Antiviral Therapy 2012; 17:255264 (doi: 10.
3851/IMP2011)
Review
Antivirals for management of herpes zoster
including ophthalmicus: a systematic review of
high-quality randomized controlled trials
Elissa M McDonald1*, Johannes de Kock2, Felix SF Ram3
1
Department of Ophthalmology, New Zealand National Eye Centre, Faculty of Medical and Health Sciences, University of Auckland,
Auckland, New Zealand
2
Department of Ophthalmology, Wanganui Hospital, Wanganui, New Zealand
3
School of Health Sciences, Massey University, Auckland, New Zealand
*Corresponding author e-mail: elissa_mcdonald@hotmail.com
Background: There is lack of consensus from randomized
controlled trials on the efficacy of antivirals in the man-
agement of herpes zoster. Therefore, a systematic review
and meta-analysis was undertaken to provide better
understanding of effectiveness of antivirals in manage-
ment of herpes zoster.
Methods: A total of 12 randomized controlled trials
with 7,277 patients were included in the review. Trials
compared one antiviral to another (aciclovir, valaciclo-
vir, famciclovir or brivudin) for a minimum of 7days in
immunocompetent patients presenting with herpes zos-
ter diagnosed within 72 h of symptom onset. Primary
outcome was reduction in pain.
Results: Compared with aciclovir, valaciclovir showed sig-
nificant reduction in herpes-zoster-associated pain up to
112 days. The largest risk reduction in pain (36%) was seen
Introduction
Primary infection with varicella zoster virus (VZV; also
known as chickenpox) results in viral latency in dorsal
root, cranial and autonomic system ganglia [1]. Reac-
tivation of the varicella virus, due to decreased cellular
immunity, often associated with ageing, chemotherapy or
immunosuppression, results in herpes zoster (HZ) infec-
tion, also known as shingles [2]. VZV migrates from a
single ganglion to neural tissue of the affected segment
and the corresponding cutaneous dermatome [1]. The
migration and resultant neural inflammation result in
altered sensation and pain (prodrome) to the dermat-
ome, usually followed by development of fluid filled
lesions, shallow ulcers and crusts. Ninety percent of pri-
mary infection with VZV occurs in children under thir-
teen years of age. Latent VZV is usually suppressed by
2012 International Medical Press 1359-6535 (print) 2040-2058 (online)
AVT-11-RV-2024_McDonald.indd 255
at 2130 days (relative risk [RR] 0.64, 95% CI 0.59, 0.70)
with number needed to treat to benefit (NNT) of 3 (95%
CI 2.7, 3.8). Famciclovir was also superior to aciclovir with
a 46% reduction in risk of pain at 2830 days (RR 0.54,
95% CI 0.48, 0.68) with NNT of 3 (95% CI 2, 5). Time to
lesion healing and adverse effect profile was comparable.
Conclusions: Evidence from quality trials have shown
significant reduction in risk of pain with valaciclovir and
famciclovir for management of herpes zoster including
ophthalmicus. Valaciclovir or famciclovir should be pre-
ferred treatment options in patients with herpes zoster as
they both provide significant reduction in risk of herpes-
zoster-associated pain. Furthermore, the superior phar-
macokinetics and more convenient dosing regimens with
the use of valaciclovir and famciclovir clearly make them
the preferred treatment option.
a healthy immune system [2] although reactivation can
occur through decreased cellular immunity [2]. The virus
returns to the ganglion and latency. Viral multiplication
within ganglionic nerve cells causes acute inflammation
and neuronal necrosis, which, combined with neuritis
following viral migration down sensory nerves, results
in lasting pain known as post-herpetic neuralgia (PHN)
occurring in approximately 20% of patients [3]. If VZV
affects the ophthalmic division of the trigeminal nerve,
ocular involvement may occur causing ocular manifesta-
tions including conjunctivitis, epithelial or stromal kera-
titis and ocular pain [4]. If corneal scarring occurs, visual
loss may be irreversible.
Nucleoside analogues (for example, aciclovir [ACV],
valaciclovir [VACV], famciclovir [FAM] and brivudin
255
09/02/2012 10:26:08
 EM McDonald et al.
[BVDU]), are the mainstay of clinical treatment for HZ
and are most effective if started within 72h of prodro-
mal symptoms (for example, altered sensation, tingling
and rash) [4] and continued for a minimum of 7days.
ACV and FAM are nucleoside analogues that have
higher affinity for viral DNA binding sites compared
to viral deoxyguanosine [5]. Hence, ACV and FAM
disable viral DNA by preventing viral replication once
encoded in viral DNA. ACV is the standard treatment
for HZ. However, due to its poor intestinal absorption
ACV dosing needs to be high and frequent (800mg, 5
daily) in order to maintain inhibitory plasma concen-
trations. The prodrug of ACV is VACV, which is readily
absorbed from the gastrointestinal tract and metabo-
