Prevention of post-herpetic neuralgia: acyclovir and

Acta Anaesthesiol Scand 2000; 44: 910–918
Printed in Denmark. All rights reserved
Copyright C Acta Anaesthesiol Scand 2000
ISSN 0001-5172
Prevention of post-herpetic neuralgia: acyclovir and
prednisolone versus epidural local anesthetic and
Department of Anesthesioloy and Intensive Care, University of Udine, 3Department of Dermatology, Hospital of Udine, 2Institute of Anesthesiology,
Intensive Care and Pain Clinic, University of Perugia and 4Division of Medical Oncology, Hospital of Perugia, Italy
Background: Treatment of herpes zoster (HZ) includes the use
of acyclovir with or without steroids. An alternative therapy is
the epidural administration of local anesthetics with or without
steroids. This trial compared the efficacy of these two treatment
regimens in the prevention of post-herpetic neuralgia (PHN).
Methods: Six hundred adults over 55 years of age with a rash
of less than 7 days duration, and severe pain due to HZ, were
enrolled and randomized to receive either intravenous acyclovir
(10 mg/kg three times daily) for 9 daysπprednisolone (60 mg
per day with progressive reduction) for 21 days, or 6–12 ml
bupivacaine (0.25%) every 6–8 or 12 hπmethylprednisolone 40
mg every 3–4 days by epidural catheter during a period ranging
from 7 to 21 days. Efficacy was evaluated at 1, 3, 6 and 12
months. PHN was assessed as pain and/or allodynia, and ‘‘abnormal sensations’’ (hypoesthesia, burning, itching, etc.). Statistical analysis was performed based on the intent-to-treat population.
Results: In the 485 patients who completed the study, the inci-
dence of pain after 1 year was 22.2% (51 patients of 230) after
acyclovir π steroids, and 1.6% (4 patients of 255) after epidural
analgesiaπsteroids. The incidence of abnormal sensations was
12.2% (28 patients) after acyclovirπsteroids, and 4.3% (11 patients) in group B.
Conclusions: Epidural administration of local anesthetic and
methylprednisolone is significantly more effective in preventing
PHN at 12 months compared to intravenous acyclovir and prednisolone.
aging. In fact, not only does this therapy have a favorable effect on resolving the acute phase, but it also
appears to be capable of preventing the onset of PHN
(7–9, 11, 13–16). However, there are no well-designed
randomized and controlled studies with both longterm follow-up and assessment of pain intensity.
There is also no current rational explanation for the
favorable results observed with this treatment regimen. Therefore, this therapy has been considered
strictly empirical and not entirely reliable (3, 5, 17).
The difficulties in verifying the validity of the use
of either acyclovirπsteroids or epidural anestheticπ
steroids are also due to lack of standardization of concepts.
There is a lack of agreement on the definition of
PHN. It has been defined as pain persisting beyond
the crusting of lesions (1), or lasting more than 1 (18),
2 (19), 3 (20) or 6 months (21) after the acute infection.
neuralgia (PHN) is observed in 9–
45% of all cases of herpes zoster (HZ) and the
incidence has been reported to be as high as 50–60%
among elderly or immunosuppressed patients (1–6).
A consensus on treatment has not been reached (1–
12), but there is general agreement for the need of
early therapy that can not only treat the acute infection (2, 8, 10) but can also prevent the onset of PHN
(5, 8, 11, 13, 14).
Acyclovir is the current treatment of choice for HZ
and may be used with or without concomitant steroid
administration. It has been found to be useful in treating the acute phase if used early enough (2, 3, 5, 10,
12), but the data on its efficiency in the prevention of
PHN remain equivocal (5, 6).
Only a few studies have discussed the use of local
anesthetics administered epidurally, either with or
without steroids, but the results appear to be encourOST-HERPETIC
Received 19 April 1999, accepted for publication 24 February 2000
Key words: Acyclovir; dorsal horn; epidural route; herpes zoster; local anesthetic; post-herpetic neuralgia; prednisolone;
spinal nerve root; steroids; varicella zoster virus.
c Acta Anaesthesiologica Scandinavica 44 (2000)
Treatment for prevention of post-herpetic neuralgia
Further, only a few researchers have studied patients
with HZ in the acute phase and followed the development of HZ into PHN over a duration of 6 months (7,
10) or more (19).
Based on the literature (17, 22–24), the acute phase
can develop into five categories:
i) Constant, spontaneous, and generally deep or cutaneous burning pain.
ii) A brief recurrent, piercing or electrical-shocktype pain often described as a shooting, tic-like pain.
iii) A sharp, radiating, dysesthetic sensation of
burning pain generally evoked by a very light tactile
response, known as tactile allodynia (pain from nonnoxious stimuli). A state of hyperalgesia superimposed over the continuous component of the pain
may also be present.
iv) Reduced sensitivity and/or the appearance of
abnormal sensations in the involved area. Burning,
aching or itching, prickling, and other sensations that
vary in intensity may also be present either spontaneously or provoked by touch or movement. Concurrently, the skin may also be insensitive and tender.
v) Complete recovery.
These clinically relevant outcomes may be grouped
as ‘‘pain’’ (i, ii, iii), ‘‘abnormal sensations’’ (iv) and
‘‘complete recovery’’ (v). However, clinical studies
have made little use of the distinctions among the
above symptoms, and patients are generally divided
into two groups, those with PHN, and those without.
