issue of atherosclerosis and HCV infection [13-15,18-

26]. Our data suggest that HCV infection is another risk factor for carotid atherosclerosis, and that

this association is affected by the smoking habit in populations with a high prevalence of smokers.

However, the small number of patients in studied populations with a low prevalence of smokers

limits the strength of this conclusion.

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPTAnother recently debated issue regards the potential association between HCV infection and

CCV events. Some studies in the literature have reported that HCV infection is a risk factor of CCV

events, whereas others have reported a protective effect of infection [27-45]. In our meta-analysis,

we found a significant increase in the overall risk of CCV events among HCV-infected individuals

compared to uninfected controls. As expected, we also found a high heterogeneity among the

studies. Therefore, we performed further analyses to identify groups of patients in whom the effect

of the infection was either pronounced or lacking. Nevertheless, our subgroup analyses were unable

to explain the observed heterogeneity fully, and the association of HCV infection remained

significant in populations with a higher prevalence of cardiovascular risk factors, DM, or

hypertension. Our data agree with those of Hsu and colleagues [43], who recently showed that the

risk of peripheral artery disease is higher in HCV-infected patients than in uninfected patients, and

that this risk further increases with the presence of comorbidities.

Although our meta-analysis was not designed to explore the reasons for the association

between HCV infection and increased cardiovascular risk, some hypotheses can be proposed

according to the available experimental and clinical data. Several studies have identified a higher

prevalence of metabolic comorbidities related or not to HCV infection. Others have identified

correlations between CVDs and the proinflammatory-profibrogenetic HCV-related environment

and/or the severity of liver damage. A direct viral activity could also potentially explain these

correlations have also been reported [57].

From a clinical standpoint, the results of our meta-analysis suggest that HCV infection

increases the cardiovascular risk, particular for individuals who already have cardiovascular risk

factors such as DM and hypertension. Although effective and safe oral antiviral regimens are

available [46], more information is needed to confirm whether anti-HCV medications will decrease

the cardiovascular risk, as suggested in some studies [34,37].

The results of this meta-analysis are subject to several limitations, such as differences in the

nature of the control groups among the different studies. We attempted to control for these

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPTdifferences by including patient- and study-level covariates. However, there were likely other

potentially important confounders for which we did not control and that might have affected the

results. Furthermore, the results only describe variations between studies, and not between patients,

because they reflect group averages rather than individual data. More detailed comparisons could be

achieved with meta-analyses of individual patient data. Lack of data on the severity of liver fibrosis,

a factor associated with the severity of cardiovascular alterations in CHC [17,23] and nonalcoholic

fatty liver disease [58,59], could affect the accuracy of the results.

With our extensive computer search for studies, we are confident that no important

published trials were overlooked. Publication bias was not substantial and was considered unlikely

to change the direction of our pooled estimates of observed HCV effects. Although issues of quality

assessment may be somewhat important in this review, the quality of the individual studies did not

seem to bias the results of our meta-analysis.

The available evidence is sufficient to conclude that HCV infection increases cardiovascular

risk, including risks of subclinical carotid atherosclerosis, CCV events, and cardiovascular

mortality. Furthermore, the effect of HCV infection on cardiovascular risk appears to be especially

pronounced in populations with a high prevalence of smoking, hypertension, or DM.

References

1. Global burden of disease (GBD) for hepatitis C. Global Burden Of Hepatitis C

Working Group. J Clin Pharmacol 2004;44:20-29.

2. Hepatitis B and C in the EU neighbourhood: prevalence, burden of disease and

screening policies. European Centre for Disease Prevention and Control,

September 2010.

3. Neal KR ; Trent Hepatitis C Study Group, Ramsay S, Thomson BJ, Irving WL.

Excess mortality rates in a cohort of patients infected with the hepatitis C virus: a

prospective study. Gut. 2007;56:1098-104.