lized in the liver to produce ACV phosphate [6,7]. FAM
is a prodrug of penciclovir triphosphate, which is an
acyclic guanosine analogue with a longer intracellular
half-life than ACV or VACV [5] and, as a result, can be
given less frequently than ACV and in lower doses than
ACV or VACV, improving both patient compliance and
maintenance of steady-state plasma concentration.
ACV and FAM, when used as 7-day treatments,
improve patient outcomes (that is, pain reduction, shorter
duration of viral activation and decreased PHN [8,9].
VACV has been found to be comparable to both ACV
and FAM [1012], with a decrease in ocular complica-
tions (for example, conjunctivitis, epithelial or stromal
keratitis and ocular pain) also noted [13]. A systematic
review in 1995 on primary care management of acute HZ
[14] failed to identify significant benefit of ACV due to a
paucity of randomized controlled trials. A recent narra-
tive review on treatment of HZ suggested antiviral medi-
cation had marginal effects on relieving acute pain and
rate of lesion healing [15]. A retrospective study compar-
ing VACV, ACV and placebo for HZ found VACV signifi-
cantly reduced incidence of PHN compared to ACV [16].
In the past 15 years, many clinical trials comparing
antivirals against each other for HZ have been con-
ducted with varying results. Studies found advantages
of VACV over ACV for pain resolution, decreased inci-
dence of PHN, and proportion of patients with pain
6 months after cutaneous healing [10,17]. However,
other studies have reported no significant difference
between the two drugs except a simpler dosing schedule
[10,11]. There are also studies that have shown no clini-
cal difference between FAM and ACV in the treatment
of HZ [13,18]; however, one study reported a more
favourable adverse effect profile with FAM [19]. Fur-
thermore, a study by Tyring et al. [12] found no signifi-
cant difference between VACV and FAM, and another
[20] found BVDU to provide greater efficacy than ACV.
Although recommendations for the management
of HZ have been published [21], due to lack of con-
sensus from the many published trials on the efficacy
of antivirals in the management of HZ there are no
256
AVT-11-RV-2024_McDonald.indd 256
internationally accepted guidelines. Therefore, an up-
to-date systematic review was warranted to provide a
better understanding of the effectiveness of antivirals
in the management of HZ. This systematic review com-
pared the efficacy of ACV, VACV, FAM and BVDU in
the management of immunocompetent patients with
active HZ diagnosed within 72h of symptom onset.
Methods
A systematic review was conducted using Cochrane
Collaboration methodology and software from the
Nordic Cochrane Centre (Rigshospitalet, Copenhagen,
Denmark) [22] of randomized controlled trials in the
management of HZ (diagnosed within 72 h of symp-
tom onset) in immunocompetent patients >18 years of
age comparing one antiviral to another (ACV, VACV,
FAM or BVDU) treated for a minimum of 7days.
Primary outcome was the proportion of patients with
reduction in pain. Pain was defined as any degree of
HZ-associated dermal discomfort ranging from mild
discomfort to discomfort that prevents activities of
daily life. Secondary outcomes included rate of lesion
healing and adverse effects.
Search strategy and selection criteria
Trials were excluded if they were non-experimental
in design, compared antiviral with placebo, involved
patients who had acute retinal necrosis, progressive
outer retinal necrosis, or herpes simplex type I or II.
The following databases were searched for potential
studies with no language restrictions: Cochrane Central
Register of Controlled Trials (CENTRAL), MEDLINE,
EMBASE, BioMed Central, Trials Central, Clinical
Trials, Controlled Clinical Trials, Web of Science and
online relevant medical journal web sites. All data-
bases were searched from the date of inception until
May 2011. The reference lists of retrieved articles were
searched to identify potential studies. Authors of iden-
tified trials and pharmaceutical companies producing
antivirals were contacted for further published, unpub-
lished or ongoing studies.
The search of databases was initially conducted
by one reviewer (EMM), and two reviewers (EMM
and FSFR) independently selected trials for inclusion
and assessed all trials that appeared potentially rel-
evant. Full articles retrieved were obtained and trans-
lated where necessary. A third reviewer (JdK) would
have been invited to arbitrate if there was failure in
resolving disagreement between the first two review-
ers. Figure1 details selection of trials with reasons for
exclusions.
A five-point Jadad scale was used to assess meth-
odological quality of included studies [23]. Allocation
concealment was also assessed using the Cochrane
2012 International Medical Press
09/02/2012 10:26:08