Consequently, it is not clear whether the patients with
abnormal sensations are included in the group with
pain or in the complete recovery group.
The objective of the present study was to compare
the 1-year efficacy of acyclovirπsteroids with epidurally administered anesthetics π steroids, taking
into account pain and abnormal sensations as separate endpoints. Through a 1-year follow-up, we
attempted to define when patients should be considered as having PHN, as well as the true incidence
and intensity of pain and abnormal sensations. Based
on this study and a review of the literature, we propose a hypothesis to explain our observations.
Patients and methods
This randomized prospective trial was performed at
two study centers on 600 patients between February
1992 and June 1995. A design with two randomized
samples of different dimension (340 and 260 for the
two centers) was used. The study protocol was approved by the ethics committees of both institutions
and an informed consent was obtained from each patient.
Patients were enrolled if they were over 55 years
with acute herpes zoster, had a skin rash of less than
7 days duration, and a pain score ⬎7 on a 10-cm visual analog scale (VAS). Exclusion criteria included
kidney failure, liver damage, peptic ulcer, generalized
or skin infections, alterations in blood-clotting tests,
and previous systemic treatment with antiviral
agents, steroids, local anesthetics, capsaicin or acetylsalicylic acid.
The patients were randomized to one of the following treatment groups:
Group A: intravenous acyclovir (10 mg/kg) three
times a day, from day 0 to day 9, plus intravenous
prednisolone 60 mg/d from day 0 to day 8 with a
change to oral prednisolone administration and progressive reduction as follows: 30 mg/d from day 9 to
day 13, 15 mg/d from day 14 to day 18, and 7.5 mg/
d from day 19 to day 21. Patients were hospitalized
during the intravenous treatment.
Group B: an epidural catheter was positioned and
taped at the involved spinal segments (correct placement of the catheter was verified by direct fluoroscopy). When HZ was localized in the trigeminal area,
the catheter was placed at C5–C6 and inserted upwards
for approximately 3–4 cm. Bupivacaine (0.25%) with
epinephrine 1:200 000 in a volume of 6–12 ml (the volume was determined on the basis of the spinal segments involved) was administered through the catheter every 6–8 or 12 h (timing to ensure pain-free
periods of 24 h). Methylprednisolone acetate retard 40
mg was also administered through the catheter every
3–4 days for 7 consecutive days. After a 24-h break during which the catheter was left in place, this cycle of
therapy was repeated for an additional 7-day period if
pain persisted. Using the same procedure, the therapeutic cycle was repeated a third time, if needed, and
the epidural catheter was removed at the end of the
third cycle. The injection of the local anesthetic was always followed by verification of complete coverage of
the dermatomes involved by assessing the disappearance of pain. If coverage was not complete, additional
2-ml doses were injected until this objective was
achieved. Patients were only hospitalized for the initial
24-h period.
Patients were asked to come to the hospital for subsequent doses. Hovewer, the doses were often administered by the family doctor or by the patient’s
relatives after appropriate instruction. These patients
were controlled every 24–48 h until the end of the
Blood samples were obtained for biochemical and
hematological monitoring before starting treatment
and again on days 3–9 and 21.
A. Pasqualucci et al.
Follow-up examinations were performed at 1, 3, 6
and 12 months after the end of the treatment. During
each follow-up examination, the patients were asked
to categorize and assess the intensity of their symptoms using the following parameters:
1. Presence of ‘‘pain’’ (see ‘‘introduction’’: patterns
i, ii, iii) and its intensity, as measured on a 10-cm VAS.
These patients demonstrated spontaneous, continuous or intermittent pain and/or allodynia (provoked
or otherwise (see below)).
2. Presence of ‘‘abnormal sensations’’ (see ‘‘introduction’’: pattern iv). These patients presented with
hypoesthesia, itching, prickling, burning sensations
and so on. The intensity of these symptoms was measured using a 4-point verbal rating scale (VRS): 1Ω
light, 2Ωmild, 3Ωmoderate and 4Ωintense.
3. ‘‘Complete recovery’’ (patients without pain:
VASΩ0 and without abnormal sensations: VRSΩ0).
The patients enrolled in the study were explained
the difference between pain and abnormal sensations
and were given instructions on how to use the VAS
and VRS scales. During the follow-up examination the
patients had to determine whether their symptoms
were referable to ‘‘pain’’ or to ‘‘abnormal sensations’’.
Provoked pain assessment was performed with the
patient’s eyes closed and consisted of both the involved skin area and the healthy contralateral dermatome being rubbed lightly with a cotton ball or pinched lightly with tweezers. Although this study was not
designed to be double-blind, follow-up examinations
were performed by researchers who were not aware
of the treatment group to which the patient was assigned. Baseline sensory evaluations were not performed prior to initiation of the treatment since entry
criteria required only the presence of pain ⬎7 cm on
the VAS, and the change in these values from baseline
was not an endpoint of the study.
Rescue medication (analgesics) was available during follow-up, and any adverse effects correlated with
the treatment being used were also recorded.
Statistical analysis
Statistical analyses were performed on the intent-totreat population.
Two-tailed Z tests and c2 tests (with Yates’ correction) were used to verify that the two groups were
balanced in terms of demographic and clinical variables (sex, age, number of days elapsed from onset of
skin eruption and the beginning of treatment, localization of HZ, ‘‘pain’’ and ‘‘abnormal sensations’’).
