Nitazoxanide for chronic hepatitis C (Review)
Nikolova K, Gluud C, Grevstad B, Jakobsen JC

This is a reprint of a Cochrane review, prepared and maintained by The Cochrane Collaboration and published in The Cochrane Library
2014, Issue 4
http://www.thecochranelibrary.com

Nitazoxanide for chronic hepatitis C (Review)
Copyright © 2014 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.

TABLE OF CONTENTS

HEADER . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
ABSTRACT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
PLAIN LANGUAGE SUMMARY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
SUMMARY OF FINDINGS FOR THE MAIN COMPARISON . . . . . . . . . . . . . . . . . . .
BACKGROUND . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
OBJECTIVES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
METHODS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
RESULTS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Figure 1.
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Figure 2.
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Figure 3.
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Figure 4.
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Figure 5.
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Figure 6.
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Figure 7.
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
ADDITIONAL SUMMARY OF FINDINGS . . . . . . . . . . . . . . . . . . . . . . . . . .
DISCUSSION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
AUTHORS’ CONCLUSIONS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
ACKNOWLEDGEMENTS
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
REFERENCES . . . . . . ; . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
CHARACTERISTICS OF STUDIES . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
DATA AND ANALYSES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Analysis 1.1. Comparison 1 Nitazoxanide versus placebo or no intervention, Outcome 1 All-cause mortality. . . .
Analysis 1.2. Comparison 1 Nitazoxanide versus placebo or no intervention, Outcome 2 Chronic hepatitis C-related
mortality. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Analysis 1.4. Comparison 1 Nitazoxanide versus placebo or no intervention, Outcome 4 Adverse events. . . . . .
Analysis 1.6. Comparison 1 Nitazoxanide versus placebo or no intervention, Outcome 6 Failure of sustained virological
response. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Analysis 1.7. Comparison 1 Nitazoxanide versus placebo or no intervention, Outcome 7 Failure of virological end-oftreatment response. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Analysis 1.8. Comparison 1 Nitazoxanide versus placebo or no intervention, Outcome 8 Participants without improvement
in ALT and/or AST serum levels. . . . . . . . . . . . . . . . . . . . . . . . . . . .
Analysis 2.1. Comparison 2 Subgroup: treatment-naives, relapsers and non-responders, Outcome 1 All-cause mortality.
Analysis 2.2. Comparison 2 Subgroup: treatment-naives, relapsers and non-responders, Outcome 2 Chronic hepatitis Crelated mortality. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Analysis 2.4. Comparison 2 Subgroup: treatment-naives, relapsers and non-responders, Outcome 4 Adverse events. .
Analysis 2.6. Comparison 2 Subgroup: treatment-naives, relapsers and non-responders, Outcome 6 Failure of sustained
virological response. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Analysis 2.7. Comparison 2 Subgroup: treatment-naives, relapsers and non-responders, Outcome 7 Failure of virological
end-of-treatment response. . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Analysis 2.8. Comparison 2 Subgroup: treatment-naives, relapsers and non-responders, Outcome 8 Participants without
improvement in ALT and/or AST serum levels. . . . . . . . . . . . . . . . . . . . . . .
Analysis 3.1. Comparison 3 Subgroup: genotype 1 compared to genotype 4, Outcome 1 All-cause mortality. . . .
Analysis 3.2. Comparison 3 Subgroup: genotype 1 compared to genotype 4, Outcome 2 Chronic hepatitis C-related
mortality. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Analysis 3.4. Comparison 3 Subgroup: genotype 1 compared to genotype 4, Outcome 4 Adverse events. . . . . .
Analysis 3.6. Comparison 3 Subgroup: genotype 1 compared to genotype 4, Outcome 6 Failure of sustained virological
response. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Analysis 3.7. Comparison 3 Subgroup: genotype 1 compared to genotype 4, Outcome 7 Failure of virological end-oftreatment response. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Nitazoxanide for chronic hepatitis C (Review)
Copyright © 2014 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.

1
1
2
4
6
7
7
11
12
15
16
18
20
22
24
28
37
40
40
40
47
68
72
73
74
75
76
77
78
79
80
81
82
83
84
84
85
86
87
i

Analysis 3.8. Comparison 3 Subgroup: genotype 1 compared to genotype 4, Outcome 8 Participants without improvement
in ALT and/or AST serum levels. . . . . . . . . . . . . . . . . . . . . . . . . . . .
Analysis 4.1. Comparison 4 Subgroup: nitazoxanide dose comparison, Outcome 1 All-cause mortality. . . . . .
Analysis 4.2. Comparison 4 Subgroup: nitazoxanide dose comparison, Outcome 2 Chronic hepatitis C-related mortality.
Analysis 4.4. Comparison 4 Subgroup: nitazoxanide dose comparison, Outcome 4 Adverse events. . . . . . . .
Analysis 4.6. Comparison 4 Subgroup: nitazoxanide dose comparison, Outcome 6 Failure of sustained virological
response. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Analysis 4.7. Comparison 4 Subgroup: nitazoxanide dose comparison, Outcome 7 Failure of virological end-of-treatment
response. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Analysis 4.8. Comparison 4 Subgroup: nitazoxanide dose comparison, Outcome 8 Participants without improvement in
ALT and/or AST serum levels. . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Analysis 5.1. Comparison 5 Best-worst case scenario analysis, Outcome 1 Failure of sustained virological response. .
Analysis 6.1. Comparison 6 Worst-best case scenario analysis, Outcome 1 Failure of sustained virological response. .
ADDITIONAL TABLES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
APPENDICES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
WHAT’S NEW . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
CONTRIBUTIONS OF AUTHORS . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
DECLARATIONS OF INTEREST . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
SOURCES OF SUPPORT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
DIFFERENCES BETWEEN PROTOCOL AND REVIEW . . . . . . . . . . . . . . . . . . . . .

Nitazoxanide for chronic hepatitis C (Review)
Copyright © 2014 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.

88
89
89
90
91
92
93
94
95
95
102
103
104
104
104
104

ii

[Intervention Review]

Nitazoxanide for chronic hepatitis C
Kristiana Nikolova1 , Christian Gluud1 , Berit Grevstad2 , Janus C Jakobsen1
1 The Cochrane Hepato-Biliary

Group, Copenhagen Trial Unit, Centre for Clinical Intervention Research, Department 7812, Rigshospitalet, Copenhagen University Hospital, Copenhagen, Denmark. 2 Copenhagen Trial Unit, Centre for Clinical Intervention Research,
Department 7812, Rigshospitalet, Copenhagen University Hospital, Copenhagen, Denmark
Contact address: Kristiana Nikolova, The Cochrane Hepato-Biliary Group, Copenhagen Trial Unit, Centre for Clinical Intervention
Research, Department 7812, Rigshospitalet, Copenhagen University Hospital, Blegdamsvej 9, Copenhagen, DK-2100, Denmark.
kristiana_nikolova@yahoo.com.
Editorial group: Cochrane Hepato-Biliary Group.
Publication status and date: Edited (no change to conclusions), published in Issue 4, 2014.
Review content assessed as up-to-date: 8 March 2014.
Citation: Nikolova K, Gluud C, Grevstad B, Jakobsen JC. Nitazoxanide for chronic hepatitis C. Cochrane Database of Systematic
Reviews 2014, Issue 4. Art. No.: CD009182. DOI: 10.1002/14651858.CD009182.pub2.
Copyright © 2014 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.

ABSTRACT
Background
Hepatitis C infection is a disease of the liver caused by the hepatitis C virus. The estimated number of chronically infected people
with hepatitis C virus worldwide is about 150 million people. Every year, another three to four million people acquire the infection.
Chronic hepatitis C is a leading cause of liver-related mortality and morbidity. It is estimated that around 5% to 20% of people with the
infection will develop liver cirrhosis, which increases the risk of hepatocellular carcinoma and liver failure. Until 2011, the combination
therapy of pegylated interferon-alpha (peginterferon) and ribavirin was the approved standard treatment for chronic hepatitis C. In
2011, first-generation direct-acting antivirals have been licensed, for use in combination with peginterferon and ribavirin for treating
hepatitis C virus genotype 1 infection. Nitazoxanide is another antiviral drug with broad antiviral activity and may have potential as
an effective alternative, or an addition to standard treatment for the treatment of the hepatitis C virus.
Objectives
To assess the benefits and harms of nitazoxanide in people with chronic hepatitis C virus infection.
Search methods
We searched The Cochrane Hepato-Biliary Group Controlled Trials Register (last searched April 2013), The Cochrane Central Register
of Controlled Trials (CENTRAL) (2013, Issue 3), MEDLINE (Ovid SP, 1948 to April 2013), EMBASE (Ovid SP, 1980 to April
2013), LILACS (1983 to April 2013), and Science Citation Index EXPANDED (ISI Web of Knowledge, 1900 to April 2013), using
the search strategies and the expected time spans. We also scanned reference lists of identified studies.
We also searched ClinicalTrials.gov and the World Health Organization’s International Clinical Trials Registry Platform search portal
for registered trials, either completed or ongoing (April 2013).
Selection criteria
We included randomised clinical trials that examined the effects of nitazoxanide versus placebo, no intervention, or any other intervention
in patients with chronic hepatitis C. We considered any co-intervention, including standard treatment, if delivered to all intervention
groups of the randomised trial concerned.
Nitazoxanide for chronic hepatitis C (Review)
Copyright © 2014 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.