Collaboration scale [24]. Table 1 lists both methodo-
logical quality scores.
All data was analysed on an intention-to-treat basis.
Mantel-Haenszel fixed-effect model analysis was used
for all dichotomous data [25,26]. Data was analysed
using weighted mean differences and 95% CIs for con-
tinuous outcomes and relative risks (RR) calculated for
dichotomous outcome with 95% CI. Where appropri-
ate, number needed to treat to benefit (NNT) and 95%
CI was calculated [27].
Results
Trial selection and quality
An electronic search yielded 1,740 abstracts, 18 trials
were identified as potentially suitable, 6 were excluded
and the remaining 12 [913,1720,2830] with points between 1 to 112 days. This risk reduction in
7,277 patients included. There was full agreement
Figure 1. Flow diagram of selected and excluded trials based on the Quorom statement
Potentially relevant RCTs identified
and screened for retrieval (n=2,054)
RTCs retrieved for more
detailed evaluation (n=18)
Potentially appropriate RTCs to be
included in meta-analysis (n=12)
RTCs included in
meta-analysis (n=12)
RCTs with usable information,
by outcome (n=12)
Pain n=9
Lesion healing n=7
Adverse effects n=9
Abnormal sensation n=2
Ocular complications n=2
RCT, randomized controlled trial.
Antiviral Therapy 17.2
AVT-11-RV-2024_McDonald.indd 257
Antivirals for herpes zoster
between two reviewers on exclusion and inclusion of
trials. Characteristics of included trials are shown in
Table 1. Included trials were of good methodologi-
cal quality with six trials scoring A and six scoring
B for Cochrane allocation concealment rating. In
addition, included trials had an average Jadad score
of 3.58, which is considered to be of good methodo-
logical quality [23]. Details for outcome comparisons
which included two or more trials are summarized in
Table2.
Reduction in pain primary outcome measure
VACV 1,000 mg three times daily was compared to
ACV 800mg five times daily in three studies [10,11,17]
with 927 patients (Figure2). Risk of pain was signifi-
cantly less with VACV compared to ACV for all time
pain ranged from 8% to 36% with VACV, with NNT
RTCs/abstracts excluded
as not relevant (n=2,036)
RCTs excluded with reasons (n=6)
Immunocompromised n=1
Patients did not have
herpes zoster n=2
Not an RCT n=3
RCTs excluded from
meta-analysis with reasons (n=0)
RCTs withdrawn, by outcome,
with reasons (n=0)
257
09/02/2012 10:26:09
 EM McDonald et al.