The Mann-Whitney U-test was used to analyze VAS
and VRS scores between groups, and the c2 test and
Fisher’s exact test (for variables that had a low fre-
quency in at least one cell) were used to compare the
presence or absence of ‘‘pain’’ and of ‘‘abnormal sensation’’.
VAS and VRS are expressed as the median and
range, the localization of HZ is expressed as a percentage, whereas all the other parameters are expressed
as mean∫standard deviation (SD). Stated P values are
for two-tailed tests and were considered significant if
Of the 600 patients who were enrolled, 31 were excluded due to protocol violations (wrong assessment
of rash onset time, use of dosage and/or drugs that
differed from the study protocol). This left 569 patients (279 in Group A and 290 in Group B) that we
were able to assess as an intent-to-treat population.
The demographic characteristics of the patients are
shown in Table 1, and the cutaneous areas of HZ
localization at baseline are shown in Table 2. These
characteristics were similar for the two groups except
for the mean time from onset of rash to beginning of
treatment which was shorter in Group A (3.1∫2.1
days) compared with Group B (4.3∫2.3 days).
Of the 569 patients, 84 did not complete the followup due to death (causes not related to the pathology
or therapy), change of residency, or refusal or inability
to return to the hospital for assessment. Sixteen of
these patients dropped out after the first assessment
(1 month), 15 after the second (3 months), and 53 after
the 6-month follow-up visit.
The mean time required for treatment in Group
Table 1
Demographic and disease characteristics of the patient population.
Group A, acylovirπprednisolone; Group B bupivacaineπ
Group A
Group B
Male, n (%)
Female, n (%)
125 (44.8)
154 (55.2)
131 (45.2)
159 (54.8)
Age (years)
Mean (∫SD)
66.9 (17.1)
68.7 (18.2)
Baseline VAS
Median (range)
8.25 (7–10)
8.75 (7–10)
3.1 (2.1)
4.3 (2.3)
Time (days) from onset of
skin eruption to beginning of
Mean (∫SD)
* N.S.Ωnot significant.
Treatment for prevention of post-herpetic neuralgia
Table 2
Localization of HZ in Group A (acyclovirπprednisolone) and Group
B (bupivacaineπmethylprednisolone) at baseline (pre-treatment).
Number (%) of patients
Group A
HZ Site
I division
II division
III division
I and II division
Ramsay Hunt**
Group B
* N.S.Ωnot significant.
** Ramsay Hunt syndrome is Bell’s palsy associated with HZ of the
geniculated ganglion, herpetic lesions being visible on the tympanic membrane, external auditory canal, and on the pinna.
B was 12.5∫5.5 days, and during treatment,
bupivacaine was administered an average 2.4
Table 3 shows that at 1, 3, and 6 months post-therapy, significantly fewer patients in Group B (epidural
anestheticπsteroid) had PHN (painπabnormal sensations) than in Group A (intravenous acyclovirπ
After 1 year, the incidence of PHN (painπabnormal
sensations) was still significantly lower in Group B
(5.9%) than in Group A (34.3%; P⬍0.0001) (Table 3).
When the PHN symptoms are considered individually (i.e., either pain or abnormal sensations), a difference for pain is still observed between the treatments
at all time points. However, for abnormal sensations,
a difference between treatments is not observed until
3 months post-treatment (Table 3).
After 1 year, only 4 patients (1.6%) in Group B reported pain. This was lower than the 22.2% of the patients who reported pain in Group A. Likewise, the
incidence of abnormal sensations was almost threefold higher in Group A (12.2%) than in Group B
(4.3%) (Table 3).
For those patients who did report pain or abnormal
sensations, the intensity of pain was similar in the two
groups at all time points (Table 4). Furthermore, at
12 months, the dermatomal distribution of pain and
abnormal sensations was uniform between the two
groups (Table 5). Control tests on the contralateral
dermatome were negative in all patients.
Of the treated patients, 21 were immunocompromised, 9 patients had Hodgkin’s lymphoma, 6 had
non-Hodgkin’s lymphoma, and 6 had solid tumors.
The distribution of these patients was 8 in Group A
and 13 in group B. However, their immunocompromised state was neither clinically evident nor debilitating when they were enrolled, and in fact eight
of the patients were not known to be immunocompromised. This only became evident when five patients were hospitalized for tests. The routine tests for
enrollment revealed neoplasms in three patients.
None of the immunocompromised patients in group
B showed worsening of pathology or systemic dis-
Table 3
Comparative efficacy (incidence) of the treatment regimens at follow-up. Group A, acylovirπprednisolone; Group B bupivacaineπ
One month
Three months
Six months
One year
Group A
Group B
Total no. of patients: 569
Pts. (%) without PHN (complete recovery)
Pts. (%) with pain
Pts. (%) with ‘‘abn. sens.’’
39 (14.0)
113 (40.5)
127 (45.5)
133 (45.9)
22 (7.6)
135 (46.5)
Total no. of patients: 553
Pts. (%) without PHN (complete recovery)
Pts. (%) with pain
Pts. (%) with ‘‘abn. sens.’’
157 (57.3)
81 (29.6)
36 (13.1)
243 (87.1)
15 (5.4)
21 (7.5)
Total no. of patients: 538
Pts. (%) without PHN (complete recovery)
Pts. (%) with pain
Pts. (%) with ‘‘abn. sens.’’
170 (65.1)
60 (22.0)
31 (11.9)
252 (91.0)
10 (3.6)
15 (5.4)
Total no. of patients: 485
Pts. (%) without PHN (complete recovery)
Pts. (%) with pain
Pts. (%) with ‘‘abn. sens.’’