1

Data collection and analysis
Two review authors extracted data independently. We assessed the risk of systematic errors (’bias’) by evaluation of bias risk domains.
We used Review Manager 5.2 for the statistical analyses of dichotomous outcome data with risk ratio (RR) and of continuous outcome
data with mean difference (MD). For meta-analyses, we used a fixed-effect model and a random-effects model, along with an assessment
of heterogeneity. We assessed risk of random errors (’play of chance’) using trial sequential analysis. We assessed the quality of the
evidence using the Grading of Recommendations Assessment, Development, and Evaluation (GRADE) system to present review results
in ’Summary of findings’ tables.
Main results
We included seven randomised clinical trials with a total of 538 participants with chronic hepatitis C. Participants were 18 years of
age or older, all diagnosed with chronic hepatitis C genotype 1 or 4. All of the trials had a high risk of bias. All of the trials compared
nitazoxanide with placebo or no intervention, and six out of seven of the trials included different antiviral co-interventions administered
equally to all intervention groups. Only one trial, comparing nitazoxanide plus peginterferon and ribavirin versus no intervention
plus peginterferon and ribavirin, provided information that there were no deaths due to any cause or due to chronic hepatitis C (100
participants, very low quality evidence). The relative effect of nitazoxanide versus placebo or no intervention on adverse events was
uncertain (37 out of 179 (21%) versus 30 out of 152 (20%); RR 1.10; 95% CI 0.71 to 1.71; I2 = 65%; four trials; very low quality
evidence). Nitazoxanide decreased the risk of failure to achieve sustained virological response when compared with placebo or no
intervention (159 out of 290 (55%) versus 133 out of 208 (64%); RR 0.85; 95% CI 0.75 to 0.97; I2 = 0%; seven trials; low quality
evidence) and also the risk of failure to achieve virological end-of-treatment response (125 out of 290 (43%) versus 110 out of 208
(53%); RR 0.81; 95% CI 0.69 to 0.96; I2 = 46%; seven trials; low quality evidence). Trial sequential analysis supported the metaanalysis result for sustained virological response, but not the meta-analysis for virological end-of-treatment response. Meta-analysis
also showed that nitazoxanide did not decrease the number of participants who showed no improvement in alanine aminotransferase
and aspartate aminotransferase serum levels when compared with placebo or no intervention (52 out of 97 (54%) versus 47 out of
95 (49%); RR 1.09; 95% CI 0.84 to 1.42; I2 = 0%; three trials; very low quality evidence). None of the included trials assessed the
effects of nitazoxanide on morbidity or on quality of life. Histological changes were only reported on a subset of three participants out
of thirteen participants included in a long term-follow-up trial.
Authors’ conclusions
We found very low quality, or no, evidence on nitazoxanide for clinically- or patient-relevant outcomes, such as all-cause mortality,
chronic hepatitis C-related mortality, morbidity, and adverse events in participants with chronic hepatitis C genotype 1 or 4 infection.
Our results of no improvement in alanine aminotransferase and aspartate aminotransferase serum levels were also uncertain. No
conclusion could be drawn about liver histology because of a lack of data. Our results indicate that nitazoxanide might have an effect
on sustained virological response and virological end-of-treatment response. However, both results could be influenced by systematic
errors because all the trials included in the review had a high risk of bias. Furthermore, only the beneficial effect on number of
participants achieving sustained virological response was supported when we applied trial sequential analysis. The results on virological
end-of-treatment response might, therefore, be caused by a random error. We totally lack information on the effects of nitazoxanide in
participants with chronic hepatitis C genotypes 2 or 3 infection. More randomised clinical trials with a low risk of bias are needed to
assess the effects of nitazoxanide for chronic hepatitis C.

PLAIN LANGUAGE SUMMARY
Nitazoxanide for chronic hepatitis C
What is chronic hepatitis C, and why is it a problem?
Hepatitis C is a virus that infects people’s liver. When an infection goes on for a long time, it is said to be ‘chronic’.
Hepatitis C is mainly transmitted between people through contact with infected blood (mostly through illegal drug use that involves
needle-sharing, but possibly from mother to baby in the womb, or through having sex with an infected person).
Usually people infected with hepatitis C have no symptoms in the early stages of the infection; however, about 60% to 70% of them
go on to develop severe liver-related problems which eventually lead to death. There are about 150 million people in the world with
chronic hepatitis C infection and more than 350,000 of them die from this liver infection each year.
Nitazoxanide for chronic hepatitis C (Review)
Copyright © 2014 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.

2

What are the treatments for chronic hepatitis C infection?
The current standard treatment for chronic hepatitis C is a combination of two medicines, pegylated interferon-alpha (interferon)
and ribavirin. Interferon is not widely available globally and can have some adverse (harmful) effects. It works better on some types
(genotypes) of the hepatitis C virus than others.
Nitazoxanide belongs to a class of antiviral medicines with activity against a broad range of viruses. It is the first medicine in this class
to be investigated for its effect on chronic hepatitis C infection.
The purpose of this review
This Cochrane review tried to identify the benefits and harms of treating chronic hepatitis C infection with nitazoxanide.
Findings of this review
The review authors searched the medical literature up to April 2013, and identified seven relevant medical trials, with a total of 538
participants. The trials were performed in two countries, the USA and Egypt. All the trials had low numbers of participants, and the
methods used in them were not of a high quality, which makes potential overestimation of benefits and underestimation of harms more
likely. All the trials only included participants with chronic hepatitis C genotype (type) 1 or 4 infection.
Outcomes important to people who suffer from this infection include: death from all causes, death from chronic hepatitis C infection,
how unwell you feel (morbidity), quality of life, and adverse events caused by the medicines. This review found no information, or
very little low quality evidence, about the effects of nitazoxanide on any of these outcomes.
Nitazoxanide might have a beneficial effect on virus activity that can be monitored by analysis of blood samples (sustained virological
response (SVR) and virological end-of-treatment response (ETR), but this is not certain. Indeed, the review authors could not draw
any conclusions about the benefits or harms of nitazoxanide for people with chronic hepatitis C infection.
More randomised clinical trials of high methodological quality are needed to establish the effects of nitazoxanide in people with chronic
hepatitis C.

Nitazoxanide for chronic hepatitis C (Review)
Copyright © 2014 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.

3

Nitazoxanide for chronic hepatitis C (Review)
Copyright © 2014 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.

S U M M A R Y O F F I N D I N G S F O R T H E M A I N C O M P A R I S O N [Explanation]

Nitazoxanide compared with placebo or no intervention for chronic hepatitis C (Overall ’Summary of findings’ table)
Patient or population: people with chronic hepatitis C
Settings: primary care in Egypt and USA
Intervention: nitazoxanide
Comparison: placebo or no intervention
Outcomes

Illustrative comparative risks* (95% CI)

Assumed risk

Relative effect
(95% CI)

No of participants
(studies)

Quality of the evidence
(GRADE)

Comments

Corresponding risk

Placebo or no interven- Nitazoxanide
tion
No deaths

No deaths

Not estimable

100
(1)

1,2,4,5
⊕
very low

-

Chronic hepatitis C-re- No deaths
lated mortality

No deaths

Not estimable

100
(1)

1,2,4,5
⊕
very low

-

Morbidity

No reports on participants developing morbidity in any trials making
this comparison

No reports on partici- Not estimable
pants developing morbidity in any trials making
this comparison

0

-

-

Adverse events

197 per 1000

217 per 1000

RR 1.10 (0.71 to 1.71)

383
(4)

1,2,3,4
⊕
very low

-

Failure of sustained vi- 639 per 1000
rological response

544 per 1000

RR 0.85 (0.75 to 0.97)

498
(7)

⊕⊕
low

1,2,4

-

Failure of virologi- 529 per 1000
cal end-of-treatment response

428 per 1000

RR 0.81 (0.69 to 0.96)

498
(7)

⊕⊕
low

1,2,4

-

All-cause mortality

4

Nitazoxanide for chronic hepatitis C (Review)
Copyright © 2014 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.

Participants without im- 495 per 1000
provement in ALT and/or
AST serum levels

539 per 1000

RR 1.09 (0.84 to 1.42)

271
(3)

1,2,3,4
⊕
very low

-

*The basis for the assumed risk (e.g. the median control group risk across studies) is provided in footnotes. The corresponding risk (and its 95% confidence interval) is based on the
assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI).
CI: confidence interval; RR: risk ratio; ALT: alanine aminotransferase; AST: aspartate aminotransferase
GRADE Working Group grades of evidence
High quality: further research is very unlikely to change our confidence in the estimate of effect
Moderate quality: further research is likely to have an important impact on our confidence in the estimate of effect and may change the estimate
Low quality: further research is very likely to have an important impact on our confidence in the estimate of effect and is likely to change the estimate
Very low quality: we are very uncertain about the estimate
1 The

trial(s) was (were) of high risk of bias
number of events in the intervention and control group was less than 300
3
The confidence intervals overlapped 1, and either 0.75 or 1.25, or both
4 The assumed risk is the control group risk
5 Only one trial reported on mortality
2 The

5

BACKGROUND

Description of the condition
Hepatitis C is a disease of the liver caused by the hepatitis C virus.
Hepatitis C virus is an enveloped RNA virus that constitutes the
genus Hepacivirus within the Flaviviridae family (van Regenmortel
2000; Penin 2004). The hepatitis C virus is classified into 11 major virus genotypes, with six of the genotypes occurring more frequently (Foster 2009). The hepatitis C virus genotypes differ from
each other by up to 30% in the nucleotide sequence and there is a
large and growing number of subtypes (Rosenberg 2001). Furthermore, hepatitis C virus genotypes differ according to geographic
region (Davis 1999). Genotype 1, followed by genotypes 2 and
3 are most common in the USA, while genotype 4 is more common in Africa and the Middle East, including Egypt. However,
in Europe, as a result of an increase in the number of people who
inject drugs, there has been an increase in infections of hepatitis C
virus genotype 4. Genotypes 5 and 6 are common in South Africa
and Southeast Africa (McKenzie 2011). Although a genotype does
not predict the clinical outcome of the infection, it does predict
the likelihood of viral response to treatment, and this may guide
the duration of treatment (Manns 2001; Fried 2002; Hadziyannis
2004).
The estimated number of people around the world who are chronically infected with hepatitis C virus is about 150 million (WHO
2011). Every year, another three to four million people acquire
the infection (WHO 2010).
Chronic hepatitis C virus infection is diagnosed when hepatitis C
virus antibodies are present in the blood, and serum aminotransferase levels are high and remain elevated for more than six months,
as spontaneous viral clearance by the body is very rare after four to
six months of infection (McKenzie 2011). Polymerase chain reaction (PCR) testing for hepatitis C virus RNA is used to establish
the presence of chronic hepatitis C in the blood. The PCR will
detect the presence of the viral genome in serum in almost all people with chronic hepatitis C infection (Ghany 2009). Diagnosis of
chronic hepatitis C may be difficult in patients who have a solidorgan transplant, are on dialysis, are taking corticosteroids, or have
agammaglobulinaemia (genetic condition that affects the body’s
ability to fight infection), and this is why it is recommended that
RNA testing is used in these patients. RNA testing should also be
used for patients with anti-hepatitis C virus who have liver disease
caused injury to the liver, i.e., due to alcohol, iron overload, or
autoimmunity (Ghany 2009). A study performed by Amon et al
in the USA has shown that the prevalence of hepatitis C virus in
injecting drug users is very high (Amon 2008). Amon et al found
that one in three people who abuse drugs become infected with
hepatitis C virus within five years (Amon 2008).
About 85% of infected people become chronic carriers of the
hepatitis C virus (Seeff 2002). These people are relatively young
(McKenzie 2011). Carriers of the virus are often unaware that they

are infected, as many remain asymptomatic. In many patients,
hepatitis C infection is only recognised when it reaches a chronic
symptomatic phase (Hodgson 2003). The interval between infection with the virus and the symptoms of cirrhosis may exceed 30
years in people with favourable risk factors. However, 20% of the
chronic hepatitis C virus carriers may develop cirrhosis within 20
years, or less, of being infected, and this happens most often to
people who abuse alcohol or are co-infected with human immunodeficiency virus (HIV) type 1 or hepatitis B virus (Lauer 2001).
Between 1% to 4% of patients with advanced fibrosis or cirrhosis
are at risk of developing hepatocellular carcinoma each year (Lauer
2001). Chronic hepatitis C virus infection is a common indication
for liver transplantation (OPTN 2005).
There are some diseases that may be confused with chronic hepatitis C; these include alcoholism, hepatitis B, hepatitis D, nonalcoholic steatohepatitis, and others (Ghany 2009).
Each year, around 350,000 people are likely to die from hepatitis
C-related liver diseases worldwide (WHO 2011).