Table 1. Characteristics of included trials with methodological quality scores

Total number of Trial quality score
Reference participants in trial Population Interventions Outcome (Cochrane/Jadad)a

Beutner 1,141 Multinational study of patients VACV 1,000mg 3 daily VACV accelerates resolution of A/4
et al. [10] aged 50 years presenting with for 7 days; VACV 1,000mg pain, has simpler dosing regimen
HZ within 72h. Mean age 68 3 daily for 14 days; ACV and comparable safety profile to
years, 57% male, 43% female. 800mg 5 daily for 7 days. ACV.
Colin 110 French study of patients aged VACV 1,000mg 3 daily VACV is comparable to ACV in B/3
et al. [11] 18 years presenting with HZO for 7 days; ACV 800mg pain resolution, preventing ocular
within 72 h. Mean age 62 years, 5 daily for 7 days. complications and adverse event
48% male, 52% female. profile.
Degreef 545 Multinational study of patients FAM 250 mg 3 daily All three doses of FAM are B/3
[28] aged 50 years presenting with for 7 days; FAM 500 mg comparable to ACV for pain
HZ within 72 h. Mean age 51 3 daily for 7 days; FAM resolution, lesion healing and
years, 44% male, 56% female. 750 mg 3 daily for 7 days; adverse event profile.
ACV 800 mg 5 daily for
7 days.
Lin 57 Taiwanese study of patients VACV 1,000 mg 3 daily VACV accelerates resolution of B/2
et al. [17] aged 18 years presenting with for 7 days; ACV 800 mg pain, has simpler dosing regimen
HZ within 72 h. Mean age 63 5 daily for 7 days. and comparable safety profile to
years, 65% male, 35% female. ACV.
Shafran 559 Multinational study of patients FAM 250 mg 3 daily All three doses of FAM were B/3
et al. [18] aged 18 years presenting with for 7 days; FAM 500 mg comparable to ACV for lesion
HZ within 72 h. Mean age 55 twice daily for 7 days; FAM healing.
years, 42% male, 58% female. 750 mg 1 daily for 7 days;
ACV 800 mg 5 daily for
7 days.
Shen 55 Taiwanese study of patients FAM 250 mg 3 daily for FAM is comparable to ACV for B/3
et al. [19] aged 18 years presenting with 7 days; ACV 800 mg 5 lesion healing but has a more
HZ within 72 h. Mean age 60 daily for 7 days. favourable adverse event profile.
years, 65% male, 35% female.
Tyring 419 Multinational study of patients FAM 500 mg 3 daily for Both doses of FAM accelerated A/3
et al. [9] aged 18 years presenting with 7 days; FAM 750 mg 3 resolution of pain and have
HZ within 72 h. Mean age 50 daily for 7 days. similar adverse event profile.
years, 53% male, 47% female. No difference between FAM
doses.
Tyring 597 American study of patients VACV 1,000 mg 3 daily VACV is comparable to FAM for A/5
et al. [12] aged 50 years presenting with for 7 days; FAM 500 mg resolution of pain, lesion healing
HZ within 72 h. Mean age 69 3 daily for 7 days. and adverse event profile.
years, 37% male, 63% female.
Tyring 454 Multinational study of patients FAM 500 mg 3 daily for FAM is comparable to ACV for A/5
et al. [13] aged 18 years presenting with 7 days; ACV 800 mg 5 ocular manifestations.
HZO within 72 h. Mean age 58 daily for 7 days.
years, 47% male, 53% female.
Wassilew 1,225 Multinational study of patients BVDU 125 mg 1 daily for BVDU accelerates lesion healing A/5
et al. [20] aged 18 years presenting with 7 days; ACV 800 mg 5 compared to ACV whilst having
HZ within 72 h. Mean age 53 daily for 7 days. comparable adverse event profile.
years, 44% male, 56% female.
Wassilew 2,027 Multinational study of patients BVDU 125 mg 1 daily BVDU is comparable to FAM for A/5
et al. [29] aged 50 years presenting with for 7 days; FAM 250 mg resolution of pain, lesion healing
HZ within 72 h. Mean age 63 3 daily for 7 days. and adverse event profile.
years, 40% male, 60% female.
Xu 88 Chinese study of patients aged FAM 250 mg 3 daily FAM accelerates resolution of B/2
et al. [30] 18 years presenting with HZ for 7 days; ACV 200 mg pain and shortens length of
within 72 h. Mean age and 5 daily for 7 days. inflammation compared to ACV.
gender ratio not stated in
translation.
a
Cochrane study quality score (A= adequate, B= unclear, C= inadequate and D= unused); Jadad trial quality score (05). ACV, aciclovir; BVDU, brivudin; FAM,
famciclovir; HZ, herpes zoster; HZO, herpes zoster ophthalmicus; VACV, valaciclovir.