151 (65.6)
51 (22.2)
28 (12.2)
240 (94.1)
4 (1.6)
11 (4.3)
A. Pasqualucci et al.
Table 4
Comparative efficacy (severity of symptoms) of the treatment regimens at follow-up. Group A, acylovirπprednisolone; Group B bupivacaineπ
Group A
Group B
One month
VAS median (range); in pts. with pain
VRS median (range); in pts. with ‘‘abn. sens.’’
7 (3.5–10)
3.5 (2–4)
6.5 (3–10)
2.5 (1–4)
Three months
VAS median (range); in pts. with pain
VRS median (range); in pts. with ‘‘abn. sens.’’
6.5 (3–10)
3 (2–4)
6.5 (2.5–9.5)
2 (1–4)
Six months
VAS median (range); in pts. with pain
VRS median (range); in pts. with ‘‘abn. sens.’’
5.75 (2.5–9)
3 (2–4)
5.5 (2.5–8.5)
2 (1–4)
One year
VAS median (range); in pts. with pain
VRS median (range); in pts. with ‘‘abn. sens.’’
4 (2.5–7.5)
3 (2–4)
4 (2.5–7)
2 (1–4)
Table 5
Localization of HZ and number (%) of patients with pain and abnormal sensation after 12 months. Group A, acylovirπprednisolone; Group B,
Group A (nΩ230)
HZ site
Number of
I division
II division
III division
I and II division
Ramsay Hunt
Total no. of patients
Group B (nΩ255)
semination of varicella-zoster virus (VZV). Furthermore, none of these patients developed any pain or
abnormal sensations.
No significant differences were observed between
immunocompromised and immunocompetent patients treated with acyclovir.
Adverse events reported in Group A were predominantly limited to nausea and/or vomiting, diarrhea
and dyspepsia. Nineteen patients in this group failed
to complete therapy; seven patients spontaneously
stopped therapy, eleven patients had adverse effects
(nausea and/or vomiting, gastralgia), and one patient
had serious renal failure that required a brief period
of dialysis. Changes in serum creatinine levels and
liver function tests were observed in 21 patients, but
these changes were transitory in nature and they were
not considered clinically significant.
Thirteen patients in Group B failed to complete
therapy. Of these, two patients had frequent sweating
and fainting spells (one with a cervical epidural for
Number of
1 (2.9)
2 (1.7)
1 (2.4)
HZ of the trigeminal nerve), one had neck pain and
stiffness following cervical epidural, one had paresis
of the left leg and oliguria, and in nine patients the
epidural catheter came out. All adverse effects resolved completely.
In 16 patients in Group B, the dura mater was
pierced when the epidural catheter was being positioned. In these patients, the catheter was put into its
final position by shifting the insertion point by one
dermatome either in a cephalic or caudal direction,
and delaying the start of therapy by 48 h. Three of
these patients developed a dural puncture headache
that was fully resolved by immobilization in bed for
5 to 7 days. We were able to complete the study in all
of these patients and at final follow-up, none of them
had pain and only one had abnormal sensations.
No clinically important ocular complications were
observed, even among the immunocompromised patients. However, there was temporary inflammation
of the conjunctiva and/or the cornea in 27 patients. In
Treatment for prevention of post-herpetic neuralgia
13 of these patients (7 in Group A and 6 in Group B),
it was decided that the use of an antiviral ophthalmic
lotion during the first few days of therapy would provide an extra margin of safety.
The results of this randomized study demonstrate that
the administration of a local anesthetic π steroid via
the epidural route within 7 days of rash onset prevents the development of PHN (pain and abnormal
sensations) to a significantly greater extent (94.1%)
than acyclovirπsteroids (65.7%; P⬍0.0001). The difference between treatments was observed as early as at
1 month (45.9% vs 14.0%; P⬍0.0001) and it lasted for
the entire duration of the trial.
The prevention of the onset of pain in Group B was
especially dramatic at 1 month (92.4%) compared with
acyclovir (59.5%). Although 1 month may not be considered a clinically relevant time point for the development of PHN, these results were maintained until
the end of the trial (98.4% and 77.8% for groups B and
A, respectively).
In contrast, the incidence of abnormal sensations
was similar in the two groups at 1 month. Only after
3 months was a lower incidence observed in Group B
(92.5%) compared with Group A (86.9%). This difference was also maintained for the duration of the trial
(95.7% and 87.8% for groups B and A, respectively).
In agreement with a previous report (21), it appeared that PHN stabilized within 6 months from the
onset of the skin rash. The apparent reduction in the
number of patients with these symptoms in both
groups is primarily due to patients who were lost to
follow-up between the 6-month and 12-month time
points. We do not know the status of these patients.
For this reason, we are taking the conservative approach and stating that there was stabilization rather
than further improvement.
The intensity of pain (VAS) and abnormal sensations (VRS) was similar in the two groups at 4 follow-up points, suggesting that there are qualitative
differences rather than quantitative differences between the treatments. In addition, the dermatomal
distribution of the lesions did not appear to have any
effect on the results obtained using either treatment.
In the 21 immunocompromised patients, neither
treatment resulted in systemic dissemination or a
worsening of condition. Furthermore, no significant
differences were observed between the immunocompromised patients and the immunocompetent patients
in the development of either pain or abnormal sensations. This is in agreement with previous sugges-
tions that immunocompromised patients are not at
higher risk for developing PHN than immunocompetent patients (25, 26).