Description of the intervention
Thiazolides are a class of antiviral drugs with activity against bacteria and a broad range of DNA and RNA viruses (Rossignol 2001).
Nitazoxanide is a thiazolide with activity against anaerobic bacteria and also protozoa (Pankuch 2006). Nitazoxanide was originally developed for the treatment of infectious diarrhoea caused
by Cryptosporidium parvum andGiardia lamblia (Pankuch 2006).
The antiviral activity of nitazoxanide was discovered by chance
in patients with acquired immunodeficiency syndrome (AIDS)
who were treated for cryptosporidial diarrhoea and had co-infection with the hepatitis B or hepatitis C virus (Rossignol 2001).
Nitazoxanide is the first thiazolide undergoing clinical development for treatment of chronic hepatitis C, and new generation
thiazolides are proposed for treating chronic hepatitis B and C,
enteric viruses, herpes viruses, and influenza (Rossignol 2009b).
Currently, nitazoxanide is licensed as Alinia® (Romark Laboratories, USA) in the USA, in the form of tablets and oral suspension,
and it is administered for diarrhoea and enteritis caused by Cryptosporidium parvum and Giardia lamblia to adults and children
from 12 months of age. Nitazoxanide may provide an alternative,
more effective treatment option than the standard treatment for
chronic hepatitis C (Rossignol 2009b).

How the intervention might work
When treating chronic hepatitis C virus infection, the goal is to
prevent complications and death caused by the infection (Ghany
2009). Frequently, the sustained virological response is the outcome used to monitor hepatitis C virus treatment (Strader 2004).
Sustained virological response is defined as the “absence of hepatitis
C virus RNA from serum by a sensitive PCR-based assay 24 weeks

Nitazoxanide for chronic hepatitis C (Review)
Copyright © 2014 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.

6

following discontinuation of therapy” (Ghany 2009). If hepatitis
C virus cannot be detected at the end of a 24-week or 48-week
course of therapy, it is referred to as an end-of-treatment response.
Virological end-of-treatment response and sustained virological
response are used as a putative surrogate outcomes for clinicallyrelevant outcome measures such as mortality, morbidity, quality
of life, and adverse reactions, but they have not been validated
in this role (Gluud 2007; Brok 2009; Brok 2010; Koretz 2013).
Existing interventions, such as ribavirin alone or in combination
with interferon, are disappointing as the proportion of patients
with sustained virological response is small, and the percentage of
adverse events such as influenza-like symptoms, depression, neutropenia, thrombocytopenia are high (Strader 2004; Brok 2009;
Brok 2010). A combination of weekly subcutaneous injections of
long-acting pegylated interferon-alpha and oral ribavirin to previously untreated patients has shown an overall sustained virological
response of 56% (Strader 2004; Brok 2010). This compares with
interferon monotherapy which has a sustained virological response
in less than 20% of chronic hepatitis C patients (Myers 2002),
and with the combination of interferon plus ribavirin, which has
a sustained virological response in about 40% of chronic hepatitis
C patients (Brok 2010).
As more and more treatments for chronic hepatitis C become
available, the possibility of moving away from the current standard treatment and its limitations becomes increasingly likely
(Aman 2012). In 2011, two first-generation direct-acting antivirals (DAAs), telaprevir and boceprevir, were licensed for use in
treatment of hepatitis C virus genotype 1 infection, and in 2013
two new drugs, sofosbuvir (Sovaldi®) and simeprevir (Olysio®)
were approved as combination therapies with ribavirin (Kowdley
2013; Lawitz 2013). Studies have indicated that sofosbuvir and
simeprevir have effects on sustained virological responses (FDA
2014). Although these DAAs might achieve high sustained virological response, all of the trials conducted had sustained virological response as their primary outcome, and very few of them assessed clinically relevant outcomes, such as mortality, morbidity,
or quality of life.
Nitazoxanide may have potential among the many antiviral drugs
that have become available (Stockis 2002; Korba 2008a; Korba
2008b). Stockis et al studied healthy participants and found that
a 500 mg tablet of nitazoxanide, administered orally with food,
is partially absorbed from the gastrointestinal tract and rapidly
hydrolyzed in the plasma, forming the active circulating metabolite tizoxanide (desacetyl-nitazoxanide) (Stockis 2002). Nitazoxanide is not detected in the plasma. Both nitazoxanide and tizoxanide are believed to inhibit replication of the hepatitis C virus
(Korba 2008a). The maximum serum concentrations of tizoxanide
recorded by Stockis et al reached approximately 10 g/mL (37 M).
Tizoxanide is glucurono-conjugated in the liver and excreted in
urine and bile (Stockis 2002). Korba et al suggested that the inhibitory effect of nitazoxanide on viral replication of hepatitis C
may be due to cellular processes required for virus protein produc-

tion and assembly (Korba 2008a). Korba et al found that nitazoxanide induced changes in the intracellular environment that may
lead to a change in the effect of subsequent treatment with other
anti-hepatitis C virus drugs, particularly interferon alpha (Korba
2008b).
The genotype of the hepatitis C virus is of importance as it might
predict viral response to treatment. The new approved antiviral
drugs all target genotype 1 which is the most prevalent genotype
worldwide. So far, only sofosbuvir shows activity against genotypes
1, 2, 3, and 4 and does not work in all patients. Therefore, there
is a need to identify new antivirals that are effective against other
genotypes.

Why it is important to do this review
Several randomised clinical trials investigating the efficacy and adverse events of nitazoxanide for the treatment of chronic hepatitis
C have been finalised and published. However, a systematic review
assessing the benefits and harms of nitazoxanide in the treatment
of patients with chronic hepatitis C has not been conducted. We
have not been able to identify any meta-analyses on the benefits
and harms of nitazoxanide, administered alone or in combination with other drugs, for chronic hepatitis C virus infection. The
results of our Cochrane systematic review may have an impact
on treatment of people with chronic hepatitis C virus infection,
healthcare costs, and may also be used to decide future research
topics.

OBJECTIVES
To assess the benefits and harms of nitazoxanide in people with
chronic hepatitis C virus infection.

METHODS

Criteria for considering studies for this review

Types of studies
We included randomised clinical trials that examined the effects
of nitazoxanide versus placebo, no intervention, or any other intervention for chronic hepatitis C virus infection. Trials were included irrespective of language of publication, publication type,
or publication status. We also reported the information on harms
from quasi-randomised studies or observational studies retrieved
by our literature search for randomised clinical trials.

Nitazoxanide for chronic hepatitis C (Review)
Copyright © 2014 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.

7

Types of participants
We included trial participants diagnosed with chronic hepatitis C
according to any of the following criteria:
1. anti-hepatitis C virus antibodies present in the serum and
serum aminotransferase activity elevated for more than six
months (Ghany 2009);
2. evidence of inflammation due to chronic hepatitis C on
liver biopsy (Strader 2004; Ghany 2009);
3. suspicion of hepatitis C virus confirmed by quantitative
analysis of hepatitis C virus RNA through PCR or transcriptionmediated amplification testing (Ghany 2009; NIH 2010); or
4. by any other method used for diagnosis of people with
chronic hepatitis C.
Participants included in the trials were classified as treatment-naive
(not previously treated with antiviral drugs), relapsers (those with
a transient response to previous antiviral treatment), or non-responders (those who did not respond to previous antiviral treatment).
We also included trial participants with a concomitant or preceding disease, such as HIV, solid organ transplant recipients, those
receiving haemodialysis, or with another form of liver disease that
might be responsible for the liver problem, for example hepatitis
B infection, alcoholism, or iron overload.

or any medical event that might have jeopardised the patient, or
required intervention to prevent it (ICH-GCP 1997). All other
adverse events (that is, any medical occurrence not necessarily
having a causal relationship with the treatment but that did,
however, cause a dose reduction or discontinuation of the
treatment) were considered as being non-serious.
Secondary outcomes

1. Quality of life (any valid continuous outcome scale used by
the trialists).
2. Failure of virological response: number of participants
without sustained virological response: number of participants
with detectable hepatitis C virus RNA (i.e., above lower limit of
detection) in the serum by a sensitive PCR-based essay or by a
transcription-mediated amplification testing 24 weeks after end
of treatment, or at end of treatment (lack of an early virological
response can be used to predict non-responders and a lack of
sustained virological response (Fried 2011)).
3. Number of participants without improvement in alanine
aminotransferase and aspartate aminotransferase serum levels,
(aspartate aminotransferase more than 40 IU/L) from baseline to
weeks 8, 16, end of treatment, and end of maximum follow-up.
4. Number of participants without histological improvement.

Types of interventions
We included all randomised clinical trials examining the effects of
nitazoxanide for chronic hepatitis C comparing:
1. nitazoxanide versus placebo or no intervention;
2. nitazoxanide in combination with any other intervention
for chronic hepatitis C versus placebo or no intervention;
3. nitazoxanide versus any other active intervention.
All co-interventions, including any form of standard treatment,
were allowed if administered equally to all intervention groups.
Types of outcome measures

Primary outcomes

1. All-cause mortality.
2. Chronic hepatitis C-related mortality (as defined by the
trialists).
3. Morbidity: number of participants who developed cirrhosis,
ascites, variceal bleeding, hepatic encephalopathy, or
hepatocellular carcinoma.
4. Adverse events: serious adverse events were defined
according to the International Conference on Harmonisation
(ICH) Guidelines for Good Clinical Practice as any untoward
medical occurrence that at any dose resulted in death, was lifethreatening, required inpatient hospitalisation or prolongation of
existing hospitalisation, or resulted in persistent or significant
disability or incapacity, or was a congenital anomaly/birth defect,

Search methods for identification of studies

Electronic searches
We searched The Cochrane Hepato-Biliary Group Controlled
Trials Register (last searched April 2013) (Gluud 2013), The
Cochrane Central Register of Controlled Trials (CENTRAL)
(2013, Issue 3), MEDLINE (Ovid SP, 1948 to April 2013), EMBASE (Ovid SP, 1980 to April 2013) , LILACS (1983 to April
2013), and Science Citation Index EXPANDED (ISI Web of
Knowledge, 1900 to April 2013) (Royle 2003), using the search
strategies and time spans presented in Appendix 1.
We also searched Clinical trials.gov at www.clinicaltrials.gov and
the World Health Organisation’s International Clinical Trials Registry Platform Search portal (ICTRP) (WHO) at apps.who.int/
trialsearch/Default.aspx for registered trials, either completed or
ongoing.
Searching other resources
We searched bibliographic references, conference abstracts, journals, and grey literature for further trials. We searched for publications from 1980 onward, as testing of nitazoxanide started after
1980 (Reddy 2012). Furthermore, we reviewed the reference lists
and contacted the principal authors of identified trials. We also
contacted pharmaceutical companies that produce nitazoxanide

Nitazoxanide for chronic hepatitis C (Review)
Copyright © 2014 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.

8

to inquire about information and request data relating to unpublished randomised clinical trials.