4. Kristiansen MG , Løchen ML, Gutteberg TJ, Mortensen L, Eriksen BO, Florholmen

J. Total and cause-specific mortality rates in a prospective study of community-

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPTacquired hepatitis C virus infection in northern Norway. J Viral Hepat 2010 Mar 8,

PubMed PMID: 20337923.

5. Guiltinan AM , Kaidarova Z, Custer B, Orland J, Strollo A, Cyrus S et al. Increased

all-cause, liver, and cardiac mortality among hepatitis C virus-seropositive blood

donors. Am J Epidemiol 2008;167:743-50.

6. Lee MH, Yang HI, Wang CH, Jen CL, Yeh SH, Liu CJ, et al. Hepatitis C virus

infection and increased risk of cerebrovascular disease. Stroke 2010

Dec;41(12):2894-900.

7. Lee MH, Yang HI, Lu SN, Jen CL, You SL, Wang LY, et al; R.E.V.E.A.L.-HCV

Study Group. Chronic hepatitis C virus infection increases mortality from hepatic

and extrahepatic diseases: a community-based long-term prospective study. J

Infect Dis 2012;206:469-77.

8. Vajdic CM , Pour SM, Olivier J, Swart A, O'Connell DL, Falster MO, et al. The

impact of blood-borne viruses on cause-specific mortality among opioid dependent

people: An Australian population-based cohort study. Drug Alcohol Depend

2015;152:264-71.

9. Wang CS , Wang ST, Yao WJ, Chang TT, Chou P. Hepatitis C virus infection and

the development of type 2 diabetes in a community-based longitudinal study. Am J

Epidemiol 2007;166:196-203.

10. Petta S , Cammà C, Di Marco V, Alessi N, Cabibi D, Caldarella R, et al. Insulin

resistance and diabetes increase fibrosis in the liver of patients with genotype 1

HCV infection. Am J Gastroenterol 2008;103:1136-44.

11. Petta S , Amato M, Cabibi D, Cammà C, Di Marco V, Giordano C, et al. Visceral

adiposity index is associated with histological findings and high viral load in patients

with chronic hepatitis C due to genotype 1. Hepatology 2010;52:1543-52.

12. Tomiyama H , Arai T, Hirose K, Hori S, Yamamoto Y, Yamashina A. Hepatitis C

virus seropositivity, but not hepatitis B virus carrier or seropositivity, associated with

increased pulse wave velocity. Atherosclerosis 2003;166:401-3.

13. Sosner P , Wangermez M, Chagneau-Derrode C, Le Moal G, Silvain C.

Atherosclerosis risk in HIV-infected patients: the influence of hepatitis C virus co-

infection. Atherosclerosis 2012;222:274-7.

14. Völzke H , Schwahn C, Wolff B, Mentel R, Robinson DM, Kleine V, et al. Hepatitis

B and C virus infection and the risk of atherosclerosis in a general population.

Atherosclerosis 2004;174:99-103.

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT15. Masiá M , Padilla S, Robledano C, Ramos JM, Gutiérrez F. Evaluation of

endothelial function and subclinical atherosclerosis in association with hepatitis C

virus in HIV-infected patients: a cross-sectional study. BMC Infect Dis 2011;11:265.

16. Maruyama S , Koda M, Oyake N, Sato H, Fujii Y, Horie Y, et al. Myocardial injury in

patients with chronic hepatitis C infection. J Hepatol 2013;58:1

17. Perticone M , Miceli S, Maio R, Caroleo B, Sciacqua A, Tassone EJ, et al. Chronic

HCV infection increases cardiac left ventricular mass index in normotensive

patients. J Hepatol. 2014 Oct;61(4):755-60.

18. Mostafa A , Mohamed MK, Saeed M, Hasan A, Fontanet A, Godsland I, et al.

Hepatitis C infection and clearance: impact on atherosclerosis and cardiometabolic

risk factors. Gut 2010;59:1135-40.

19. Ishizaka Y , Ishizaka N, Takahashi E, Unuma T, Tooda E, Hashimoto H, et al.

Association between hepatitis C virus core protein and carotid atherosclerosis. Circ

J 2003;67:26-30.