Both treatments in our study incorporated steroids.
Their use, although contentious, is still frequent.
Many authors agree that these drugs are effective for
both the acute phase pain and for the prevention of
PHN, and attribute their efficacy to the anti-inflammatory effects and to lysosomal protection which
could reduce neuronal damage (15, 27). Other
authors, however, doubt the efficacy of these drugs
and suggest that steroids increase the risk of herpes
dissemination (28, 29). Our choice of the dose was
based on other reported studies (10, 13, 14, 25), and
our results do not support the latter viewpoint that
these drugs are ineffective and contribute to systemic
The results observed in Group B are consistent with
previous uncontrolled studies that reported an incidence of PHN of 1.5–9% using similar epidural therapy (7–9, 11, 13, 14). The slight variability among the
studies may be attributable to a partial displacement
of the catheter, to insufficient coverage of the dermatomes involved, or to the definition of PHN itself, i.e.
whether patients with abnormal sensations were considered recovered or with pain.
It is agreed that treatment of HZ should begin as
early as possible for best efficacy (2, 7–11, 13, 14).
Many authors (uncontrolled studies) agree that the
epidural block seems to maintain its efficacy if performed within 10–15 days (7–9, 11, 13, 14) whereas
there does not seem to be any data with regard to
when acyclovir has to be administered. In our study,
treatment after rash onset was started significantly
later with anestheticπsteroid (4.3∫2.3 days) than with
acyclovirπsteroid (3.1∫2.1 days; P⬍0.001) due to
either administrative reasons (the patients from
Group B generally started treatment 24–36 h after
their control visit: laboratory tests and availability of
the physician responsible for positioning the epidural
catheter) or accidental piercing of the dura mater in
16 patients. Despite this delay, epidural treatment still
resulted in a significantly lower incidence of PHN
compared with the other treatment. The incidence of
PHN in Group A (acyclovir π steroid) was consistent
with other studies, even though our initiation of treatment occurred 24–48 h later than recommended and
used in other trials (2, 5, 10, 12).
Of the 279 patients who were treated with acyclovir,
196 were treated within 3 days from onset of skin
eruption (26 patients had a skin rash of 1 day, 41 patients of 2 days, 129 patients of 3 days). Of the 196
patients, 161 completed follow-up: 30 (18.6%) had
A. Pasqualucci et al.
pain and 23 (14.2%) referred abnormal sensations. Incidence of PHN (32.8%) in patients treated with acyclovir within 3 days from onset of skin rash was in
agreement with total data and therefore it is higher
than in patients treated with epidural block.
The efficacy of acyclovir in preventing PHN is not
unequivocal, and may depend upon the definition of
PHN. In our study, if the definition of PHN includes
both pain and abnormal sensations, the incidence of
PHN at 12 months (34.3%) is comparable to that reported in trials that did not specify whether the abnormal sensations were included in pain or recovery
(1, 2, 4, 5, 10, 12). However, if we separate out abnormal sensations and consider only pain, our data seem
to support the recent meta-analysis that suggests
some efficacy for acyclovir (6). During our study, it
was not infrequent to see abnormal sensations in addition to pain in some patients. However, in these patients, the abnormal sensations were always marginal
and of reduced intensity compared with the pain.
We therefore believe that it is necessary to further
differentiate the symptomatology and we propose
such a differentiation which may be correlated with
specific pathology (Table 6). According to our proposal, if abnormal sensations are present, there is
likely widespread inflammation and damage of the
cutaneous branches of the sensory nerves. This would
result in late onset phenomena involving reorganiza-
Table 6
Proposed differentiation of PHN into 3 symptomatologic groups and
pathophysiologic correlates.
1. Complete recovery.
Limited inflammation; no damage or minimal damage to ganglia, posterior roots and nerve endings; good immune response limits viral
spreading (34–37). It was observed in 5 patients with complete recovery (32, 39).
2. Abnormal sensations.
Widespread inflammation and damage to cutaneous sensory nerves
and spinal nerve root; phenomena of excitability that arose during
the acute phase (35–36); reorganization of nerve fibers, residual
modifications of the dorsal horn (peripheral and central sensitization
mechanisms) 23, 30, 31, 35, 36, 39). It was observed in 4 patients
with abnormal sensations (32, 39; case 21 of 31).
3. Pain:
– Allodynia. New connections between nonnociceptive large diameter primary afferents and central pain transmission neurons (pain
associated with small fiber deafferentation) (23, 34, 35, 40), in addition to the phenomena that arose in patients with abnormal sensations.
– Brief recurrent or constant and spontaneous pain. Damage of the
dorsal horn in addition to the damage of the spinal nerve root (with
or without a loss of both large and small diameter fibers) with the
mechanisms and characteristics of ‘‘central-type’’ pain (22–24, 34,
35, 40, 41). It was observed in 5 patients who had brief recurrent
or constant and spontaneous pain (38, 39; case 7 of 31).
tion of the peripheral nerve fibers and/or residual
modifications of the dorsal horn in addition to the development of excitability that took place during the
acute phase (23, 30–39). When ‘‘pain’’ is present, there
could be either the formation of new connections between primary afferents and central pain transmission
neurons (small fiber deafferentation leading to allodynia), or damage to the dorsal horn (leading to recurrent or constant spontaneous pain) (23, 31, 33–36, 38–
Therefore, in agreement with Fields et al. (35), it is
likely that more than one mechanism is operative in
one individual with PHN (patients with abnormal
sensations or with allodynia) but, in agreement with
Mondelli et al. (41) too, we believe that the only difference between subjects with and without brief recurrent or constant spontaneous pain is related to the
damage caused by the VZV to the dorsal horn and
caudal trigeminal nuclei.