Data collection and analysis
We performed the systematic review and meta-analyses following
the recommendations of The Cochrane Collaboration (Higgins
2011), and The Cochrane Hepato-Biliary Group (Gluud 2013).
In the case of cross-over trials, we had planned to include data only
from the first period (Higgins 2011). The analyses were performed
using Review Manager 5.2 (RevMan 2012), and TSA version 0.9
(Wetterslev 2008; CTU 2011; Thorlund 2011). We calculated the
risk ratio (RR) or mean difference (MD) with 95% confidence intervals (CI) using fixed-effect and random-effects models of metaanalyses based on intention-to-treat or available data analysis. We
reported results from both meta-analysis models, discussing any
discrepancies.

Selection of studies
Two of the authors, KN and BG, independently identified the
trials for inclusion in accordance with the inclusion criteria of the
review protocol and listed the excluded studies with the reasons
for their exclusion. KN and BG resolved disagreements through
discussions. CG was the arbitrator in case of disagreements.

Data extraction and management
KN and BG used a standardised data collection form to extract
data regarding source, eligibility, methods, participants, genotype,
disease, interventions, outcomes, and results of trials. Data were
collected from either the published reports or were obtained directly from authors of the publications by email.
KN and BG extracted all data independently. KN and BG resolved
disagreements through discussions. CG was the arbitrator in case
of disagreements.

Assessment of risk of bias in included studies
Methodological quality, hence risk of bias, was defined as confidence that the design and report of the randomised clinical trials restrict bias in the comparison of the intervention (Moher
1998). We assessed the following risk of bias domains using the
revised Cochrane Collaboration’s ’Risk of bias’ tool provided in
theCochrane Handbook for Systematic Reviews of Interventions as
well as in the Cochrane Hepato-Biliary Group Module. The
domains are based on empirical evidence (Schulz 1995; Moher
1998; Kjaergard 2001; Wood 2008; Lundh 2012; Savovi 2012a;
Savovi 2012b).

Allocation sequence generation

• Low risk of bias: sequence generation was achieved using
computer random number generation or a random number
table. Drawing lots, tossing a coin, shuffling cards, and throwing
dice were adequate if performed by an independent person not
otherwise involved in the trial.
• Uncertain risk of bias: the method of sequence generation
was not specified.
• High risk of bias: the sequence generation method was not
random.
Allocation concealment

• Low risk of bias: the participant allocations could not have
been foreseen in advance of, or during, enrolment. Allocation
was controlled by a central and independent randomisation unit.
The allocation sequence was unknown to the investigators (for
example, if the allocation sequence was hidden in sequentiallynumbered, opaque, and sealed envelopes).
• Uncertain risk of bias: the method used to conceal the
allocation was not described so that intervention allocations may
have been foreseen in advance of, or during, enrolment.
• High risk of bias: the allocation sequence was likely to be
known to the investigators who assigned the participants.
Blinding of participants, personnel, and outcome assessors

• Low risk of bias: blinding was performed adequately, or the
assessment of outcomes was not likely to be influenced by lack of
blinding.
• Uncertain risk of bias: there was insufficient information to
assess whether blinding was likely to induce bias on the results.
• High risk of bias: no blinding or incomplete blinding, and
the assessment of outcomes were likely to be influenced by lack
of blinding.
Incomplete outcome data

• Low risk of bias: missing data were unlikely to make
treatment effects depart from plausible values. Sufficient
methods, such as multiple imputation, were employed to handle
missing data.
• Uncertain risk of bias: there was insufficient information to
assess whether missing data in combination with the method
used to handle missing data were likely to induce bias in the
results.
• High risk of bias: the results were likely to be biased due to
missing data.
Selective outcome reporting

• Low risk of bias: all outcomes were pre-defined and
reported, or all clinically relevant and reasonably expected

Nitazoxanide for chronic hepatitis C (Review)
Copyright © 2014 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.

9

outcomes were reported. The trial was registered either on the
www.clinicaltrials.gov web site or a similar register, or there was a
published protocol.
• Uncertain risk of bias: it was unclear whether all predefined and clinically relevant and reasonably expected outcomes
were reported.
• High risk of bias: one or more clinically relevant and
reasonably expected outcomes were not reported, and data on
these outcomes were likely to have been recorded.

hepatitis C-related mortality, and morbidity (Hollis 1999; Gluud
2013):
1. ’Best-worst’ case scenario analyses: participants with
missing outcome data were considered to be successes in the
experimental group and failures in the control group. The
denominator included all the participants in the trial.
2. ’Worst-best’ case scenario analyses: participants with
missing outcome data were considered to be failures in the
experimental group and successes in the control group. The
denominator included all the participants in the trial.

Other bias

• Low risk of bias: the trial appeared to be free of other
components that could put it at risk of bias.
• Uncertain risk of bias: it was uncertain whether the trial was
free of other components that could put it at risk of bias.
• High risk of bias: there were other factors in the trial that
could put it at risk of bias, e.g., for-profit involvement, authors
have conducted trials on the same topic, etc.
We assessed trials as being at low risk of bias if they were judged to
have a low risk of bias in all the above domains. We assessed a trial
as being at high risk of bias if it was judged to have an uncertain
or high risk of bias in any of the above domains.
KN and BG resolved disagreements through discussions. CG was
the arbitrator in case of disagreements.
Measures of treatment effect
For dichotomous outcomes we calculated the risk ratio (RR), and
for continuous outcomes we used the mean difference (MD), both
with a 95% confidence interval (CI).

Assessment of heterogeneity
We attempted to assess heterogeneity in three ways (Higgins
2002): graphically, by using forest plots; by the Chi2 test where P
values of less than 0.10 determine statistical significance; and by
the I2 statistic. Roughly, we read the I2 test value in the following way: from 0% to 40%, heterogeneity may not be important;
from 30% to 60%, heterogeneity may be moderate; from 50% to
90%, heterogeneity may be substantial; and from 75% to 100%,
heterogeneity may be considerable.

Assessment of reporting biases
For any type of outcome where there were at least ten trials included
in the meta-analysis, we planned to test for funnel plot asymmetry
(Higgins 2011). Nevertheless, asymmetric funnel plots are not
necessarily caused by publication bias, and publication bias does
not necessarily lead to asymmetry in a funnel plot (Egger 1997).

Data synthesis
Dealing with missing data
KN contacted investigators of the trials to request missing data.
We performed our analyses based on the intention-to-treat principle, using imputation of dichotomous data for the outcomes:
1. If data were missing regarding all-cause mortality and
chronic hepatitis C-related mortality, we assumed that the
participants lost to follow-up had survived. Therefore, the
denominator included the number of randomised participants in
the respective trial.
2. If data were missing regarding morbidity, adverse events,
failure of sustained virological response, failure of virological
end-of-treatment response, participants without improvement in
alanine aminotransferase or aspartate aminotransferase serum
levels, or number of participants without histological
improvement, we assumed that participants lost to follow-up had
such an event. Therefore, we included all the participants
randomised in the trial in the denominator.
We had planned to perform sensitivity analyses using two extreme
scenarios for participants with incomplete or missing data for
the first three primary outcomes, i.e., all-cause mortality, chronic

For the statistical analyses, we used Review Manager 5.2 (RevMan
2012). For meta-analyses with more than one trial we used both
a fixed-effect model (DeMets 1987) and a random-effects model
(DerSimonian 1986), along with an assessment of heterogeneity.
We presented the results with both meta-analytic methods. If there
were no differences in the results, we presented the results with
the fixed-effect model only.

Trial sequential analysis

We applied trial sequential analysis to minimise the risks of random error (Brok 2008; Wetterslev 2008; Brok 2009; Thorlund
2009, Wetterslev 2009; Thorlund 2010; CTU 2011; Thorlund
2011). Cumulative meta-analyses are at risk of producing random
errors due to sparse data and repetitive testing on the accumulating
data (Brok 2008; Wetterslev 2008). To minimise random errors,
we calculated the required information size, i.e., the number of
participants needed in a meta-analysis to detect or reject a certain
intervention effect (Wetterslev 2008). The information size calculation should also account for the diversity present in the meta-

Nitazoxanide for chronic hepatitis C (Review)
Copyright © 2014 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.

10

analysis (Wetterslev 2009). In our meta-analysis, the required information size was based on the assumption of a plausible RR reduction of 20% or on the RR reduction observed in the included
trials with low risk of bias (Wetterslev 2008).
The underlying assumption of trial sequential analysis is that testing for significance may be performed each time a new trial is
added to the meta-analysis. We added the trials according to the
year of publication, and if more than one trial was published in a
year, trials were added alphabetically according to the last name of
the first author (Wetterslev 2008).
On the basis of the required information size and risk for type I
(5%) and type II (20%) errors, trial sequential monitoring boundaries can be constructed (Wetterslev 2008). These boundaries determine the statistical inference one may draw regarding the cumulative meta-analysis that has not reached the required information size. If the trial sequential monitoring boundary is crossed
before the required information size is reached, firm evidence may
be established and further trials may turn out to be superfluous.
On the other hand, if the boundary is not surpassed, it is probably
necessary to perform more trials in order to detect or reject a certain intervention effect. As default, type I error of 5%, type II error of 20%, and diversity-adjusted required information size were
used unless otherwise stated (Wetterslev 2008; Wetterslev 2009).

’Summary of findings’ table
We assessed the quality of the evidence using the Grading of
Recommendations Assessment, Development, and Evaluation
(GRADE) system to present review results in ’Summary of
findings’ (SoF) tables (GRADEPro 3.6 (http://ims.cochrane.org/
revman/gradepro)). A SoF table consists of three parts: information about the review, a summary of the statistical results, and the
grade of the quality of evidence. The quality assessment of the
available evidence is comprised of the number of studies, the types
of studies (randomised or observational), and five factors including risk of bias, inconsistency of results, indirectness of evidence,
imprecision, and publication bias that affect the quality of the evidence. The five factors are used to judge whether the quality of
the collected evidence should be decreased or increased.

RESULTS

Description of studies
See: Characteristics of included studies; Characteristics of excluded
studies.

Subgroup analysis and investigation of heterogeneity
The following subgroup analyses were considered and performed
whenever possible.
1. Risk of bias: trials that were judged to be at low risk of bias
compared to trials judged to be at high risk of bias.
2. Participants: trials with treatment-naive patients compared
to relapsers and non-responders to previous antiviral
interventions.
3. Genotype: trials with participants infected with different
hepatitis C virus genotypes.
4. Dose of nitazoxanide: trials with different doses of
nitazoxanide.
5. Different regimens of nitazoxanide in combination with
any other intervention for chronic hepatitis C.
6. Co-infections and co-morbidities: participants with HIV,
solid organ transplant recipients, participants receiving
haemodialysis, with hepatitis B, alcoholism, iron overload, etc.
7. Adults compared to adolescents and children.
The difference between subgroups was assessed by the test of interaction (Altman 2003).