258 2012 International Medical Press

AVT-11-RV-2024_McDonald.indd 258 09/02/2012 10:26:09

 Antivirals for herpes zoster

Table 2. Outcomes with two or more included studies comparing antivirals for herpes zoster
Comparisons/outcomes Number of studies (subjects) RR (95% CI) P-value NNT (95% CI)

VACV 1,000 mg 3 daily versus ACV 800mg 5 daily

Pain
At presentation 3 (927) 1.01 (0.99, 1.03) 0.41
110 Days 3 (927) 0.92 (0.88, 0.97) 0.004a 14 (9, 50)
1120 Days 3 (927) 0.91 (0.84, 0.98) 0.02a 14 (8, 100)
2130 Days 3 (927) 0.64 (0.59, 0.70) <0.00001a 3 (2.7, 3.8)
3160 Days 2 (870) 0.85 (0.73, 0.99) 0.04a 14 (8, 100)
61112 Days 2 (870) 0.79 (0.63, 0.98) 0.03a 17 (8, 100)
113170 Days 2 (870) 0.81 (0.62, 1.05) 0.11
FAM 250, 500, 750mg 3 daily versus ACV 800mg 5 daily
Pain
FAM 250mg, 2830 days 2 (328) 0.54 (0.43, 0.68) <0.00001a 3 (2, 5)
Loss of vesicles
FAM 250 mg versus ACV 800 mg, 57 days 3 (605) 0.97 (0.93, 1.02) 0.31
FAM 500 mg versus ACV 800 mg, 56 days 2 (553) 0.98 (0.93, 1.03) 0.38
FAM 750 mg versus ACV 800 mg, 57 days 2 (555) 0.99 (0.95, 1.04) 0.74
Cessation of new lesions
FAM 250 mg versus ACV 800 mg, 3 days 2 (550) 0.89 (0.76, 1.04) 0.15
FAM 500 mg versus ACV 800 mg, 3 days 2 (553) 0.93 (0.80, 1.09) 0.37
FAM 750 mg versus ACV 800 mg, 3 days 2 (555) 0.95 (0.81, 1.10) 0.47
Full crusting
FAM 250 mg versus ACV 800 mg, 611 days 3 (605) 1.04 (0.97, 1.10) 0.26
FAM 500 mg versus ACV 800 mg, 67 days 2 (553) 0.99 (0.92, 1.06) 0.69
FAM 750 mg versus ACV 800 mg, 68 days 2 (555) 1.03 (0.97, 1.10) 0.37
Loss of crusts
FAM 250 mg versus ACV 800 mg, 2027 days 3 (605) 1.02 (0.97, 1.06) 0.45
FAM 500 mg versus ACV 800 mg, 1921 days 2 (553) 0.99 (0.95, 1.04) 0.73
FAM 750 mg versus ACV 800 mg, 2021 days 2 (555) 1.00 (0.96, 1.05) 0.86
FAM 500 mg 3 daily versus FAM 750 mg 3 daily
Pain
10 Days 2 (545) 0.83 (0.76, 0.91) <0.0001a 7 (5, 13)
20 Days 2 (545) 0.95 (0.83, 1.09) 0.42
30 Days 2 (545) 0.94 (0.81, 1.09) 0.40
60 Days 2 (545) 1.03 (0.85, 1.26) 0.74
110 Days 2 (545) 0.88 (0.63, 1.21) 0.43

a
Statistically significant outcome. ACV, aciclovir; FAM, famciclovir; NNT, number needed to treat to benefit (computed only for significant outcomes); RR, relative risk;
VACV, valaciclovir.

ranging from 3 to 17 (Table 2). There was no further
reduction in risk of pain from 113 to 170 days and
there was no data available beyond 170 days. Colin et
al. [11] compared VACV to ACV in patients with HZ
ophthalmicus (HZO) and found no difference in the
incidence of ocular complications or use of analgesics
(topical or systemic) for ocular pain.
FAM 250mg three times daily was compared to ACV
800mg five times daily in two studies [19,28] with 328
patients (Figure3). Risk of pain was reduced by 46%
(RR 0.54, 95% CI 0.43, 0.68) compared to ACV at 28
to 30 days post-treatment, with an NNT of 3 (95% CI
2, 5; Table 2). Three studies [18,19,28] comparing FAM
to ACV found no difference in adverse effects or lesion
healing.
Vesicles, lesions, crusts and adverse effects
There was no difference in median time to loss of vesi-
cles, cessation of new lesions, full crusting or loss of
crusts when VACV or FAM was compared to ACV at
any of the doses used in the included trials (Table 2).
There was no difference in risk of the adverse effects
when VACV or FAM was compared to ACV (Table 3).
Two studies with 545 patients [9,28] compared FAM
500mg three times daily to FAM 750mg three times
daily with the 500mg dose showing significantly greater
reduction in risk of pain by 17% (RR 0.83, 95% CI
0.76, 0.91, with an NNT of 7) at 10 days compared to
the 750mg dose (Figure4 and Table 2). Adverse effects
profiles were similar between the two doses. Degreef
[28] also showed that when the three doses (250, 500

Antiviral Therapy 17.2 259

AVT-11-RV-2024_McDonald.indd 259 09/02/2012 10:26:09

 EM McDonald et al.

