hepatitis c virus infection is associated with increased ...groups. in some studies, hcv infection...
TRANSCRIPT
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: [email protected];
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