Results of the search
We identified a total of 282 bibliographic references through the
electronic searches of The Cochrane Hepato-Biliary Group Controlled Trials Register (n = 15), The Cochrane Central Register
of Controlled Trials (CENTRAL) (n = 11), MEDLINE (n = 36),
EMBASE (n = 142), LILACS (n = 14), and Science Citation
Index Expanded (n = 64). We identified three other references
on one ongoing trial (accessed January 2014) through searching
clinicaltrials.gov. Two other references were identified through
other sources.
We excluded 75 duplicates and screened 212 references. We excluded 164 clearly irrelevant references through reading abstracts.
Thus, we assessed 48 references for eligibility. We excluded additionally 22 of these references as they did not fulfil the inclusion
criteria. Out of the 26 references left, 23 were on seven trials fulfilling the inclusion criteria of our review, and three references were
on one still ongoing trial. We could not identify any references on
randomised clinical trials through scanning reference lists of the
identified ones. The reference flow is shown in Figure 1.

Nitazoxanide for chronic hepatitis C (Review)
Copyright © 2014 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.

11

Figure 1. Study flow diagram

Nitazoxanide for chronic hepatitis C (Review)
Copyright © 2014 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.

12

Included studies
Seven trials fulfilled all of the inclusion criteria. Details about the
trial interventions, participants characteristics, methods, and the
outcomes are shown in the ’Characteristics of included studies’
table. A summary of the included trials and a summary of the
characteristics of the included trials are shown in additional tables
(Table 1; Table 2).
The earliest included trial was published in 2008 (Rossignol
2008a), and the latest in 2012 (Basu 2012a). Four trials had a
parallel group design with two intervention groups (Rossignol
2008a; Shiffman 2009/2011; Bacon 2010; Shehab 2012), and
the other three trials included three intervention groups (Keeffe
2009a; Rossignol 2009a; Basu 2012a).

Design

Four trials compared nitazoxanide monotherapy versus placebo
(Rossignol 2008a; Keeffe 2009a; Shiffman 2009/2011; Bacon
2010). One of these trials compared nitazoxanide monotherapy versus placebo without additional antiviral drugs (Rossignol
2008a). The other three trials compared nitazoxanide monotherapy followed by an addition of the combination of peginterferon
and ribavirin (i.e., standard treatment) versus placebo followed
by an addition of standard treatment (Keeffe 2009a; Shiffman
2009/2011; Bacon 2010).
Two trials compared nitazoxanide monotherapy followed by the
addition of standard treatment versus no intervention plus standard treatment (Rossignol 2009a; Shehab 2012).
One trial compared nitazoxanide plus telaprevir and standard
treatment versus no intervention plus telaprevir and standard treatment (Basu 2012a).

Participants

Four publications reported on the sex of the participants (
Rossignol 2008a; Rossignol 2009a; Bacon 2010; Shehab 2012);
336 out of 353 were males (95.2%) based on the per protocol
analyses of the trials. Participants were 18 years of age or older, all
diagnosed with chronic hepatitis C genotypes 1 or 4. One participant in one of the trials was found to be co-infected with hepatitis
B, and an exception was made to allow this person concerned to
continue to participate in the trial (Rossignol 2008a).
A total of 350 participants were treatment-naive, i.e., had not had
any previous antiviral therapy (Keeffe 2009a; Rossignol 2009a;
Bacon 2010; Shehab 2012). A total of 51 participants were
treatment-experienced (Rossignol 2008a; Rossignol 2009a; Basu
2012a); 24 were only treatment-experienced to peginterferon
(Rossignol 2009a), five were treatment-experienced to both peginterferon and ribavirin (Rossignol 2008a), and the previous treatments given to the remaining 22 participants was unclear (Basu
2012a). Only one of these trials specified how these treatmentexperienced participants responded to their previous treatment;
participants in Basu 2012a were all stated to be partial-relapsers.
Two trials included 84 participants who were non-responders to
peginterferon and ribavirin (Shiffman 2009/2011; Basu 2012a).
Only six participants out of all the participants included in this
review were relapsers and it was unknown to which therapy (Basu
2012a). It was not known whether 47 participants in total had previously been treated with any antiviral-interventions (Rossignol
2008a; Basu 2012a).
None of the trial participants was reported to be co-infected with
any other concomitant or preceding disease except for one participant in the Rossignol 2008a trial who was co-infected with hepatitis B virus infection (HBV).

Interventions
Sample sizes

The seven trials contained a total of 538 randomised participants,
but only 507 of these were used in the per protocol analyses of the
trials.

Setting

The seven trials originated either from the USA or Egypt. Two trials were conducted in more than one centre in the USA (Shiffman
2009/2011; Bacon 2010), four trials were conducted in one or
more centres in Egypt (Rossignol 2008a; Keeffe 2009a; Rossignol
2009a; Shehab 2012), and one trial did not specify where it was
conducted (Basu 2012a). All seven trials included in our review
were published in English.

The trials were slightly heterogeneous in terms of dosage regimen
and duration of administration of nitazoxanide (see Characteristics
of included studies table). Six of the seven included trials used the
same dose of nitazoxanide, i.e., the standard dose of 500 mg twice
daily (Rossignol 2008a; Rossignol 2009a; Shiffman 2009/2011;
Bacon 2010; Basu 2012a; Shehab 2012). However, the Basu 2012a
trial also used 500 mg three times daily in one of its treatment
groups. The seventh trial used higher doses of nitazoxanide, i.e.,
675 mg and 1350 mg twice daily (Keeffe 2009a).
Dosages were similar in all of the trials that included the standard
treatment in their study. Peginterferon was administered weekly
at 180 µg in four trials (Keeffe 2009a; Rossignol 2009a; Shiffman
2009/2011; Bacon 2010). In the Basu 2012a trial, 180 µg peginterferon was given every other week, and in the Shehab 2012
trial, 160 µg was administered weekly. Administration of ribavirin

Nitazoxanide for chronic hepatitis C (Review)
Copyright © 2014 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.

13

was weight-based, and varied from 1000 mg to 1200 mg daily in
five trials (Keeffe 2009a; Rossignol 2009a; Shiffman 2009/2011;
Bacon 2010; Shehab 2012). One trial had a fixed dose of 1200
mg ribavirin daily (Basu 2012a). The dosage of telaprevir was 750
mg three times daily (Basu 2012a).
The duration of treatment in all trials varied. All trials had a
monotherapy of nitazoxanide or placebo administered over between four and 24 weeks, except for the trial by Basu 2012a for
which it was uncertain. Continuation of therapy in combination
with other drugs proceeded for between 24 to 48 weeks. Six of
the trials had a 24 week follow-up after completion of treatment
(Rossignol 2008a; Rossignol 2009a; Bacon 2010; Keeffe 2009a;
Shiffman 2009/2011; Shehab 2012). The follow-up was unclear
in the Basu 2012a trial. The details are displayed in Table 2.
None of the trials compared a long duration of nitazoxanide treatment with a short duration of nitazoxanide treatment.

Outcomes

All the included trials evaluated the effect and efficacy of nitazoxanide on sustained virological response and virological endof-treatment response, and four trials evaluated participants who
showed no improvement in alanine aminotransferase levels according to the authors’ definitions (See Characteristics of included
studies) (Rossignol 2008a; Rossignol 2009a; Shiffman 2009/2011;
Shehab 2012). Five trials evaluated the safety of nitazoxanide for
treatment of chronic hepatitis C by looking at number of participants with adverse events (Rossignol 2008a; Keeffe 2009a;
Rossignol 2009a; Bacon 2010; Shehab 2012). For more details,
see the Characteristics of included studies table.
Only one trial reported on all-cause mortality and chronic hepatitis
C-related mortality (Shehab 2012).
None of the trials reported on quality of life, development of
morbidity, and only one trial with a long-term follow-up reported

on the number of participants with histological changes (Rossignol
2008a).
We wrote to the contact authors of four of the trials in order to
obtain additional information about the trials (Rossignol 2008a;
Rossignol 2009a; Basu 2012a; Shehab 2012), and received answers relating to Shehab 2012, Basu 2012a, and Keeffe 2009a.
Detailed information is given in the Characteristics of included
studies table.

Excluded studies
We excluded 22 references to 16 studies. Four studies were not
randomised (Basu 2009; Keeffe 2009; Yoffe 2009; Asrani 2010)
and the remaining 12 studies were review articles of no interest to
our review (Characteristics of excluded studies).
Data on adverse events were reported as not serious, mild to moderate, and intermittent in nature in the study by Yoffe 2009. This
study enrolled 14 treatment-naive patients with chronic hepatitis
C genotype 1 to receive nitazoxanide plus peginterferon and ribavirin. Four patients developed adverse events including diarrhoea
(n = 3), rash (n = 2), cellulitis (n = 3), and laryngitis (n = 1). These
data are not part of the meta-analysis on adverse events.

Ongoing studies
We identified one ongoing trial (Kohla ongoing); we hope that
the results from this trial will be included in the next update of
this review.

Risk of bias in included studies
All of the included trials had high risk of bias, as they were judged
to be at high risk or unclear risk in at least one domain (Risk of bias
in included studies). A summary of our assessment, by domain
and trial, is presented in Figure 2 and Figure 3.

Nitazoxanide for chronic hepatitis C (Review)
Copyright © 2014 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.

14

Figure 2. Risk of bias summary: review authors’ judgements about each risk of bias item for each included
study

Nitazoxanide for chronic hepatitis C (Review)
Copyright © 2014 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.

15

Figure 3. Risk of bias graph: review authors’ judgements about each risk of bias item presented as
percentages across all included studies

All of the included trials stated that they allocated participants
randomly to comparison groups. The generation of the allocation
sequence was described in three trials (Rossignol 2008a; Rossignol
2009a; Basu 2012a) (Table 2). Both Basu 2012a and Rossignol
2008a used sealed envelopes, however, there was no information
about whether the envelopes had been sequentially-numbered and
were opaque. Rossignol 2009a randomised participants sequentially to one of the three treatment groups using the next available
package of study medication, however, as the investigators were
not blinded to intervention groups, it is possible that the allocation
may have been foreseeable. Accordingly, allocation concealment
was assessed as being at high risk of bias in all the trials.
The method of blinding of participants and personnel was partially
described in two of the trials (Rossignol 2008a; Rossignol 2009a).
One of the included trials was an open-label trial and was judged
as being at high risk of bias (Shehab 2012). The four remaining
trials did not provide information about blinding in the trial, and,
therefore, the risk of bias was judged to be high (Keeffe 2009a;
Shiffman 2009/2011; Bacon 2010; Basu 2012a).
Incomplete data were addressed adequately in five of the seven trials (Rossignol 2008a; Rossignol 2009a; Bacon 2010; Basu 2012a;
Shehab 2012), which were judged to have a low risk of bias for
this domain. The other two trials were judged to have a high risk
of bias for this domain (Keeffe 2009a; Shiffman 2009/2011).
Six of the seven trials were judged to have a low risk of selective re-

porting bias, as primary and secondary outcomes were adequately
assessed. Only one of the trials had a high risk of bias for this domain as not all the outcomes were adequately described (Shiffman
2009/2011).
Other potential sources of bias
All except two of the trials had received or seem to have received funding from the medical industry (Romark Laboratories,
LCTampa, FL, USA) (Basu 2012a; Shehab 2012), which put the
trials at high risk of bias in terms of for-profit bias.