Figure 2. Details of pain outcome for valaciclovir 1,000mg three times daily versus aciclovir 800mg five times daily

Study or subcategory VACV, n/N ACV, n/N RR (fixed) 95% CI Weight, % RR (fixed) 95% CI

01 110 days
Beutner 1995 334/384 346/376 86.40 0.95 (0.90, 0.99)
Colin 2000 24/56 32/54 8.05 0.72 (0.50, 1.05)
Lin 2001 23/32 20/25 5.55 0.90 (0.67, 1.20)
Subtotal (95% CI) 472 455 100.00 0.92 (0.88, 0.97)
Total events:
381 (VACV), 398 (ACV)
Test for heterogeneity:
2=2.46, df=2 (P=0.29)
Test for overall effect:
Z=2.91 (P=0.004)

02 1120 days
Beutner 1995 288/384 301/376 87.00 0.94 (0.87, 1.01)
Colin 2000 19/56 27/54 7.86 0.68 (0.43, 1.07)
Lin 2001 17/32 16/25 5.14 0.83 (0.54, 1.29)
Subtotal (95% CI) 472 455 100.00 0.91 (0.84, 0.98)
Total events:
324 (VACV), 344 (ACV)
Test for heterogeneity:
2=2.31, df=2 (P=0.31)
Test for overall effect:
Z=2.36 (P=0.02)

03 2130 days
Beutner 1995 238/384 363/376 92.00 0.64 (0.59, 0.70)
Colin 2000 14/56 17/54 4.34 0.79 (0.44, 1.45)
Lin 2001 9/32 13/25 3.66 0.54 (0.28, 1.06)
Subtotal (95% CI) 472 455 100.00 0.64 (0.59, 0.70)
Total events:
261 (VACV), 393 (ACV)
Test for heterogeneity:
2=0.74, df=2 (P=0.69)
Test for overall effect:
Z=10.41 (P<0.00001)

04 3160 days
Beutner 1995 165/384 188/376 94.91 0.86 (0.74, 1.00)
Colin 2000 8/56 10/54 5.09 0.77 (0.33, 1.81)
Subtotal (95% CI) 440 430 100.00 0.85 (0.73, 0.99)
Total events:
173 (VACV), 198 (ACV)
Test for heterogeneity:
2=0.06, df=1 (P=0.81)
Test for overall effect:
Z=2.03 (P=0.04)

05 61112 days
Beutner 1995 100/384 120/376 93.71 0.82 (0.65, 1.02)
Colin 2000 3/56 8/54 6.29 0.36 (0.10, 1.29)
Subtotal (95% CI) 440 430 100.00 0.79 (0.63, 0.98)
Total events:
103 (VACV), 128 (ACV)
Test for heterogeneity:
2=1.53, df=1 (P=0.22)
Test for overall effect:
Z=2.12 (P=0.03)

0.5 0.7 1 1.5 2

Less with VACV Less with ACV

The mean value for each trial is indicated by a square box with the line through it representing the 95% CI. The size of the square represents the weight that the
corresponding study exerts in the meta-analysis; this is the MantelHaenszel weight. Mean values left of the vertical line of no effect (value of 1) favour less risk of pain
with valaciclovir (VACV) and values on the right favour less risk of pain with aciclovir (ACV). The solid diamond indicates the overall mean effect VACV has on pain with
the lateral points indicating confidence intervals of this estimate. A percentage weighting, which is dependent on the precision and sample size of the estimation of
the mean value for each randomized controlled trial, is allocated to each study. The c2 and the degrees of freedom (df) values for each time point provide a measure of
heterogeneity of the results. The Z statistic indicates the level of significance for the overall result. P-values 0.05 are considered statistically significant. RR, relative risk.

260 2012 International Medical Press

AVT-11-RV-2024_McDonald.indd 260 09/02/2012 10:26:10

 Antivirals for herpes zoster

Figure 3. Famciclovir 250mg three times daily reduces risk of pain versus aciclovir 800mg five times daily at 2830 days

Study or subcategory FAM, n/N ACV, n/N RR (fixed) 95% CI Weight, % RR (fixed) 95% CI

01 2830 days
Degreef 1994 47/134 96/139 84.95 0.51 (0.39, 0.66)
Shen 2004 12/27 17/28 15.05 0.73 (0.44, 1.23)
Subtotal (95% CI) 161 167 100.00 0.54 (0.43, 0.68)
Total events:
59 (FAM), 113 (ACV)
Test for heterogeneity:
2=1.55, df=1 (P=0.21)
Test for overall effect:
Z=5.25 (P<0.00001)

0.2 0.5 1 2 5
Less with FAM Less with ACV

The mean value for each trial is indicated by a square box with the line through it representing the 95% CI. The size of the square represents the weight that the
corresponding study exerts in the meta-analysis. The solid diamond indicates the overall mean effect FAM has on pain with the lateral points indicating confidence
intervals of this estimate. ACV, aciclovir; df, degrees of freedom; FAM, famciclovir; RR, relative risk.