These distinctions between ‘‘pain’’ and abnormal
sensations would appear to be crucial, and a more
complete differentiation (i.e. pain and allodynia, pain
and/or allodynia with or without other abnormal sensations, different symptomatologies of abnormal sensations) could facilitate comparisons between studies.
From a pathophysiologic perspective, we believe
that it is possible to explain the efficacy of local anesthetics in preventing PHN. Following VZV infection,
the virus remains quiescent within the dorsal root
ganglia, but if the immune response is weakened,
there is a change in the balance between the virus and
the host, and viral latency is broken, leading to acute
localized recrudescence of the disease (42–45). When
this occurs, viral spreading takes place by the axon
transport mechanism characteristic of all epidermoneurotrophic viruses (46–48), as well as by the anterograde/retrograde routes and transneuronally (30–49),
since it is also observed in the autonomic system
fibers (46). Once reactivated, VZV is especially damaging to the dorsal root ganglia, the peripheral nerves,
and the nerve endings, the latter resulting in skin
rash. The damage may diffuse centrally to the dorsal
horn of the spinal cord via the neurons (31, 32, 37, 42,
45, 50, 51). It is generally thought that 9–12 days are
necessary for central lesions to appear (31, 45, 50).
This variability in time, along with the different extension and seriousness of the damage (31, 32, 38, 39)
could be due to the immune status of the host and/
or the capacity to react to viral spreading (18, 33, 34,
45, 50, 51). Although it is possible that viral diffusion
could also occur via cell-to-cell contact, this is believed to be a secondary mechanism (31, 32, 42, 50). A
recent study (34) suggests that subclinical extension
Treatment for prevention of post-herpetic neuralgia
of viral inflammation into the spinal cord is common
in HZ, and another study suggests that VZV reactivation can result in acute aseptic meningitis even in
the absence of rash (33).
We propose that local anesthetics prevent PHN by
blocking axonal transport and hindering axonal and
transneuronal spread of the virus (52–56) and perhaps, indirectly, also its replication. Although the actual mechanism has not been elucidated, it is known
that axonal transport is independent of nervous conduction interruption (52, 53), and the axonal transport
blocking capacity of different local anesthetics has
been demonstrated at lower concentrations than the
ones adopted in clinical settings (53–56). Additionally,
pharmacokinetic studies have shown that drugs administered epidurally spread along the axon of the
spinal nerve roots to the most superficial layers of the
spinal cord (57–60). It is important to bear in mind
that VZV damage that occurs at the spinal root and
the dorsal horn (31, 38, 39, 42, 45, 50, 61) is equivalent
to the damage occurring at the spinal root of the trigeminal nerve and of the sensory nucleus of the trigeminal nerve, localized in the medulla oblongata and
in the first cervical segments (62).
Consequently, early treatment with local anesthetic
could be related to the possibility of blocking the
spread of VZV, giving the host enough time for its
immune system to react adequately.
In summary, this study demonstrates that continuous epidural anesthetic block in combination with
steroids is significantly more effective than treatment
with acyclovir and steroids in preventing PHN. This
efficacy may be due to the capacity of the local anesthetics to block the axonal diffusion of VZV.
We suggest that this treatment be employed as early
as possible in immunocompetent patients at high risk
of PHN. Although immunocompromised patients in
this study responded in a similar manner to the immunocompetent patients, the low number of immunocompromised patients does not permit the specific
recommendation of this therapy. Until further data are
available, concomitant administration of an epidural
block and an antiviral might be prudent to reduce the
possibility of disseminated infection or other serious
The authors thank Claudia Galli, PhD Tappo, for her advice and
help in translating, and Anna Boanelli, MD, Piero Cirulli, MD
(University of Perugia), Veronica Bonfreschi, MD, Riccardo Contardo, MD and Paola Antares, MD (University of Udine), for
their great assistance and dedication in collecting data.
1. Burgoon CF Jr, Burgoon JS, Baldridge GD. The natural history of herpes zoster. JAMA 1957: 164: 265–269.
2. Esman V, Ipsen J, Peterslund NA, Seyer-Hansen K,
Schonheyder H, Juhl H. Therapy of acute herpes zoster with
acyclovir in the non-immunocompromised host. Am J Med
1982: 73: 320–325.
3. Loeser JD. Herpes zoster and postherpetic neuralgia. Pain
1986: 25: 149–164.
4. Dworkin RH, Portenoy RK. Pain and its persistence in herpes zoster. Pain 1996: 67: 241–251.
5. Kost RJ, Straus SE. Postherpetic neuralgia – pathogenesis,
treatment, and prevention. N Engl J Med 1996: 335: 32–42.
6. Jackson JL, Gibbons R, Meyer G, Inouye L. The effect of
treating herpes zoster with oral acyclovir in preventing postherpetic neuralgia. Arch Intern Med 1997: 157: 909–912.
7. Colding A. The effect of regional sympathetic blocks in
treatment of herpes zoster. Acta Anaesthesiol Scand 1969: 13:
8. Manabe H, Dan K, Higa K. Continuous epidural infusion of
local anesthetics and shorter duration of acute zoster-associated pain. Clin J Pain 1995: 11: 220–228.