Effects of interventions
See: Summary of findings for the main comparison
Nitazoxanide for chronic hepatitis C; Summary of findings 2
Nitazoxanide for chronic hepatitis C; Summary of findings 3
Nitazoxanide for chronic hepatitis C; Summary of findings 4
Nitazoxanide for chronic hepatitis C; Summary of findings 5
Nitazoxanide for chronic hepatitis C

Nitazoxanide versus placebo or no intervention

Primary outcomes

Nitazoxanide for chronic hepatitis C (Review)
Copyright © 2014 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.

16

All-cause mortality or chronic hepatitis C-related mortality
Only one trial provided information on all-cause mortality and on
chronic hepatitis C-related mortality (Shehab 2012). None of the
participants in the trial died during the follow-up period (Analysis
1.1; Analysis 1.2).

Morbidity
None of the included trials reported on the number of participants who developed cirrhosis, ascites, variceal bleeding, hepatic
encephalopathy, or hepatocellular carcinoma during the trial. Only
baseline data on morbidity were assessed, and this was done in only
four of the seven included trials (Rossignol 2008a; Keeffe 2009a;
Rossignol 2009a; Bacon 2010). Therefore, data on this outcome
could not be included in the analyses (Analysis 1.3).

Adverse events

Overall effect of nitazoxanide versus placebo or no
intervention including trials with and without cointerventions
Four trials reported on the number of participants who experienced serious and non-serious adverse events (Rossignol 2008a;
Rossignol 2009a; Bacon 2010; Shehab 2012). Sixty-seven out of
a total of 331 participants (20%) were reported as having had adverse events. The meta-analysis with both the fixed-effect model
and random-effects model showed no significant difference in the
effect of nitazoxanide versus placebo or versus no intervention on
the number of participants who experienced adverse events (37
out of 179 (21%) versus 30 out of 152 (20%); fixed-effect model:
RR 1.10; 95% CI 0.71 to 1.71; I2 = 65%, P value 0.68) (Analysis
1.4).
As there were no trials with an overall low risk of bias, we performed a trial sequential analysis on all included trials that reported on adverse events (Figure 4). The trial sequential analysis
showed no evidence to support or refute the theory that nitazoxanide influences the number of participants with adverse events
(Figure 4).

Nitazoxanide for chronic hepatitis C (Review)
Copyright © 2014 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.

17

Figure 4. Trial sequential analysis of the fixed-effect meta-analysis of the effect of nitazoxanide versus
placebo or no intervention on adverse events in chronic hepatitis C infected patients. The trial sequential
analysis is performed with a type 1 error of 5% (two sided), a power of 80%, an assumed control proportion of
number of patients with adverse events of 20%, and an anticipated relative risk reduction (RRR) of 40%. The
diversity-adjusted required information size (DARIS) to detect or reject a RRR of 40% with a between trial
heterogeneity in the meta-analysis is estimated to 1946 participants. The actually accrued number of
participants is 331, which is only 17% of the required information size. The blue cumulative Z-curve does not
cross the conventional statistical boundaries or the red inward sloping trial sequential monitoring boundaries
for benefit or harm. Not even with a large RRR of 40% is there evidence to support that nitazoxanide has any
effect on adverse events. The cumulative Z-curve does not reach the futility area, demonstrating that further
trials may be needed.

Nitazoxanide for chronic hepatitis C (Review)
Copyright © 2014 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.

18

Nitazoxanide versus placebo
Adverse events were reported in 11 out of 25 (44%) of the participants in the nitazoxanide group versus six out of 25 (24%) in
the placebo group in the Rossignol 2008a trial; this difference was
not statistically significant (P value 0.15). Two of these 17 participants had serious adverse events that required hospitalisation; one
in each of the two intervention groups. None of the participants
with adverse events discontinued the intervention (Analysis 1.4).
The Rossignol 2008a trial provided no information on dose-reductions in participants that had adverse events.

Nitazoxanide plus standard treatment versus placebo plus
standard treatment
Bacon 2010 reported adverse events in 11 out of 75 (15%) participants in the nitazoxanide plus standard treatment group versus
seven out of 37 (19%) participants in the placebo plus standard
treatment group; there was no significant difference between the
groups (P value 0.56). Of the participants who experienced adverse effects, seven from the nitazoxanide group and four from the
placebo group discontinued treatment due to the adverse events
(Analysis 1.4).
The Bacon 2010 trial provided no information on dose-reductions
due to adverse events.

five in the no intervention plus standard treatment group. The
intervention dose reductions were not specified in the Rossignol
2009a trial. The Shehab 2012 trial reported dose reductions of
peginterferon in four participants in the nitazoxanide plus standard treatment group versus seven participants in the no intervention plus standard treatment group. The dose of ribavirin was also
reduced in three participants in the standard treatment group.

Subgroup differences
Tests for subgroup differences showed no significant difference in
effect between trials assessing:
1. the effects of nitazoxanide versus placebo;
2. nitazoxanide plus standard treatment versus placebo plus
standard treatment; and
3. nitazoxanide plus standard treatment versus no intervention
plus standard treatment (P value 0.34).

Secondary outcomes

Quality of life
None of the included trials reported on quality of life.
Failure to achieve sustained virological response

Nitazoxanide plus standard treatment versus no intervention
plus standard treatment
Adverse events were reported in 15 out of 79 (19%) participants
in the nitazoxanide plus standard treatment group versus 17 out
of 90 (19%) participants in the no intervention plus standard
treatment group in the two trials with this comparison (Rossignol
2009a; Shehab 2012). In the Shehab 2012 trial, none of the participants had serious adverse events that required hospitalisation.
Three participants discontinued treatment due to adverse events
in the standard treatment group, compared with two in the nitazoxanide group. In the Rossignol 2009a trial one participant in
the standard treatment group had an adverse event that required
hospitalisation. Five participants in the standard treatment group
discontinued medication because of adverse events. The metaanalysis showed no statistically significant difference between the
effect of nitazoxanide plus standard treatment and standard treatment alone (fixed-effect model: RR 1.00; 95% CI 0.51 to 1.96; I
2 = 85%, P value 0.99) (Analysis 1.4).
The Rossignol 2009a trial reported that seven participants experienced dose-reductions as a result of adverse events; two participants in the nitazoxanide plus standard treatment group versus

Overall effect of nitazoxanide versus placebo or no
intervention
All of the included trials reported on the proportion of participants who failed to achieve sustained virological response. In the
nitazoxanide group a total of 159 out of 290 (55%) participants
failed to achieve sustained virological response versus 133 out of
208 (64%) participants in the placebo or no intervention group.
The meta-analysis with both the fixed-effect model and randomeffects model showed a significant difference in favour of nitazoxanide versus placebo or no intervention (fixed-effect model: RR
0.85; 95% CI 0.75 to 0.97; I2 = 0%, P value 0.02) (Analysis 1.6).
As there were no trials at low risk of bias, we performed a trial
sequential analysis on all included trials that reported on failure of
sustained virological response. The trial sequential analysis of the
data supports the statistically significant effect of nitazoxanide versus placebo or no intervention for sustained virological response
(Figure 5). The result of the trial sequential analysis is shown by
the cumulated Z-curve (blue curve) that crosses the trial sequential monitoring boundary for benefit (red inward sloping curve),

Nitazoxanide for chronic hepatitis C (Review)
Copyright © 2014 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.

19

which implies that there is evidence for a beneficial effect of nitazoxanide in encouraging sustained virological response; this result
is free of risk of random errors. However, as stated above, we cannot exclude risks of systematic errors (bias).
Figure 5. Trial sequential analysis of the fixed-effect meta-analysis of the effect of nitazoxanide versus
placebo or no intervention on risk of failure of sustained virological response in patients with chronic hepatitis
C virus infection. The trial sequential analysis is performed with a type 1 error of 5% (two sided), a power of
80%, an assumed control proportion of number of participants with failure of sustained virological response of
64%, and an anticipated relative risk reduction (RRR) of 20%. The diversity-adjusted required information size
(DARIS) to detect or reject a RRR of 20% with a between trial heterogeneity in the meta-analysis is estimated
to 419 participants. The actually accrued number of participants is 498. The blue cumulative Z-curve crosses
the red inward sloping trial sequential monitoring boundaries for benefit. This implies that there is no risk of
random error in the estimate of a beneficial effect of nitazoxanide versus placebo or no intervention.

Nitazoxanide for chronic hepatitis C (Review)
Copyright © 2014 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.

20

We performed sensitivity analyses using two extreme scenarios for
participants with incomplete or missing data regarding sustained
virological response.
Firstly, we performed a best-worst case scenario analysis assuming
that participants with data missing for this outcome in the nitazoxanide group did achieve sustained virological response, and
that all participants in the placebo or no intervention group for
whom the relevant data were missing did not achieve sustained
virological response. This sensitivity analysis with the fixed-effect
model revealed a statistically significant effect favouring nitazoxanide versus placebo or no intervention (best-worst case scenario:
RR 0.81; 95% CI 0.70 to 0.94, I2 = 63%, P value 0.004; four
trials) (Analysis 5.1).
Secondly, we performed a worst-best case scenario analysis where
we assumed that participants with data missing for this outcome
in the nitazoxanide group did not achieve sustained virological
response, and that all participants in the placebo or no intervention group for whom the relevant data were missing did achieve
sustained virological response. This sensitivity analysis with the
fixed-effect model did not reveal a statistically significant effect of
nitazoxanide versus placebo or no intervention (worst-best case
scenario: RR 0.87; 95% CI 0.76 to 1.01; I2 = 4%, P value 0.06)
(Analysis 6.1).
Nitazoxanide versus placebo
Rossignol 2008a reported on failure of sustained virological response in 21 out of 25 (84%) participants in the nitazoxanide
group versus 25 out of 25 (100%) participants in the placebo
group; there was no significant difference between the groups (P
value 0.07).
Nitazoxanide plus standard treatment versus placebo plus
standard treatment
Three trials reported on failure of sustained virological response
in participants treated with nitazoxanide plus standard treatment
versus placebo plus standard treatment (Keeffe 2009a; Shiffman
2009/2011; Bacon 2010). Out of 150 participants given nitazoxanide plus standard treatment, 89 failed to achieve sustained virological response (59%) versus 51 out of 67 (76%) participants
given placebo plus standard treatment. The meta-analysis showed
no statistically significant difference between the treatment groups
on sustained virological response using both the fixed-effect model
and random effects-model (fixed-effect model: RR 0.85; 95% CI
0.72 to 1.00; I2 = 64%, P value 0.05) (Analysis 1.6).
Nitazoxanide plus standard treatment versus no intervention
plus standard treatment
Two trials reported on failure of sustained virological response
in participants treated with nitazoxanide plus standard treatment
versus no intervention plus standard treatment (Rossignol 2009a;
Shehab 2012). Forty-one out of 91 (45%) participants given nita-

zoxanide plus standard treatment failed to achieve sustained virological response versus 46 participants out of 90 (51%) in the standard treatment group. The meta-analysis showed no statistically
significant difference between the treatment groups for failure of
sustained virological response (fixed-effect model: RR 0.88; 95%
CI 0.65 to 1.19; I2 = 0%, P value 0.05) (Analysis 1.6).
Nitazoxanide plus standard treatment plus telaprevir versus
no intervention plus standard treatment plus telaprevir
Only one trial compared nitazoxanide plus standard treatment
plus telaprevir versus no intervention plus standard treatment plus
telaprevir (Basu 2012a). The trial reported failure of sustained
virological response in eight out of 24 (33%) participants in the
nitazoxanide plus standard treatment plus telaprevir group versus
11 out of 26 (42%) participants in the standard treatment plus
telaprevir group there was no significant difference between the
groups (P value 0.52).
Subgroup differences
Tests for subgroup differences showed no significant difference in
effect between the trials assessing:
1. the effects of nitazoxanide versus placebo;
2. nitazoxanide plus standard treatment versus placebo plus
standard treatment; and
3. nitazoxanide plus standard treatment versus no intervention
plus standard treatment (P value 0.99).
Failure of virological end-of-treatment response