20. Targher G , Bertolini L, Padovani R, Rodella S, Arcaro G, Day C. Differences and

similarities in early atherosclerosis between patients with non-alcoholic

steatohepatitis and chronic hepatitis B and C. J Hepatol 2007;46:1126-32.

21. Caliskan Y , Oflaz H, Pusuroglu H, Boz H, Yazici H, Tamer S, et al. Hepatitis C

virus infection in hemodialysis patients is not associated with insulin resistance,

inflammation and atherosclerosis. Clin Nephrol 2009;71:147-57.

22. Adinolfi LE , Restivo L, Zampino R, Guerrera B, Lonardo A, Ruggiero L, et al.

Chronic HCV infection is a risk of atherosclerosis. Role of HCV and HCV-related

steatosis. Atherosclerosis 2012;221:496-502.

23. Petta S , Torres D, Fazio G, Cammà C, Cabibi D, Di Marco V, et al. Carotid

atherosclerosis and chronic hepatitis C: a prospective study of risk associations.

Hepatology 2012;55:1317-23.

24. Bilora F , Campagnolo E, Rinaldi R, Rossato A, Arzenton M, Petrobelli F. Carotid

and femoral atherosclerosis in chronic hepatitis C: a 5-year follow-up. Angiology

2008 Dec-2009;59:717-20.

25. Ishizaka N , Ishizaka Y, Takahashi E, Tooda Ei, Hashimoto H, Nagai R, Yamakado

M. Association between hepatitis C virus seropositivity, carotid-artery plaque, and

intima-media thickening. Lancet. 2002 Jan 12;359(9301):133-5.

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT26. Tien PC , Schneider MF, Cole SR, Cohen MH, Glesby MJ, Lazar J, et al.

Association of hepatitis C virus and HIV infection with subclinical atherosclerosis in

the women's interagency HIV study. AIDS 2009;23:1781-4.

27. Arcari CM , Nelson KE, Netski DM, Nieto FJ, Gaydos CA. No association between

hepatitis C virus seropositivity and acute myocardial infarction. Clin Infect Dis 2006

15;43:e53-6. 1-5.

28. Liao CC , Su TC, Sung FC, Chou WH, Chen TL. Does hepatitis C virus infection

increase risk for stroke? A population-based cohort study. PLoS One

2012;7:e31527.

29. Hsu CS , Kao JH, Chao YC, Lin HH, Fan YC, Huang CJ, et al. Interferon-based

therapy reduces risk of stroke in chronic hepatitis C patients: a population-based

cohort study in Taiwan. Aliment Pharmacol Ther 2013;38:415-23.

30. Butt AA , Xiaoqiang W, Budoff M, Leaf D, Kuller LH, Justice AC. Hepatitis C virus

infection and the risk of coronary disease. Clin Infect Dis 2009;49:225-32.

31. Forde KA , Haynes K, Troxel AB, Trooskin S, Osterman MT, Kimmel SE, et al.

Risk of myocardial infarction associated with chronic hepatitis C virus infection: a

population-based cohort study. J Viral Hepat 2012;19:271-7.

32. Adinolfi LE , Restivo L, Guerrera B, Sellitto A, Ciervo A, Iuliano N, et al. Chronic

HCV infection is a risk factor of ischemic stroke. Atherosclerosis. 2013

Nov;231(1):22-6.

33. Enger C , Forssen UM, Bennett D, Theodore D, Shantakumar S, McAfee A.

Thromboembolic events among patients with hepatitis C virus infection and

cirrhosis: a matched-cohort study. Adv Ther. 2014 Aug;31(8):891-903.