Table 3. Valaciclovir, aciclovir and famciclovir adverse effects for outcomes with two or more included studies
Comparisons/outcomes Number of studies (subjects) RR (95% CI) P-value

VACV 1,000 mg 3 daily versus ACV 800 mg 5 daily

Vomiting 2 (870) 1.38 (0.89, 2.12) 0.15
Diarrhoea 2 (817) 0.83 (0.51, 1.34) 0.44
Constipation 2 (817) 1.01 (0.57, 1.81) 0.96
Dizziness 2 (817) 0.70 (0.39, 1.26) 0.24
FAM 250 mg 3 daily versus ACV 800 mg 5 daily
Headache 2 (332) 1.25 (0.72, 2.17) 0.43
Constipation 2 (332) 0.83 (0.37, 1.88) 0.66
FAM 500 mg 3 daily versus FAM 750 mg 3 daily
Headache 2 (555) 1.02 (0.73, 1.43) 0.91
Nausea 2 (555) 1.08 (0.68, 1.72) 0.75

ACV, aciclovir; FAM, famciclovir; RR, relative risk; VACV, valaciclovir.

Figure 4. Famciclovir 500mg three times daily reduces risk of pain versus FAM 750mg three times daily at 10 days

FAM FAM
Study or subcategory 500 mg, n/N 750 mg, n/N RR (fixed) 95% CI Weight, % RR (fixed) 95% CI

01 10 days
Degreef 1994 80/134 108/138 45.51 0.76 (0.65, 0.90)
Tyring 1995 115/138 126/135 54.49 0.89 (0.82, 0.97)
Subtotal (95% CI) 272 273 100.00 0.83 (0.76, 0.91)
Total events:
195 (FAM 500 mg), 234 (FAM 750 mg)
Test for heterogeneity:
2=3.49, df=1 (P=0.06)
Test for overall effect:
Z=4.10 (P<0.0001)

0.2 0.5 1 2 5
Less with Less with
500 mg 750 mg

The mean value for each trial is indicated by a square box with the line through it representing the 95% CI. The size of the square represents the weight that the
corresponding study exerts in the meta-analysis. The solid diamond indicates the overall mean effect FAM 500 mg has on pain with the lateral points indicating
confidence intervals of this estimate. df, degrees of freedom; FAM, famciclovir; RR, relative risk.

Antiviral Therapy 17.2 261

AVT-11-RV-2024_McDonald.indd 261 09/02/2012 10:26:11

 EM McDonald et al.
and 750mg three times daily) of FAM were compared
they all provided similar efficacy in reducing pain and
lesion healing, with the exception of greater reduction
in risk of pain at 10 days with FAM 500 mg, when
compared to 750mg.
Outcomes with a single included trial
A single study with 273 patients [28] demonstrated
significant reduction in risk of pain with three doses of
FAM (250, 500, 750mg three times daily) from 1170
days when compared to ACV (800mg five times daily).
The following comparisons were also included in sin-
gle trials: VACV 1,000 mg versus FAM 500 mg [12];
FAM 250mg versus ACV 200mg [30]; FAM 250mg no pain to 3= severe pain), and six-point pain scales
compared to FAM 500 mg and FAM 750 mg [28];
BVDU 125mg compared to ACV 800mg [20] and FAM
250mg [29]. Meta-analysis of a single trial is nonsensi-
cal and uninterpretable; therefore these comparisons are
not discussed in detail.
Discussion
This is the first study to show significant reduction in
risk of HZ-associated pain with VACV and FAM com-
pared to ACV. The three antivirals did not differ regard-
ing time to lesion healing, cessation of new lesions, full
crusting or loss of crusts at any of the doses used in the
trials and all antivirals showed comparable safety pro-
file. It is interesting to note that although VACV pro-
vides significant reduction in risk of pain compared to
ACV, there is no difference in the rate of skin lesion
healing. It is possible that VACV is present in greater
concentrations in the dorsal root ganglion compared to
ACV. Furthermore, as there was no difference in the rate
of skin lesion healing it is plausible that there was no
difference in drug concentration at the skin. However,
this pharmacological difference needs to be evaluated in
future studies.
Recommendations published on the management of
HZ [21] are in agreement with the results of this sys-
tematic review, preferring VACV and FAM over ACV.
VACV and FAM provide higher plasma concentrations
of acyclic nucleoside analogue required for inhibition
of VZV, reducing acute viral shedding and minimizing
neural damage [21]. As the dosing schedule for VACV
and FAM is three times daily as opposed to ACV five
times daily, patient compliance with dosing regimes is
likely to be higher.
There has been ongoing debate as to the best way
to measure PHN. Studies have used differing methods
such as HZ-associated pain, and cross-sectional com-
parisons at times varying from 30120 days after onset
of pain. The Shingles Prevention Study (SPS) [31] used
90 days to define PHN but also did comparisons at 30,
60 and 120days. We have used a similar comparison
262
AVT-11-RV-2024_McDonald.indd 262
to the SPS measuring pain at varying times from 1170
days but did not use any specific time to define PHN.
Due to the changing definition of PHN over the years,
our approach was to accept the inclusion of subjects
with HZ-associated pain, from presentation to study
completion. Pain as an outcome was stratified in some
studies to include degrees of pain both during the acute
phase and after cutaneous healing, whereas in other
studies only presence or absence of pain was described
and pain was seen as a continuum from presentation to
resolution. Included studies assessed pain in a variety of
ways including Gracely scales (0= no pain to 5= pain
that requires rest or bed rest), four-point 03 scales (0=
(0= no pain to 5= unbearable). Although there is gen-
eral consensus that antiviral zoster trials, particularly
those that measure PHN, utilize pain severity scores,
we chose to use the presence or absence of pain as our
primary outcome measure to enable comparability
between studies and utilized comparable time points
without defining post-rash status. The effect of using
this definition as our outcome measure would under-
estimate treatment effect as participants with mild pain
were included in the pain group.
In any systematic review it may be possible that not
all studies have been identified. However, a comprehen-
sive search for published and unpublished studies was
performed to decrease the chance of missing any poten-
tial studies. No language restriction was applied and in
studies with insufficient data, authors were contacted
for further details.
A single trial conducted on 454 patients with HZO
demonstrated no difference in the incidence of ocular
manifestations between FAM (500mg three times daily)
compared to ACV (800mg five times daily) indicating
no increased risk of ocular manifestations when FAM is
used for HZO [13]. Another study in 88patients [30]
showed reduction in time to cessation of acute pain, by
3 days with FAM (250mg three times daily) compared
to ACV (200mg five times daily). This study [30] also
demonstrated that lesion healing occurred a day earlier
with FAM (250mg three times daily) compared to ACV
(200 mg five times daily) and loss of crusts occurred
3 days earlier with FAM (weighted mean difference
2.93 days, 95% CI 1.02, 4.84). It is possible, however
that the subtherapeutic dose (200mg five times daily) of
ACV did not provide adequate plasma concentrations to
inhibit VZV, thus skewing the results in favour of FAM.
A recent systematic review [32] investigating antivi-
ral treatment for preventing PHN stated ACV provided
significant reduction in risk of PHN 1 month after
onset of acute herpetic rash. This result, however, was
heterogeneous (P=0.01, I2=72%) and therefore a valid
conclusion cannot be derived. The authors did not find
significant difference between ACV and control groups
2012 International Medical Press
09/02/2012 10:26:12