Overall effect of nitazoxanide versus placebo or no
intervention
All of the included trials reported on risk of failure of virological
end-of-treatment response. Failure of virological end-of-treatment
response was reported in 235 out of 498 (47%) participants of
the included trials. In the nitazoxanide groups, a total of 125 out
of 290 (43%) participants failed to achieve virological end-oftreatment response versus 110 out of 208 (53%) participants in
the placebo or no intervention group. Meta-analysis with both
the fixed-effect model and the random-effects model showed a
significant beneficial effect for nitazoxanide versus placebo or no
intervention (fixed-effect model: RR 0.81; 95% CI 0.69 to 0.96;
I2 = 46%, P value 0.01) (Analysis 1.7).
As there were no trials with a low risk of bias, we performed a trial
sequential analysis on all included trials that reported on failure
to achieve virological end-of-treatment response (Figure 6). The
trial sequential analysis does not support the statistically significant findings of the meta-analysis for a beneficial effect of nitazoxanide in preventing failure to achieve virological end-of-treatment
response.

Nitazoxanide for chronic hepatitis C (Review)
Copyright © 2014 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.

21

Figure 6. Trial sequential analysis of the fixed-effect meta-analysis of the effect of nitazoxanide versus
placebo or no intervention on risk of failure of virological end-of-treatment response in patients with chronic
hepatitis C virus infection. The trial sequential analysis is performed with a type 1 error of 5% (two sided), a
power of 80%, an assumed control proportion of number of patients who failed to achieve a virological end-oftreatment response of 53%, and an anticipated relative risk reduction (RRR) of 20%. The diversity-adjusted
required information size (DARIS) to detect or reject a RRR of 20% with a between trial heterogeneity in the
meta-analysis is estimated to 2263 participants. The actually accrued number of participants is 498, which is
22% of the required information size. The blue cumulative Z-curve does not cross the red inward sloping trial
sequential monitoring boundaries for benefit or harm. Therefore, there is no evidence to support that
nitazoxanide influences number of patients who failed (which is not even drawn by the programme) to achieve
virological end-of-treatment response. The cumulative Z-curve does not reach the futility area, demonstrating
that further trials may be needed.

Nitazoxanide for chronic hepatitis C (Review)
Copyright © 2014 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.

22

Nitazoxanide versus placebo
Rossignol 2008a reported failure of virological end-of-treatment
response in 18 out of 25 (72%) participants in the nitazoxanide
group versus 25 out of 25 (100%) participants in the placebo
group; the difference between the groups was statistically significant (P value 0.01).

Only one trial investigated nitazoxanide plus standard treatment
plus telaprevir versus no intervention plus standard treatment plus
telaprevir (Basu 2012a). The trial reported failure of virological
end-of-treatment response in five out of 24 (21%) participants receiving nitazoxanide in combination with telaprevir and standard
treatment versus 10 out of 26 (38%) participants in the telaprevir plus standard treatment group (Basu 2012a). The difference
between the treatment groups was not statistically significant (P
value 0.19) (Analysis 1.7).

Nitazoxanide plus standard treatment versus placebo plus
standard treatment
Three trials reported on risk of failure of virological end-of-treatment response in participants treated with nitazoxanide plus standard treatment versus placebo plus standard treatment (Keeffe
2009a; Shiffman 2009/2011; Bacon 2010). Sixty-eight out of 150
(45%) participants given nitazoxanide plus standard treatment
failed to achieve virological end-of-treatment response versus 46
participants out of 67 (69%) in the placebo plus standard treatment group. Meta-analysis using the fixed-effect model showed a
statistically significant beneficial effect for nitazoxanide plus standard treatment on failure of virological end-of-treatment response
(RR 0.73; 95% CI 0.60 to 0.89; I2 = 80%, P value 0.002), but
this significant difference was not maintained when we used the
random-effects model (RR 0.64; 95% CI 0.36 to 1.15; I2 = 80%,
P value 0.14) (Analysis 1.7).

Subgroup differences
Tests for subgroup differences showed no significant difference in
effect between the trials assessing:
1. the effects of nitazoxanide versus placebo;
2. nitazoxanide plus standard treatment versus placebo plus
standard treatment; and
3. nitazoxanide plus standard treatment versus no intervention
plus standard treatment (P value 0.16).
Participants without improvement in alanine
aminotransferase or aspartate aminotransferase serum levels

Overall effect of nitazoxanide versus placebo or no
intervention
Nitazoxanide plus standard treatment versus no intervention
plus standard treatment
Two trials reported on risk of failure of virological end-of-treatment response in participants treated with nitazoxanide plus standard treatment versus no intervention plus standard treatment
(Rossignol 2009a; Shehab 2012). Thirty-four out of 91 (37%)
participants given nitazoxanide plus standard treatment failed to
achieve virological end-of-treatment response versus 29 out of 90
(32%) participants in the standard treatment group. Meta-analysis showed no statistically significant difference in effect of nitazoxanide added to standard treatment versus no intervention plus
standard treatment on failure of virological end-of-treatment response (fixed-effect model: RR 1.16; 95% CI 0.78 to 1.73; I2 =
0%, P value 0.46) (Analysis 1.7).

Nitazoxanide plus standard treatment plus telaprevir versus
no intervention plus standard treatment plus telaprevir

Three trials reported on participants who did not show improvement in alanine aminotransferase or aspartate aminotransferase
serum levels (Rossignol 2008a; Rossignol 2009a; Shehab 2012).
Ninety-nine participants out of a total of 192 (52%) participants
in the included trials, were reported as not showing improvement
in alanine aminotransferase or aspartate aminotransferase serum
levels. Meta-analysis with both the fixed-effect model and the random-effects model showed no significant effect of nitazoxanide
(52 out of 97 (54%) participants) versus placebo or other intervention (47 out of 95 (49%) participants) (fixed-effect model: RR
1.09; 95% CI 0.84 to 1.42; I2 = 0%, P value 0.50) (Analysis 1.8).
As there were no trials with a low risk of bias, we performed a
trial sequential analysis on all included trials that reported on the
number of participants who did not show improvement in alanine
aminotransferase or aspartate aminotransferase levels (Figure 7).
The trial sequential analysis showed that there is no evidence to
support or refute an effect of nitazoxanide on participants who do
not show improvement in alanine aminotransferase or aspartate
aminotransferase levels.

Nitazoxanide for chronic hepatitis C (Review)
Copyright © 2014 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.

23

Figure 7. Trial sequential analysis of the fixed-effect meta-analysis of the effect of nitazoxanide versus
placebo or no intervention on number of participants with chronic hepatitis C virus infection without
improvement in alanine aminotransferase or aspartate aminotransferase serum levels. The trial sequential
analysis is performed with a type 1 error of 5% (two sided), a power of 80%, an assumed control proportion of
number of patients with morbidity of 49%, and an anticipated relative risk reduction (RRR) of 20%. The
diversity-adjusted required information size (DARIS) to detect or reject a RRR of 20% with a between trial
heterogeneity in the meta-analysis is estimated to 806 participants. The actually accrued number of
participants is 192, which is 24% of the required information size. The blue cumulative Z-curve does not cross
the red inward sloping trial sequential monitoring boundaries for benefit or harm. Therefore, there is no
evidence to support that nitazoxanide has any effect on number of participants without change in alanine
aminotransferase or aspartate aminotransferase serum levels. The cumulative Z-curve does not reach the
futility area (which is not even drawn by the programme), demonstrating that further trials may be needed.

Nitazoxanide for chronic hepatitis C (Review)
Copyright © 2014 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.

24

Nitazoxanide versus placebo
The Rossignol 2008a trial compared nitazoxanide versus placebo,
and reported only on the number of participants who showed
no improvement in alanine aminotransferase levels. This trial reported on these participants from baseline to end-of-treatment
only as per protocol (23 participants in the active group and 24 in
the placebo group), and did not provide information on alanine
aminotransferase levels for the remaining participants. The participants were assigned as either ’normalised’, ’remained normal’,
’remained elevated’, or ’normal to elevated’. Data in the Data and
analyses section are based on the intention-to-treat principle. Out
of a total of 25 participants, 18 participants (72%) in the nitazoxanide group versus 20 out of 25 (80%) participants in the placebo
group showed no improvement in alanine aminotransferase levels; the difference between the groups was not significant (P value
0.51) (Analysis 1.8).

Nitazoxanide plus standard treatment versus no intervention
plus standard treatment
The two trials that compared nitazoxanide plus standard treatment versus no intervention plus standard treatment, also reported
on the number of participants who showed no improvement in
alanine aminotransferase levels (Rossignol 2009a; Shehab 2012).
The Rossignol 2009a trial only reported on unimproved alanine
aminotransferase levels in participants who achieved sustained virological response. The Shehab 2012 trial included all participants in the report on the number of participants who showed
no improvement in alanine aminotransferase levels. Meta-analysis showed no statistical significant difference between groups for
nitazoxanide in combination with standard treatment versus no
intervention plus standard treatment on the number of participants who showed no improvement in alanine aminotransferase
or aspartate aminotransferase serum levels (34 out of 72 (47%)
versus 27 out of 70 (39%), respectively) (fixed-effect model: RR
1.24; 95% CI 0.85 to 1.80; I2 = 0%, P value 0.27) (Analysis 1.8).

Subgroup differences
Tests for subgroup differences showed no significant difference in
effect between the trials assessing:
1. the effects of nitazoxanide versus placebo;
2. nitazoxanide plus standard treatment versus placebo plus
standard treatment; and
3. nitazoxanide plus standard treatment versus no intervention
plus standard treatment (P value 0.20).