34. Hsu Y , Lin J, Ho H, Kao Y, Huang Y, Hsiao N, et al. Antiviral treatment for hepatitis

C virus infection is associated with improved renal and cardiovascular outcomes in

diabetic patients. Hepatology 59:1293-1302, 2014

35. Pothineni NV , Delongchamp R, Vallurupalli S, Ding Z, Dai Y, Hagedorn CH, Mehta

JL. Impact of hepatitis C seropositivity on the risk of coronary heart disease events.

Am J Cardiol. 2014 Dec 15;114(12):1841-5.

36. Butt AA , Khan UA, McGinnis KA, Skanderson M, Kent Kwoh C. Co-morbid medical

and psychiatric illness and substance abuse in HCV-infected and uninfected

veterans. J Viral Hepat. 2007 Dec;14(12):890-6.

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT37. Hsu YC , Ho HJ, Huang YT, Wang HH, Wu MS, Lin JT, Wu CY. Association

between antiviral treatment and extrahepatic outcomes in patients with hepatitis C

virus infection. Gut. 2015 Mar;64(3):495-503.

38. Vassalle C , Masini S, Bianchi F, Zucchelli GC. Evidence for association between

hepatitis C virus seropositivity and coronary artery disease. Heart 2004;90:565-6.

39. Alyan O , Kacmaz F, Ozdemir O, Deveci B, Astan R, Celebi AS, et al. Hepatitis C

infection is associated with increased coronary artery atherosclerosis defined by

modified Reardon severity score system. Circ J 2008;72:1960-5.

40. Yelken B , Gorgulu N, Caliskan Y, Elitok A, Cimen AO, Yazici H, et al. Association

between chronic hepatitis C infection and coronary flow reserve in dialysis patients

with failed renal allografts. Transplant Proc 2009;41:1519-23.

41. Momiyama Y , Ohmori R, Kato R, Taniguchi H, Nakamura H, Ohsuzu F. Lack of

any association between persistent hepatitis B or C virus infection and coronary

artery disease. Atherosclerosis 2005;181:211-3.

42. Younossi ZM , Stepanova M, Nader F, Younossi Z, Elsheikh E. Associations of

chronic hepatitis C with metabolic and cardiac outcomes. Aliment Pharmacol Ther

2013;37:647-52.

43. Hsu YH , Muo CH, Liu CY, Tsai WC, Hsu CC, Sung FC, Kao CH. Hepatitis C virus

infection increases the risk of developing peripheral arterial disease: a 9-year

population-based cohort study. J Hepatol. 2015 Mar;62(3):519-25.

44. Kakinami L , Block RC, Adams MJ, Cohn SE, Maliakkal B, Fisher SG. Risk of

cardiovascular disease in HIV, hepatitis C, or HIV/hepatitis C patients compared to

the general population. Int J Clin Pract 2013;67:6-13.

45. Lin MS , Guo SE, Chen MY, Huang TJ, Huang JC, Hu JH, Lin YS. The impact of

hepatitis C infection on ischemic heart disease via ischemic electrocardiogram. Am

J Med Sci. 2014 Jun;347(6):478-84.

46. EASL Recommendations on Treatment of Hepatitis C 2015. J Hepatol (2015),

http://dx.doi.org/10.1016/j.jhep.2015.03.025

47. Shamseer L , Moher D, Clarke M, Ghersi D, Liberati A, Petticrew M, Shekelle P,

Stewart LA; PRISMA-P Group. Preferred reporting items for systematic review and

meta-analysis protocols (PRISMA-P) 2015: elaboration and explanation. BMJ. 2015

Jan 2;349:g7647.

48. Wells GA SB , O’Connell D, Peterson J, Welch V (2012) The Newcastle-Ottawa

Scale (NOS) for assessing the quality of non-randomised studies in meta-analyses

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT49. DerSimonian R , Laird N. Meta-analysis in clinical trials. Control Clin Trials. 1986

Sep;7(3):177-88.

50. Review Manager, Version 5.2, The Nordic Cochrane Centre, The Cochrane

Collaboration, Copenhagen, Denmark

51. Breslow NE , Day NE. The Analysis of Case-Control Studies. Oxford, England:

IARC Scientific Publications; 1980. Davis W, ed. Statistical Methods in Cancer

Research; vol 1.