in the incidence of PHN at 4 or 6 months following
acute herpetic rash. However, they have stated that a
limitation of their review is the lack of quality clinical
trials hampering the robustness of their conclusions and
found insufficient evidence from included trials to sup-
port the use of other antiviral agents in preventing PHN.
Our review of high-quality randomized controlled trials
addresses this gap in evidence by showing up to 36%
reduction in risk of pain with VACV and 46% reduction
in risk of pain with FAM.
FAM has been shown to be more effective than ACV
in reducing risk of pain with comparable lesion heal-
ing and safety profile. FAM 500 mg three times daily
has been shown to significantly reduce risk of pain at
10 days compared to FAM 750 mg three times daily
with no difference in adverse effects. A single study
comparing FAM 250 mg to 750 mg, and 250 mg to
500 mg found all doses to be comparable for pain,
lesion healing and adverse effects. HZO was excluded
from all trials comparing FAM to ACV for pain, lesion
healing and adverse effects. Therefore, the results of this
comparison need to be used with caution in patients
presenting with trigeminal first nerve involvement.
Further studies are required to confirm these findings
and, until such time, the lowest effective dose of FAM
should be used in the management of HZ. Further stud-
ies are required comparing FAM and ACV beyond 30
days for reduction in risk of pain, lesion healing and
adverse effects, particularly in patients with HZO.
Due to the greater clinical effectiveness of VACV
and FAM, as evidenced by significant reduction in risk
of pain, superior pharmacokinetics and easier dosing
schedule compared to ACV, it is recommended that
VACV and FAM be the treatments of choice in the man-
agement of HZ.
Acknowledgements
We would like to thank authors of included studies who
kindly responded to requests for further information.
The principal investigator had full access to all the data
in the study and takes responsibility for the integrity of
the data and the accuracy of the data analysis. No fund-
ing was received for this review.
Disclosure statement
The authors declare no competing interests.
References
1. Gnann JW, Whitley RJ. Herpes zoster. N Engl J Med 2002;
347:340346.
2. Gilden DH, Mahalingam R, Cohrs RJ, Tyler KL.
Herpesvirus infections of the nervous system. Nat Clin Pract
Neurol 2007; 3:8294.
Antiviral Therapy 17.2
AVT-11-RV-2024_McDonald.indd 263
Antivirals for herpes zoster
3. Wu CL, Marsh A, Dworkin RH. The role of sympathetic
nerve blocks in herpes zoster and postherpetic neuralgia.
Pain 2000; 87:121129.
4. Bartlett J, Jaanus S. Clinical ocular pharmacology. 5th ed.
StLouis, MO: Butterworth Heinemann Elsevier 2008.
5. Chakrabarty A, Tyring SK, Beutner K, Rauser M. Recent
clinical experience with famciclovir a third generation
nucleoside prodrug. Antivir Chem Chemother 2004;
15:251253.
6. Soul-Lawton J, Seaber E, On N, Wootton R, Rolan P,
PosnerJ. Absolute bioavailability and metabolic disposition
of valaciclovir, the l-valyl ester of acyclovir, following oral
administration to humans. Antimicrob Agents Chemother
1995; 39:27592764.
7. Wang LH, Schultz M, Weller S, Smiley ML, Blum MR.
Pharmacokinetics and safety of multiple-dose valaciclovir in
geriatric volunteers with and without concomitant diuretic
therapy. Antimicrob Agents Chemother 1996; 40:8085.
8. Huff JC, Drucker JL, Clemmer A, et al. Effect of oral
acyclovir on pain resolution in herpes zoster: a reanalysis.
JMed Virol 1993; 41:Suppl 1:9396.
9. Tyring S, Barbarash RA, Nahlik JE, et al. Famciclovir for
the treatment of acute herpes zoster: effects on acute disease
and postherpetic neuralgia. A randomized, double-blind,
placebo-controlled trial. Ann Intern Med 1995; 123:8996.
10. Beutner KR, Friedman DJ, Forszpaniak C, AndersenPL,
Wood MJ. Valaciclovir compared with acyclovir for
improved therapy for herpes zoster in immunocompetent
adults. Antimicrob Agents Chemother 1995; 39:15461553.
11. Colin J, Prisant O, Cochener B, Lescale O, Rolland B,
Hoang-Xuan T. Comparison of the efficacy and safety of
valaciclovir and acyclovir for the treatment of herpes zoster
ophthalmicus. Ophthalmology 2000; 107:15071511.
12. Tyring SK, Beutner KR, Tucker BA, Anderson WC,
CrooksRJ. Antiviral therapy for herpes zoster: randomized,
controlled clinical trial of valacyclovir and famciclovir
therapy in immunocompetent patients 50 years and older.
Arch Fam Med 2000; 9:863869.
13. Tyring S, Engst R, Corriveau C, et al. Famciclovir for
ophthalmic zoster: a randomised aciclovir controlled study.
Br J Ophthalmol 2001; 85:576581.
14. Lancaster T, Silagy C, Gray S. Primary care management
of acute herpes zoster: systematic review of evidence from
randomized controlled trials. Br J Gen Pract 1995; 45:3945.
15. Opstelten W, Eekhof J, Neven AK, Verheij T. Treatment of
herpes zoster. Can Fam Physician 2008; 54:373377.
16. Goh CL, Khoo L. A retrospective study on the clinical
outcome of herpes zoster in patients treated with acyclovir
or valaciclovir vs patients not treated with antiviral. Int J
Dermatol 1998; 37:544546.
17. Lin WR, Lin HH, Lee SS, et al. Comparative study of the
efficacy and safety of valaciclovir versus acyclovir in the
treatment of herpes zoster. J Microbiol Immunol Infect
2001; 34:138142.
18. Shafran SD, Tyring SK, Ashton R, et al. Once, twice, or three
times daily famciclovir compared with aciclovir for the oral
treatment of herpes zoster in immunocompetent adults: a
randomized, multicenter, double-blind clinical trial. J Clin
Virol 2004; 29:248253.
19. Shen MC, Lin HH, Lee SS, Chen YS, Chiang PC, Liu YC.
Double-blind, randomized, acyclovir-controlled, parallel-
group trial comparing the safety and efficacy of famciclovir
and acyclovir in patients with uncomplicated herpes zoster.
J Microbiol Immunol Infect 2004; 37:7581.
20. Wassilew SW, Wutzler P. Oral brivudin in comparison
with acyclovir for improved therapy of herpes zoster in
immunocompetent patients: results of a randomized,
double-blind, multicentered study. Antiviral Res 2003;
59:4956.
21. Dworkin RH, Johnson RW, Breuer J, et al.
Recommendations for the management of herpes zoster.
Clin Infect Dis 2007; 44 Suppl 1:S1S26.
263
09/02/2012 10:26:12
 EM McDonald et al.
22. Review Manager (RevMan). Version 4.3 for Windows.
Copenhagen: Nordic Cochrane Centre 2006.
23. Jadad AR, Moore RA, Carroll D, et al. Assessing the
quality of reports of randomized clinical trials: is blinding
necessary? Control Clin Trials 1996; 17:112.
24. Higgins J, Green S (Editors). Cochrane handbook for
systematic reviews of interventions 4.2.5. Chichester, UK:
John Wiley & Sons Ltd 2005.
25. Laird NM, Mosteller F. Some statistical methods for
combining experimental results. Int J Technol Assess Health
Care 1990; 6:530.
26. Mantel N, Haenszel W. Statistical aspects of the analysis
of data from retrospective studies of disease. J Natl Cancer
Inst 1959; 22:719748.
27. Altman DG, Andersen P. Calculating the number needed to
treat for trials where the outcome is time to an event. BMJ
1999; 319:14921495.
28. Degreef H. Famciclovir, a new oral antiherpes drug: results
of the first controlled clinical study demonstrating its
efficacy and safety in the treatment of uncomplicated herpes
zoster in immunocompetent patients. Int J Antimicrob
Agents 1994; 4:241246.
Accepted 20 June 2011; published online 15 December 2011
264
AVT-11-RV-2024_McDonald.indd 264
29. Wassilew S. Brivudin compared with famciclovir in the
treatment of herpes zoster: effects in acute disease and
chronic pain in immunocompetent patients. A randomized,
double-blind, multinational study. J Eur Acad Dermatol
Venereol 2005; 19:4755.
30. Xu HZ, Mao LE, Luo BG, Zhou SX, Bian ZP. A randomised
controlled trial of famciclovir versus acyclovir in the
treatment of herpes zoster. Journal of Clinical Dermatology.
2000; 29:288289.
31. Oxman MN, Levin MJ. Vaccination against herpes zoster
and postherpetic neuralgia. J Infect Dis 2008; 197 Suppl
2:S228S236.
32. Li Q, Chen N, Yang J, et al. Antiviral treatment for
preventing postherpetic neuralgia. Cochrane Database Syst
Rev 2009; 2:CD006866.
2012 International Medical Press
09/02/2012 10:26:12