9. Pasqualucci V, Cirulli P, Paoletti F. Rationale for prevention
of postherpetic pain. 5th International Congress: The Pain
Clinic. Jerusalem, Israel, September 14–18, 1992: 20A.
10. Wood MJ, Johnson RW, McKendrick MW, Taylor J, Mandal
BK, Crooks J. A randomized trial of acyclovir for 7 days or
21 days with and without prednisolone for treatment of
acute herpes zoster. N Engl J Med 1994: 330: 896–900.
11. Wicks MA, Gomesz FAR. The use of epidural blocks and
trigeminal ganglion blocks in acute herpes zoster for the prevention of post herpetic neuralgia. Pain in Europe, Congress
of European Federation of IASP Chapters. Verona, Italy, May
18–21, 1995: 153A.
12. Peterslund NA, Seyer-Hansen K, Ipsen J, Esmann V,
Schonheyder H, Juhl H. Acyclovir in herpes zoster. Lancet
1981: 2: 827–830.
13. Pasqualucci V, Pasqualucci A, Paoletti F. Efficacy of Acute
R.A. intervention. Symposium Faculty: Herpes zoster and
postherpetic neuralgia. The Cleveland Clinic Foundation.
Chateau Whistler Resort, Whistler Mountain, British Columbia, Canada, August 13–14, 1996.
14. Pernak J, Erdmann W, Bryant JD. Acute herpes zoster of the
trigeminal nerve and its treatment. J Pain Therapy 1993: 3:
15. Lipton S. Relief of pain in clinical practice. Oxford: Blackwell,
1979: 223–237.
16. Perkins HM, Hanlon PR. Epidural injection of local anesthetic and steroids for relief of pain secondary to herpes zoster. Arch Surg 1978: 113: 253–254.
17. Portenoy RK, Duma C, Foley KM. Acute herpetic and postherpetic neuralgia: Clinical review and current management. Ann Neurol 1986: 20: 651–664.
18. Hope-Simpson RE. The nature of herpes zoster. A long term
study and a new hypothesis. Proc R Soc Med 1965: 58: 9–20.
19. Eaglstein WH, Katz R, Brown JA. The effects of early corticosteroid therapy on the skin eruption and pain of herpes zoster. JAMA 1970: 211: 1681–1683.
20. Dworkin RH, Portenoy RK. Proposed classification of herpes zoster pain (letter). Lancet 1994: 343: 1648.
21. Riopelle JM, Naraghi M, Grush KP. Chronic neuralgia incidence following local anesthetic therapy for herpes zoster.
Arch Dermatol 1984: 120: 747–750.
22. Rowbotham MC, Fields HL. Post-herpetic neuralgia: the relation of pain complaint, sensory disturbance, and skin temperature. Pain 1989: 39: 129–144.
A. Pasqualucci et al.
23. Nurmikko T, Bowsher D. Somatosensory findings in postherpetic neuralgia. J Neurol Neurosurg Psychiatry 1990: 53:
24. Baron R, Sauger M. Postherpetic neuralgia. Are C-nociceptors involved in signalling and maintenance of tactile allodynia? Brain 1993: 116: 1477–1496.
25. Whitley RJ, Weiss H, Gnann JW Jr, Tyring S, Mertz GJ, Pappas PG et al. Acyclovir with and without prednisolone for the
treatment of herpes zoster. A randomized, placebo-controlled trial. The National Institute of Allergy and Infectious
Diseases Collaborative Antiviral Study Group. Ann Intern
Med 1996: 125: 376–383.
26. Balfour HH. Varicella-zoster virus infections in immunocompromised hosts: a review of the natural history and management. Am J Med 1988: 85 (Suppl 2A): 68–73.
27. Moesker A, Boersma EP. The effect of extradural administration of corticosteroids as pain treatment of acute herpes zoster and prevention of postherpetic neuralgia. Pain Clinic
1975: 1: 273.
28. Huff JC. Herpes zoster. Curr Probl Dermatol 1988: 1: 1–4.
29. Esmann V, Geil JP, Kroon S et al. Prednisolone does not prevent postherpetic neuralgia. Lancet 1987: 2: 126–129.
30. Norgren RB Jr, Lehman MN. Retrograde transneuronal
transport of Herpes Simplex Virus in the retina after injection in the superior colliculus, hypothalamus and optic chiasm. Brain Res 1989: 479: 374–378.
31. Head H, Campbell AW. The pathology of herpes zoster and
its bearing on sensory localisation. Brain 1900: 23: 353–523.
32. Muller SA, Winkelmann RK. Cutaneous nerve changes in
zoster. J Invest Dermatol 1969: 52: 71–77.
33. Echevarria JM, Casas I, Martinez-Martin P. Infections of the
nervous system caused by varicella-zoster virus: a review.
Intervirology 1997: 40: 72–84.
34. Haanpaa M, Dastidar P, Weinberg A, Levin M, Miettinen A,
Lapinlampi A et al. CSF and MRI finding in patients with
acute herpes zoster. Neurology 1998: 51: 1405–1411.
35. Fields H, Rowbotham M, Baron R. Postherpetic neuralgia:
irritable nociceptors and deafferentation. Neurobiol Dis 1998:
5: 209–227.