52. Thompson SG , Sharp SJ. Explaining heterogeneity in meta-analysis: a comparison

of methods. Stat Med. 1999 Oct 30;18(20):2693-708.

53. R Core Team 2014. R: A language and environment for statistical computing. R

Foundation for Statistical Computing, Vienna, Austria. URL http://www.R-

project.org/.

54. http://www.who.int/gho/tobacco/use/en/

55. http://www.who.int/mediacentre/factsheets/fs312/en/)

56. http://www.who.int/gho/ncd/risk_factors/blood_pressure_text/en/

57. Petta S , Macaluso FS, Craxì A. Cardiovascular diseases and HCV infection: a

simple association or more? Gut. 2014 Mar;63(3):369-75.

58. Ekstedt M , Hagström H, Nasr P, Fredrikson M, Stål P, Kechagias S, Hultcrantz R.

Fibrosis stage is the strongest predictor for disease-specific mortality in NAFLD

after up to 33 years of follow-up. Hepatology. 2015 May;61(5):1547-54.

59. Angulo P , Kleiner DE, Dam-Larsen S, Adams LA, Bjornsson ES, et al. Liver

Fibrosis, but no Other Histologic Features, Associates with Long-term Outcomes of

Patients With Nonalcoholic Fatty Liver Disease. Gastroenterology. 2015 Apr 29. pii:

S0016-5085(15)00599-5. doi: 10.1053/j.gastro.2015.04.043.

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPTFigure Legends

Figure 1. PRISMA flow diagram of the search process.

Figure 2. Meta-analysis of three studies that assessed the impact of HCV infection on CVD-

related mortality, using the random-effects model. OR and 95% CI are shown on a logarithm

scale. Studies are arranged by publication year. Study names are provided in the corresponding

references.

Figure 3. Meta-analysis of nine studies that assessed the impact of HCV infection on the

presence of carotid plaques, using the random-effects model. (A) Overall impact. (B) Impact

according to the prevalence of smoking habit in the population (<20% versus >20%). OR and 95%

CI are shown on a logarithm scale. Studies are arranged by publication year. Study names are

provided in the corresponding references.

Figure 4. Meta-analysis of eight studies that assessed the impact of HCV infection on cerebro-

cadiovascular events. (A) Overall impact. (B–E) Impact in subgroups according to prevalence of

diabetes (B), prevalence of hypertension (C), study design (D), and age (E). OR and 95% CI are

shown on a logarithm scale. Studies are arranged by publication year. Study names are provided in

the corresponding references.

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

Table 1. Patients and Characteristic of the Studies in the Meta-analysis

Author [ref] Year,

Country

Study

Design

Outcomes No. of

subjects

Mean

Age,

years

Male,

%

Caucasian,

%

Hyperte

nsion, %

Diabetes

, %

Mean

Cholesterol,

mg/dL

Hyperlipi

demia, %

Obesity,

%

Mean

BMI

Smokers,

% NOS

Score

Ishizaka [25]

2002, Japan

Case-

control CA

HCV+ 104 NR 74 NR NR NR NR NR NR NR NR 7

HCV- 4680 NR 65 NR NR NR NR NR NR NR NR

Ishizaka [19]

2003, Japan

Case-

control CA

IMT

HCV+ 25 61 68 NR NR NR 178 NR NR NR 56 7

HCV- 1967 57 65 NR NR NR 205 NR NR NR 52

Arcari [27]

2006, USA

Case-

control CAD

HCV+ 52 NR 0 61 NR NR NR NR NR NR NR 6

HCV- 530 NR 0 61 NR NR NR NR NR NR NR

Butt [36]

2007, USA

Case-

control CAD

Stroke

HCV+ 126971 51 97 48.5 NR 14 NR NR NR NR NR 5

HCV- 126971 52 97 46 NR 13 NR NR NR NR NR

Targher [20]