Number of participants without histological improvement
None of the trials reported on participants who showed no histological improvement (Analysis 1.9).
Long-term follow-up
Only one of the included trials assessed outcomes at a long-term
follow-up (Rossignol 2008a).
Thirteen out of the 50 participants included in the multicentre
Rossignol 2008a trial were assessed in a long-term follow-up trial
by Kabil 2011. Of the 13 participants, nine were from the original
nitazoxanide group that had included 25 participants, and four
were from the original placebo group that had included 25 participants. How the 13 participants were selected for the long-term
follow-up was not specified (Kabil 2011). During the follow-up
period the participants did not receive any antiviral treatment.
Although no serious adverse events were reported in either of the
treatment groups, and none of the total of 13 participants developed cirrhosis, two of the nine participants given nitazoxanide experienced adverse events. None of the participants required dose
reduction or discontinued treatment due to adverse events.
Six out of the nine participants given nitazoxanide failed to achieve
sustained virological response and virological end-of-treatment response versus all of the participants in the placebo group who
failed to achieve these virological responses.
No significant improvement was identified between treatment
groups for mean alanine aminotransferase levels before and after
treatment. Three participants achieved sustained virological response; two were associated with normalisation of alanine aminotransferase levels, while the third fluctuated. Participants who were
non-responders showed no improvement in alanine aminotransferase levels.
The Rossignol 2008a trial only reported data on histological improvement for the long-term follow-up study (Kabil 2011). Paired
liver biopsies before and after treatment were performed in three
participants. In a post treatment liver biopsy one of the participants
that had achieved sustained virological response, showed no worsening of fibrosis and minimal improvement in his necroinflammatory score. This participant was from the nitazoxanide group. The
other two participants who had liver biopsy after treatment were
non-responders from the placebo group. In one of them fibrosis
had worsened and there was a minimal increase in the necroinflammatory score. The other participant showed regression of both
fibrosis- and necroinflammatory stage. Only one participant had
a liver biopsy at the long-term follow-up; this participant was a
non-responder and showed no worsening of fibrosis.
Subgroup analyses

Nitazoxanide for chronic hepatitis C (Review)
Copyright © 2014 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.

25

We conducted subgroup analyses stratifying the trials according
to hepatitis C virus genotype and dose of nitazoxanide. We also
performed a subgroup analysis comparing treatment-naive participants to relapsers and non-responders to previous antiviral interventions.
In the Rossignol 2008a trial, five out of the 50 participants were
treatment-experienced with peginterferon and ribavirin; three
were in the nitazoxanide group (n = 25), and two were in the
placebo group.

Treatment-naive compared to relapsers and non-responders
Trials reporting on the number of treatment-naive participants
with adverse events showed no statistically significant difference
between nitazoxanide (17%) versus placebo or no intervention
(19%) using both the fixed-effect model and the random-effects
model (fixed-effect model: RR 0.92; 95% CI 0.54 to 1.56; I2 =
71%, P value 0.75). None of the trials that included relapsers and
non-responders participants reported on adverse events (Analysis
2.4).
The subgroup analyses stratifying the trials according to previous
antiviral therapy showed a statistically significant difference for
failure of sustained virological response between the nitazoxanide
group (52%) versus the placebo or no intervention group (60%)
(fixed-effect model: RR 0.82; 95% CI 0.69 to 0.96; I2 = 55%, P
value 0.02) (Analysis 2.6). Meta-analysis of trials with treatmentnaive participants showed a statistically significant effect on achieving sustained virological response, favouring nitazoxanide; eightytwo out of 187 participants (44%) failed to achieve sustained virological response versus 75 out of 135 participants (56%) in the
placebo or no intervention group using both the fixed-effect model
and the random-effects model (fixed-effect model: RR 0.78; 95%
CI 0.62 to 0.97; I2 = 37%, P value 0.03). Meta-analysis of trials
that included relapsers and non-responders did not show any significant difference between the nitazoxanide group (80%) versus
the placebo or no intervention group (78%) (fixed-effect model:
RR 0.91; 95% CI 0.77 to 1.09; I2 = 0%, P value 0.31) for failure
of sustained virological response. The test for subgroup difference
showed no significant difference (P value 0.26) (Analysis 2.6).
Meta-analysis of subgroups stratifying the trials according to previous antiviral therapy on failure of virological end-of-treatment
response showed no statistically significant difference between the
nitazoxanide group (41%) versus the placebo or no intervention
group (48%) (fixed-effect model: RR 0.82; 95% CI 0.67 to 1.00;
I2 = 49%, P value 0.05) (Analysis 2.7). Meta-analysis of trials
with treatment-naive participants showed no statistically significant difference on failure of virological end-of-treatment response
between the nitazoxanide group (32%) versus the placebo or no
intervention group (39%) when both the fixed-effect model and
the random-effects model were used (fixed-effect model: RR 0.80;
95% CI 0.60 to 1.06; I2 = 58%, P value 0.12) (Analysis 2.7). The
trials that included relapsers and non-responders also did not show

any significant difference between the nitazoxanide group (86%)
versus the placebo or no intervention group (100%) (fixed-effect
model: RR 0.87; 95% CI 0.75 to 1.00, P value 0.05) for failure
of virological end-of-treatment response (Analysis 2.7). Testing
for subgroup differences showed no significant difference (P value
0.60).
Those trials that included treatment-naive participants that reported on the number of participants who showed no improvement in alanine aminotransferase or aspartate aminotransferase
serum levels showed no statistically significant difference between
groups for nitazoxanide (47%) versus placebo or no intervention
(39%) using both the fixed-effect model and the random-effects
model (fixed-effect model: RR 1.24; 95% CI 0.85 to 1.80; I2 =
0%, P value 0.27). None of the trials that included relapsers and
non-responders reported on the number of participants who did
not show improvement in alanine aminotransferase or aspartate
aminotransferase serum levels (Analysis 2.8).

Genotype 1- compared to genotype 4-infected participants
A subgroup meta-analysis of the number of genotype 4 participants with adverse events showed no significant difference between
nitazoxanide (25%) versus placebo or no intervention (20%) using both a fixed-effect model and a random-effects model (fixed
effects-model: RR 1.24; 95% CI 0.74 to 2.08; I2 = 72%, P value
0.42) (Analysis 3.4). Only one trial with genotype 1 participants
reported on the number with adverse events; there was no significant difference between the nitazoxanide group (15%) versus
placebo or no intervention group (18%) (fixed-effect model: RR
0.78; 95% CI 0.33 to 1.84, P value 0.56) for adverse events. Testing for subgroup differences showed no significant difference (P
value 0.36) (Analysis 3.4).
Meta-analysis using both the fixed-effect model and the randomeffects model showed no significant difference between nitazoxanide versus placebo or no intervention for failure of sustained
virological response in either genotype 1 participants (64% in the
nitazoxanide group versus 68% in the placebo or no intervention
group) (fixed-effect model: RR 0.88; 95% CI 0.74 to 1.04; I2
= 0%, P value 0.13), or in genotype 4 participants (46% in the
nitazoxanide group versus 61% in the placebo or no intervention
group) (fixed-effect model: RR 0.83; 95% CI 0.68 to 1.02; I2 =
0%, P value 0.07). Testing for subgroup differences showed no
significant difference (P value 0.70) (Analysis 3.6).
The three trials that included genotype 1 participants showed a
statistically significant difference between the nitazoxanide group
(50%) versus placebo or no intervention group (61%), favouring
nitazoxanide, on failure of virological end-of-treatment response
using the fixed-effect model and the random-effects model (fixedeffect model: RR 0.76; 95% CI 0.61 to 0.93; I2 = 52%, P value
0.010) (Bacon 2010; Shiffman 2009/2011; Basu 2012a). The trials including genotype 4 participants showed no statistically significant difference between the nitazoxanide group (37%) versus

Nitazoxanide for chronic hepatitis C (Review)
Copyright © 2014 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.

26

placebo or the no intervention group (47%) for failure of virological end-of-treatment response (fixed-effect model: RR 0.88; 95%
CI 0.69 to 1.12; I2 = 70%, P value 0.29). Testing for subgroup difference showed no significant difference (P value 0.38) (Analysis
3.7).
Subgroup meta-analysis on the number of participants with no
improvement in alanine aminotransferase serum levels in trials including genotype 4 participants showed no significant difference
between the nitazoxanide group (54%) versus placebo or no intervention group (49%) using both the fixed-effect model and random-effects model (fixed-effect model: RR 1.09; 95% CI 0.84 to
1.42; I2 = 0%, P value 0.50). No trials with genotype 1 participants
reported on the number of participants without improvement in
alanine aminotransferase or aspartate aminotransferase serum levels (Analysis 3.8).

anide, between the nitazoxanide group (13%) versus the placebo
or no intervention group (43%) on the risk in failure of virological
end-of-treatment response using the fixed-effect model and the
random-effects model (fixed-effect model: RR 0.36; 95% CI 0.17
to 0.76; I2 = 0%, P value 0.008). Testing for subgroup differences
did show a statistical significant difference (P value 0.03) (Analysis
4.7).
We planed to perform subgroup analyses on the number of participants who showed no improvement in alanine aminotransferase
or aspartate aminotransferase serum levels, stratifying the trials according to the dose of nitazoxanide administered, but we could
not do this because all the trials that provided information on this
outcome had used the standard dose of nitazoxanide. However,
subgroup analysis did not show any significant difference between
the nitazoxanide group (54%) and the placebo or no intervention
group (49%) (fixed-effect model: RR 1.09; 95% CI 0.84 to 1.42;
I2 = 0%, P = 0.50) (Analysis 4.8).

Nitazoxanide dose comparison
Six out of the seven trials included administered the standard dose
of nitazoxanide (500 mg twice daily) to participants (Rossignol
2008a; Rossignol 2009a; Shiffman 2009/2011; Bacon 2010; Basu
2012a; Shehab 2012). Two trials used higher daily doses of nitazoxanide (Basu 2012a; Keeffe 2009a).
We planed to perform a subgroup meta-analyses for adverse events,
stratifying trials according to standard dose versus experimentaldose of nitazoxanide, but we could not do so because all trials that
provided information about adverse events had used the standard
dose of nitazoxanide, which for this subgroup did not show any
significant difference between the nitazoxanide group (21%) versus placebo or no intervention group (20%) (fixed-effect model:
RR 1.10; 95% CI 0.71 to 1.71; I2 = 65%, P value 0.68) (Analysis
4.4).
Subgroup analyses of trials that administered the standard dose of
nitazoxanide revealed no statistically significant difference in the
effect of nitazoxanide (60%) versus placebo or no intervention
(65%) for failure of sustained virological response using both the
fixed-effect model and random-effects models (fixed-effect model:
RR 0.87; 95% CI 0.76 to 1.00; I2 = 0%, P value 0.05). Metaanalysis of trials that administered experimental doses of nitazoxanide also failed to show any significant difference between nitazoxanide (24%) versus placebo or no intervention (45%) (fixedeffect model: RR 0.57; 95% CI 0.31 to 1.04; I2 = 0%, P value
0.07%). Testing for subgroup difference revealed no statistical significant difference (P value 0.17) (Analysis 4.6).
Subgroup analyses of trials that administered the standard dose of
nitazoxanide revealed no statistically significant difference in effect
of nitazoxanide (49%) versus placebo or no intervention (53%)
on failure of virological end-of-treatment response using both the
fixed-effect model and random-effects model (fixed-effect model:
RR 0.85; 95% CI 0.72 to 1.00; I2 = 0%, P value 0.06). However,
the meta-analysis of trials that administered the experimental-dose
of nitazoxanide did show a significant effect, favouring nitazox-

Abandoned subgroup analyses
We could not perform subgroup analyses in which the included
trials were to be stratified according to risk of bias, different regimens of nitazoxanide in combination with any other intervention
fo