36. McMahon S, Koltzenburg M. The changing role of primary
afferent neurones in pain. Pain 1990: 43: 269–272.
37. Haanpaa M, Laippala P, Nurmikko T. Pain and somatosensory dysfunction in acute herpes zoster. Clin J Pain 1999: 15:
38. Watson CPN, Morshead C, Van der Rooy D, Deck O, Evans
RJ. Post-herpetic neuralgia: post-mortem analysis of a case.
Pain 1988: 34: 129–138.
39. Watson CPN, Deck JEI, Morshead C, Van der Kooy D, Evans
RJ. Post-herpetic neuralgia: further post-mortem study of
cases with or without pain. Pain 1991: 44: 105–117.
40. Zacks SL, Langfitt TW, Elliott FA. Herpetic neuritis: a light
and electron microscopic study. Neurology 1964: 14: 744–750.
41. Mondelli M, Romano C, Della Porta P, Rossi A. Electrophysiological findings in peripheral fibres of subjects with and
without post-herpetic neuralgia. Electroencephalogr Clin Neurophysiol 1996: 101: 185–191.
42. Esiri MM, Tomlinson AH. Herpes zoster: demonstration of virus in trigeminal nerve and ganglion by immunofluorescence
and electron microscopy. J Neurol Sci 1972: 15: 35–48.
43. Cohrs RJ, Barbour M, Gilden DH. Varicella-zoster virus
(VZV) transcription during latency in human ganglia: detection of transcripts mapping to genes 21, 29, 62, and 63 in a
cDNA library enriched for VZV RNA. J Virol 1996: 70: 2789–
44. Lungu O, Annunziato PW, Gershon A, Staugaitis SM, Josefson D, La Russa P et al. Reactivated and latent varicella-
zoster virus in human dorsal root ganglia. Proc Natl Acad Sci
USA 1995: 92: 10980–10984.
Jemsek J, Greenberg SB, Taber L, Harvey D, Gershon A, Couch RB. Herpes zoster-associated encephalitis: clinicopathologic report of 12 cases and review of the literature. Medicine (Baltimore) 1983: 62: 81–97.
Marchand CF, Schwab ME. Binding uptake and retrograde
axonal transport of herpes virus suis in sympathetic neurons. Brain Res 1986: 383: 262–270.
Penfold MET, Armati P, Cunningham AL. Axonal transport
of herpes simplex virions to epidermal cells: evidence for a
specialized mode of virus transport and assembly. Proc Natl
Acad Sci USA 1994: 91: 6529–6533.
Kristensson K. lmplication of axoplasmic transport for the
spread of virus infections in nervous system. In: Weiss DG,
Gorio A, Eds. Axoplasmic transport in physiology and pathology.
Berlin: Springer, 1982: 153–158.
Ugolini G, Kuypers HGJM, Simmons A. Retrograde transneuronal transfer of herpes simplex virus type I (HSV1)
from motoneurones. Brain Res 1987: 422: 242–256.
Devinsky O, Cho ES, Petito CK, Price RW. Herpes zoster
myelitis. Brain 1991: 114: 1181–1196.
Ochs S, Brimijoin WS. Axonal transport. In: Dyck PJ, Thomas PK et al., eds. Peripheral Neuropathy, Vol. 1. 3rd edn. Philadelphia, USA: WB Saunders Company, 1993: 331–360.
Lavoie PA. Block of fast axonal transport in vitro by the local
anesthetics dibucaine and etidocaine. J Pharmacol Exp Ther
1982: 223: 251–256.
Lavoie PA. Inhibition du transport axonal rapide par les
anesthe´siques locaux. Union Me´dicale du Canada 1987: 116:
Lavoie PA, Khazen T, Filion PR. Mechanism of the inhibition
of fast axonal transport by local anesthetics. Neuropharmacology 1989: 28: 175–181.
Bisby MA. Inhibition of axonal transport in nerves chronically treated with local anesthetics. Exp Neurol 1975: 47: 481–
Powell HC, Kalichman MW, Garrett RS, Myers RR. Selective
vulnerability of unmyelinated fiber Schwann cells in nerves
exposed to local anesthetics. Lab Invest 1988: 59: 271–280.
Bromage PR. Analgesia epidurale. Roma: Verducci editore,
1980: 97–129.
Byrod G, Olmarker K, Konno S, Larsson K, Takahashi K,
Rydevik B. A rapid transport route between the epidural
space and the intraneural capillaries of the nerve roots. Spine
1995: 20: 138–143.
Dietz FB, Jaffe RA. Bupivacaine preferentially blocks ventral
root axons in rats. Anesthesiology 1997: 86: 172–180.
Cousins MJ, Bromage PR. Epidural neural blockade. In:
Cousins MJ, Bridenbaugh PO, eds. Neural blockade in clinical
anesthesia and management of pain. 2nd edn. Philadelphia:
Lippincott, 1988: 253–360.
Noordenbos W. Pain: problems pertaining to the transmission of
nerves impulses which give rise to pain: preliminary statement.
Amsterdam: Elsevier, 1959: Vol. I, 4–10; Vol. X, 68–80.
England MA, Wakely J, eds. Color atlas of the brain & spinal
cord. St Louis, USA: Mosby-Year Book Inc., 1991: 82–196.
Alberto Pasqualucci MD
Department of Anesthesiology, Intensive Care and Pain Clinic
University of Udine
Piaz. S.M. della Misericordia
33100 Udine, Italy
e-mail: A.Pasqualucci/