2007, Italy

Case-

control CA

IMT

HCV+ 60 46 57 NR NR NR NR NR NR 26 25 8

HCV- 60 46 57 NR NR NR NR NR NR 25 33

Bilora [24]

2008, Italy

Case-

control CA

IMT

HCV+ 40 57 55 NR 30 5 NR 5 NR NR 52.5 5

HCV- 40 57 50 NR 30 5 NR 5 NR NR 57.5

Caliskan [21]

2008, Turkey

Case-

control CA

IMT

HCV+ 36 46 36 NR NR NR 161 NR NR 22 39 6

HCV- 36 46 50 NR NR NR 189 NR NR 22 44

Tien [26]

2009, USA

Case-

control CA

IMT

HCV+ 53 49 0 NR NR 26 NR NR NR NR NR 6

HCV- 452 36 0 NR NR 6 NR NR NR NR NR

Mostafa [18]

2010, Egypy

Case-

control CA

IMT

HCV+ 187 50.8 63 NR NR 7 NR NR NR 27.7 14.2 9

HCV- 192 50.2 45 NR NR 4.6 NR NR NR 29.1 5.8

Adinolfi [22]

2012, Italy

Case-

control CA

IMT

HCV+ 326 54 51 NR 13 7 171 NR NR 26.5 42 8

HCV- 477 54 52 NR 17 9 205 NR NR 26.5 45

Petta [23]

2012, Italy

Case-

control CA

IMT

HCV+ 174 53 43 NR 29 7.5 178 NR 20 26 30.5 7

HCV- 174 53 43 NR 28 9 163 NR 17 26 29

Adinolfi [32]

2013, Italy

Case-

control Stroke HCV+ 79 73 51.5 NR 50 51.5 167 NR NR NR 39

7

HCV- 741 76 59 NR 59 48 193 NR NR NR 37

Guiltinan [5]

2008, USA

Cohort

study CV

mortality

HCV+ 10259 NR 65 48 NR NR NR NR NR NR NR 7

HCV- 10259 NR 65 52 NR NR NR NR NR NR NR

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

List of Abbreviations: BMI: Body Mass index; CA: Carotid Atherosclerosis; CAD: Coronary Artery Disease; CV: Cardiovascular; HCV: Hepatitis C Virus; IMT: Intima

Media Thickness; NOS: Newcastle-Ottawa Scale; NR: Not Reported.

Butt [30]

2009, USA

Cohort

study CAD

HCV+ 82083 51 97 55 42 21 175 39 NR NR NR

8 HCV- 89582 52 97 56 50 22 198 72 NR NR NR

Forde [31]

2012, UK

Cohort

study CAD HCV+ 4809 39 61 NR 10 5 NR 12 11.5 NR 26

7 HCV- 71668 39 61 NR 10 3 NR 13 14 NR 38

Lee [7]

2012, Taiwan

Cohort

study CV

mortality

HCV+ 1095 51 42.5 NR NR NR NR NR NR NR NR 8

HCV- 18541 47 49 NR NR NR NR NR NR NR NR

Liao [28]

2012, Taiwan

Cohort

study Stroke HCV+ 4094 NR 50 NR 42 26 NR 40 2 NR NR 8

HCV- 16376 NR 50 NR 35 14 NR 27 1 NR NR

Hsu [29]

2013, Taiwan

Cohort

study Stroke HCV+ 3113 NR 52 NR 8 9 NR 4 NR NR NR

8 HCV- 12452 NR 52 NR 8.5 4 NR 2 NR NR NR

Hsu [34]

2014, Taiwan

Cohort

study CAD

Stroke

HCV+ 1411 55 65.1 NR 42.9 100 NR 12.4 NR NR NR 8

HCV- 5644 55 65.2 NR 43.3 100 NR 12.7 NR NR NR

Enger [33]

2014, USA

Cohort

study CAD

Stroke

HCV+ 21919 49 62 44 NR NR NR NR NR NR NR 7

HCV- 67109 49 62 51 NR NR NR NR NR NR NR

Pothineni [35]

2014, USA

Cohort

study CAD

HCV+ 1434 49 57 77 31 16 156 NR 45 NR NR 7

HCV- 14799 53 54 60 12 5 185 NR 17 NR NR

Vajdic [8]

2015, Australia

Cohort

study CV

mortality

HCV+ 15523 23 69 NR NR NR NR NR NR NR NR 6

HCV- 14048 23 69 NR NR NR NR NR NR NR NR

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

Table 2. Meta-regression Analyses

Risk Factor Number of

Studies

Number of

Patients

β 95% C.I. P value

Cardiovascular Mortality and HCV Infection

Quality Score 3 69725 0.18 -0.31-0.67 0.46

Geographical source 3 69725 -0.07 -1.12-0.99 0.90

Carotid Plaques and HCV Infection

Quality Score 9 9083 -0.06 -0.36-0.23 0.67

Geographical source 9 9083 -0.34 -0.95-0.27 0.27

Age 7 3496 0.03 -0.03-0.10 0.32

Male Gender 8 4299 0.00 -0.02-0.01 0.61

BMI 5 1742 0.08 -0.15-0.30 0.50

Hypertension 3 1231 0.01 -0.03-0.06 0.50

Diabetes 5 2115 0.06 -0.21-0.33 0.64

Cholesterol 4 3215 0.03 -0.02-0.07 0.20

Smoking 7 3790 0.03 0.00-0.05 0.02

Carotid Intima-media Thickness and HCV Infection

Quality Score 7 3909 0.05 -0.01-0.10 0.09

Geographical source 7 3909 0.02 -0.19-0.23 0.87

Age 6 3106 -0.01 -0.03-0.01 0.52

Male Gender 7 3909 0.00 -0.01-0.00 0.60

BMI 5 1837 0.02 -0.02-0.06 0.42

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

Risk factors assessed in at least three studies

were tested

Hypertension 3 1231 -0.02 -0.01-0.01 0.43

Diabetes 4 1725 0.05 0.02-0.07 <0.001

Cholesterol 4 3215 0.00 -0.01-0.01 0.82

Smoking 7 3909 0.00 0.00-0.00 0.94

Cerebrocardiovascular Events and HCV Infection

Cohort Study Design 8 390602 0.59 0.10-1.08 0.01

Quality Score 8 390602 -0.12 -0.44-0.20 0.45

Geographical source 8 390602 0.33 -0.09-0.75 0.12

Age 5 354223 0.05 0.03-0.07 <0.001

Male Gender 8 390602 0.00 -0.01-0.01 0.42

Hypertension 7 390020 0.02 0.00-0.03 0.008

Diabetes 6 300992 0.03 0.01-0.05 <0.001

Hyperlipidaemia 4 283939 0.01 0.00-0.02 0.18

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

From: Moher D, Liberati A, Tetzlaff J, Altman DG, The PRISMA Group (2009). Preferred Reporting Items for Systematic Reviews and Meta-Analyses: The PRISMA Statement. PLoS Med 6(6): e1000097. doi:10.1371/journal.pmed1000097

For more information, visit www.prisma-statement.org.

Records identified through primary electronically search (PubMed

Central/Medline, Embase, Cochrane Library) (n = 1861)

Scre

enin

g In

clu

ded

El

igib

ility

Id

enti

fica

tio

n

Records after removal of duplicates (n = 1265)

Studies excluded after review of titles and abstracts, as being letters,

commentaries, or obviously irrelevant studies

(n = 1227)

Full-text articles assessed for eligibility

(n = 38)

Full-text articles excluded for irrelevant data or absence of considered outcomes: n= 8 Full-text articles excluded for presence of overlapping cohorts: n= 1 Full-text articles excluded for inability to extract number of subjects and/or number of events from cases and/or controls group: n= 7

Studies included in the meta-analysis

(n = 22)

Additional records identified by reference

list hand-search (n = 13)

Excluded because duplicates: 3 Excluded because obviously

irrelevant: 10

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPTMethods for Meta-analysis of studies assessing the impact of HCV infection on intima-media thickness . We assumed as secondary outcome the mean difference of intima media thickening between HCV positive and HCV negative subjects. The variance for each study was calculated by computing the pooled standard deviation of the intima media thickening between the HCV positive and negative groups. For the computation of pooled effects, each study was assigned a weight consisting of the reciprocal of its variance. Estimates of the pooled exposure effect and 95% confidence intervals were calculated with models based on the random effects assumptions. Heterogeneity was assessed with STATA software (STATA Corporation, College Station, Tex) [53].

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

Supplemental Table 1: Newcastle – Ottawa Quality Assessment Scale for Case Control Studies

Study, year

[ref]

Selection Comparability Outcome Score

Adequate

definition of

cases

Representativeness

of cases

Selection of

controls

Definition of

controls

Control for

important

factors

Ascertainment

of exposure

Same method

to ascertain for

cases and

controls

Non-

response

rate

Ishizaka, 2002

[25]

1 - 1 1 2 1 1 - 7

Ishizaka, 2003

[19]

1 - 1 1 2 1 1 - 7

Arcari, 2006

[27]

- 1 1 1 1 1 1 - 6

Butt, 2007 [36] - 1 1 1 1 - 1 - 5

Targher, 2007

[20]

1 1 1 1 2 1 1 - 8

Bilora, 2008

[24]

1 - - - 2 1 - 1 5

Caliskan, 2008

[21]

1 - - 1 2 1 1 - 6

Tien, 2009 [26] 1 1 - - 2 1 1 - 6

Mostafa, 2010

[18]

1 1 1 1 2 1 1 1 9

Adinolfi, 2012

[22]

1 1 1 1 2 1 1 - 8

Petta, 2012

[23]

1 1 - 1 2 1 1 - 7

Adinolfi, 2013

[32]

1 1 - 1 2 1 1 - 7

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

Supplemental Table 2: Newcastle – Ottawa Quality Assessment Scale for Cohort Studies

Study [ref] Selection Comparability Exposure Score

Representativeness

of the exposed

cohort

Selection of the

non exposed

cohort

Ascertainment

of exposure

Demonstration

that outcome

of interest was

not present at

start of study

Comparability

of cohorts on

the basis of the

design or

analysis

Assessment

of outcome

Was follow-

up long

enough for

outcomes to

occur

Adequacy

of follow-

up of

cohorts

Guiltinan,

2008 [5]

1 1 1 - 1 1 1 1 7

Butt, 2009

[30]

- 1 1 1 2 1 1 1 8

Forde, 2012

[31]

1 1 - 1 2 1 - 1 7

Lee, 2012 [7]

1 1 1 - 2 1 1 1 8

Liao, 2012

[28]

1 1 - 1 2 1 1 1 8

Hsu, 2013

[29]

1 1 - 1 2 1 1 1 8

Hsu, 2014

[34]

1 1 - 1 2 1 1 1 8

Enger, 2014

[33]

1 1 - 1 1 1 1 1 7

Pothineni

2014 [35]

1 1 1 1 2 1 - - 7

Vajdic 2015

[8]

- 1 1 - 1 1 1 1 6

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

Supplemental Figure 1. Meta-analysis of seven studies that assessed the impact of HCV

infection on IMT, using the random-effects model. OR and 95% CI are shown on a logarithm

scale. Studies are arranged by publication year. Study names are provided in the corresponding

references.

Supplemental Figure 2. Meta-analysis of studies that assessed the impact of HCV infection

on cardiovascular or cerebrovascular events, using the random-effects model. OR and 95%

CI for the effect of HCV infection on cardiovascular events (A) or cerebrovascular events (B) are

shown on a logarithm scale. Studies are arranged by publication year. Study names are provided

in the corresponding references.

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT