ARTICLE
Auteur(s) : Laurent Mardivirin1, Laurence
Valeyrie-Allanore2, Estelle Branlant-Redon3,
Nathalie Beneton2, Kaoutar Jidar4, Annick
Barbaud3, Béatrice Crickx2, Sylvie
Ranger-Rogez1, Vincent
Descamps4
1Department of Virology, Centre Hospitalier
Universitaire Dupuytren, Limoges, France
2Department of Dermatology, Hôpital Bichat-Claude
Bernard, 46 rue Henri Huchard, 75018 Paris, France
3Department of Dermatology, Centre Hospitalier
Universitaire de Nancy, Nancy, France
4Department of Infectious Diseases, Hôpital
Bichat-Claude Bernard, 46 rue Henri Huchard, 75018 Paris,
France
accepté le 9 Septembre 2009
DRESS (Drug Reaction with Eosinophilia and Systemic Symptoms) is
a severe, life-threatening, drug-induced, multi-organ system
reaction [1, 2]. It is characterized by rash, facial edema, fever,
lymphadenopathy and leukocytosis with eosinophilia. Its severity is
explained by the development of visceral manifestations including
hepatitis, encephalitis, pneumonitis, hemophagocytic syndrome and
multi-organ failure. DRESS is associated with the reactivation of
the human herpesviruses, including Human Herpesvirus 6 (HHV-6),
Epstein Barr virus (EBV) and cytomegalovirus (CMV) [3-7]. The
reactivation of herpesviruses is recognized to participate in the
pathophysiology of this syndrome. Clinical and biological
manifestations are considered to illustrate the systemic reaction
to human herpesviruses and to be the consequence of an immune
response against the reactivation of human herpesviruses [8,
9].
We observed a flare of DRESS associated with amoxicillin intake
at the beginning of DRESS in several cases induced by classical
drugs. Amoxicillin was given in a clinical setting of pharyngitis
with cervical lymphadenopathy and fever, when the diagnosis of
DRESS had not yet been established. During the following days a
flare of DRESS was observed. In the context of herpesvirus
infection, amoxicillin-induced flare in DRESS is reminiscent of an
amoxicillin/ampicillin-induced rash associated with infectious
mononucleosis [10, 11].
We recently reported that sodium valproate, a DRESS-associated
drug, increases the replication of HHV-6 in vitro [12]. We studied
the early effects of amoxicillin on HHV6 replication in vitro. An
increase of HHV-6 replication was demonstrated. Our hypothesis is
that amoxicillin-induced flares may be the consequence of a direct
effect of amoxicillin on HHV-6 replication.
Materials and methods
Cases
7 cases of DRESS with amoxicillin-induced flare were
retrospectively analysed (table 1). The
cases were retrospectively selected by members of the French Group
of Cutaneous Drug Adverse Reactions of the French Society of
Dermatology. These patients were admitted to the Department of
Dermatology of Bichat Hospital (Paris, France), Nancy Hospital
(Nancy, France) or Henri Mondor Hospital (Créteil, France). They
showed typical DRESS symptoms. The diagnosis of DRESS was
established according to the classical criteria: exanthema and
facial edema, lymphadenopathy, atypical T cell lymphocytes,
eosinophilia, and hepatitis. One case (case 6) has been previously
reported [2]. We collected the following data: age, sex, drug
intake, clinical symptoms, available biological tests, and
virological tests, when available. Because this is a descriptive,
non-interventional study, the project did not require approval by
an institutional review board.
Table 1 Characteristics of the patients
|
Patient no.
|
Age (years)/sex
|
Drug associated with DRESS: causal drug
|
Time between first culprit drug intake and first manifestations
of DRESS (duration of treatment)
|
Time between diagnosis of DRESS and amoxicillin intake (duration
of amoxicillin intake)
|
Manifestations before amoxicillin intake
|
Manifestations after amoxicillin intake (time after the first
amoxicillin intake)
|
HHV6 PCR
|
|
1
|
26/F
|
Carbamazepine
|
14 days (14 days)
|
D0 (1 day)
|
Pharyngitis with facial edema (D0)
|
Fever, rash, facial edema mononucleosis, eosinophilia (1 day)
|
ND
|
|
Re-challenge with carbamazepine at D6
|
|
Rash (D1)
|
Fever, flare of the rash, facial edema, eosinophilia hepatitis (1
day after re-challenge with carbamazepine)
|
|
2
|
43/H
|
Carbamazepine
|
17 days (39 days)
|
D2-D9 (8 days)
|
Pharyngitis, malaise, flu-like syndrome (D0)
|
Fever, rash, eosinophilia, hepatitis (14 days)
|
ND
|
|
3
|
68/M
|
Lisinopril
|
ND
|
D5-D10 (6 days)
|
Pharyngitis
|
Fever, flare of the rash, facial edema, eosinophilia hepatitis
|
ND
|
|
4
|
17/H
|
Carbamazepine
|
20 days (20 days)
|
D8 (1 day)
|
Rash (D0) Pharyngitis (D8)
|
Cutaneous flare, high fever, hepatitis, eosinophilia,
lymphadenopathy (1 day)
|
ND
|
|
5
|
76/F
|
Strontium ranelate
|
31 days (31 days)
|
D3-D9 (7 days)
|
Pharyngitis and rash (D0)
|
Cutaneous flare, fever, eosinophilia, hepatitis, renal failure (5
days)
|
ND
|
|
Lansoprazol
|
61 days (61 days)
|
|
6
|
16/F
|
Minocycline
|
12 days (15 days)
|
Amoxicillin + clavulanic acid D0-D4 (5 days)
|
Pharyngitis
|
Facial oedema, rash, high fever, lymphadenopathy, eosinophilia (4
days)
|
HHV6 PCR + (D20)
|
|
7
|
61/M
|
Allopurinol
|
19 days (29 days)
|
D5-D10 (5 days)
|
Pharyngitis, fever
|
Rash, eosinophilia, fever, lymphocytosis, eosinophilia, hepatitis,
renal failure, hemophagocytic syndrome (8 days)
|
HHV6 PCR + (D12, D19)
|
Cells and viruses
The human T lymphoblastoid MT4 cell line was cultured in a RPMI
1640 medium with 12.5% fetal calf serum (FCS). HHV-6B strain HST
was propagated in MT4 cells in RPMI 1640 with 12.5% FCS,
supplemented by 10 U.mL−1 anti-interferon-α (Sigma) and
2.5 μg.mL−1 polybrene (Sigma). All cultures were
grown at 37 °C with 5% CO2. HHV-6 cell-free viruses
were prepared as described elsewhere [13]. When more than 80%
HHV-6-infected cells showed a cytopathic effect, the culture of
infected cells was frozen and thawed twice. After centrifugation at
500 g for 10 min, the supernatant was stored at
– 80 °C as cell-free virus stock.
Virus titration by indirect immunofluorescence
Virus titre of supernatant stocks was studied by MT4 infection in
24-well microtiter plates. Several dilutions of viral inoculum were
added to 5.105 cells, and the infection was enhanced by
a centrifugation lasting for one hour at 2,000 g. After a
24 h incubation, infected cells were harvested, washed and
fixed onto slides in cold acetone. An immunofluorescence assay
(IFA) was then performed, with anti-HHV-6 p41 primary antibody
(ABI, dilution 1:100) which recognizes a protein in the early stage
of viral replication, and a fluorecein-conjugated anti-mouse
secondary antibody (Argene, dilution 1:100). Stained cells were
then washed and percentages of infected cells were determined by
fluorescence microscopy.
Cytotoxicity assay
Cell proliferation was assessed using the
3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT)
dye reduction assay [14]. MT4 cells were seeded onto 96-well
microtiter plates and incubated with a culture medium containing
different concentrations of amoxicillin (0, 5, 10, 25, 50, 100 and
200 μg.mL−1, Sigma). The incubation time was the
same as for viral replication assays: 48 and 72 h. After
incubation with amoxicillin, MTT (6 mg.mL−1) was
added and after an additional 3 h incubation time, cells were
lysed in dimethyl sulfoxide (DMSO). Absorbance at 540 nm was
determined for each well using a 96-well multiscanner. Results are
expressed in percentages of cell viability, by comparing to the
control. Experiments were performed in quadruplicate, 3 times.
Viral replication assays
MT4 cells were pre-incubated with concentrations of amoxicillin
(ranking from 0 to 200 μg.mL−1) in culture medium
during 24 h. Infection with viral supernatants was performed
(as described for virus titration), with the appropriate dilution
for a multiplicity of infections (MOI) of 0.2. Infected cells were
then incubated during 24 or 48 h, with the initial drug
concentrations. The viral replication was then quantified by IFA as
described previously, and by real-time PCR. For each concentration
of amoxicillin tested, experiments were performed 3 times.
Real-time PCR
Cells were washed and DNA was then extracted using a standard
phenol chloroform procedure [15]. Real-time PCR was performed by
amplifying the HHV-6 U22 gene, using 100 ng of purified DNA,
as described elsewhere [16].
Statistical analysis
Values presented are the means and standard deviations of 3
experiments. Comparisons with negative controls were performed
using Student’s t test. P values lower than 0.05 were considered to
be significant.
Results
The characteristics of the 7 patients are described in table 1. The causative drugs for DRESS were
carbamazepine (3 cases), strontium ranelate or lansoprazol,
minocycline, lisinopril and allopurinol. Amoxicillin was given in
six patients when the diagnosis of DRESS had not yet been made
because the patients presented symptoms of pharyngitis, and
sometimes fever and lymphadenopathy. These manifestations
(pharyngitis, fever, lymphadenopathy) were probably the first
clinical signs of DRESS. Day 0 (beginning of DRESS) was
retrospectively fixed at the beginning of the clinical
manifestations. Amoxicillin was given during an average of 4.7 days
(1-8 days). In every patient a flare was observed during the
treatment or within the following days. Every physician in charge
of the patients had the feeling that amoxicillin could contribute
to the development of DRESS. In every case the clinical and
biological manifestations were typical of DRESS. Two cases were
severe. Case 6 has already been reported [2]. In case 7, a
hemophagocytic syndrome developed at day 19.
Interestingly, in cases 1 and 2, amoxicillin was suspected to be
the causal drug for the manifestations. In case 1, after an initial
withdrawal of both carbamazepine and amoxicillin, carbamazepine was
given again from day 5 onwards with a new flare of DRESS. In case
2, only amoxicillin was stopped. Carbamazepine was only stopped 12
days later because the clinical and biological picture worsened. In
4 cases, including these two cases (cases 1, 2, 6 and 7), the
stoppage of the culprit drug was delayed. These clinical
manifestations were not recognized as possible early manifestations
of DRESS.
In two cases (cases 4 and 5), patch testing was performed for
amoxicillin and carbamazepine, and strontium ranelate and
lanzoprazol, respectively. In case 4, the amoxicillin patch test
was negative and the carbamazepine patch test was positive. In case
5, amoxicillin and lanzoprazol patch tests were positive.
A strontium ranelate patch test was negative. Prick tests were
negative for these three drugs in this patient. Amoxicillin was not
reintroduced in these patients. HHV-6 reactivation was confirmed by
PCR analysis of blood sample in two cases (cases 6 and 7) at Day 12
and Day 19 (case 7), and Day 20 (case 6). No test was available at
the very beginning of DRESS.
To investigate the influence of amoxicillin on MT4 cell
proliferation, the MTT dye reduction assay was used. The cell
viability, compared to control, is close to 100% for the range of
amoxicillin concentrations tested. No increased cell proliferation
was observed for the different concentrations studied.
Amoxicillin treatment for 24 h before infection and 24 or
48 h after infection increased HHV-6 strain HST replication in
MT4 cells. A significant increase in the percentage of
infected cells was observed for 25 μg.mL−1
amoxicillin 24 and 48 h after infection, and for
50 μg.mL−1 amoxicillin 48 h after infection
(figure 1A).
Similar results were obtained using real-time PCR for quantifying
HHV-6 replication: the number of U22 gene copies was significantly
increased after 24 and 48 h for 25 μg.mL−1
amoxicillin, and after 48 h for 50 μg.mL−1
amoxicillin (figure
1B).
The increase of viral replication did not depend on the dose in
a proportional way, but a particular concentration range was able
to stimulate HHV-6 strain HST replication in MT4 cells. This range
included 25 and 50 μg.mL−1 amoxicillin. The classic
amoxicillin plasma levels did not exceed
15 μg.mL−1, but some cases of intravenous
administration were able to induce plasma levels up to
120 μg.mL−1 or more [17].
Discussion
In seven cases we observed a flare after amoxicillin intake in
patients with DRESS associated with other drugs. Amoxicillin was in
every case administrated because the patients presented with
pharyngitis, lymphadenopathy and fever. Retrospectively it was
considered that these manifestations were the first signs of DRESS.
Within the following days an increase in clinical and biological
symptoms was observed with the development or flare of rash, facial
edema, and biological abnormalities. In every case this flare was
responsible for admission to hospital.
The beginning of DRESS is often very dramatic as it is
associated with high fever, lymphadenopathy and rash. But when we
ask the patients whether they noticed some prior manifestations in
the previous days, they often report some clinical manifestations
like pharyngitis, lymphadenopathy, fatigue, and fever.
We think that this sequence of events is not infrequent and
needs to be seriously considered. This reaction needs to be known
because sometimes amoxicillin is considered to be the culprit drug,
even though the true culprit drug had been taken for many weeks. It
is also important when faced with a clinical and biological picture
of amoxicillin-induced DRESS to investigate which other drugs have
been or are being taken. There may also be a delay in the
withdrawal of the true culprit drug.
In these cases, the role of amoxicillin intake in influencing
the development of DRESS is questionable. Several hypotheses may be
proposed: (a) a spontaneous development of DRESS, (b) an adverse
reaction against amoxicillin alone, (c) an adverse reaction against
amoxicillin that precipitates a reaction against the “culprit drug”
or inversely, (d) a direct role of amoxicillin as compared to an
amoxicillin-induced rash in mononucleosis syndrome. No patient had
a previous allergy to amoxicillin. The time between the treatment
with amoxicillin and the development of the complete manifestations
of DRESS was too short to consider amoxicillin as the culprit drug
in these cases. In some cases the flare was observed while the
patient was still taking the “culprit drug”. We think that this
reaction is reminiscent of the amoxicillin reaction in patients
with infectious mononucleosis. DRESS is associated with the
reactivation of human herpesviruses. We consider that this
reactivation is occurring at both the very beginning of DRESS,
explaining the clinical symptoms and the mononucleosis syndrome,
and during the subsequent flares of DRESS [18]. The virological
context of infectious mononucleosis is also very close to DRESS. We
recently reported that, in DRESS, the T-cell response was mainly
directed to viral antigens [8].
Amoxicillin-induced exanthema in patients with infectious
mononucleosis is still not well understood. The incidence of rash
following aminopenicillin intake in patients with infectious
mononucleosis rises in adults to 27.8-69%, with up to 100%
incidence reported in children [19]. Interestingly, when
amoxicillin is given long after the outbreak, most often no flare
is observed. It is classically recognized that viral infections may
enhance the risk of drug allergic reactions. Two mechanisms have
been proposed in the context of a breakdown of tolerance or
enhancement of the immune reaction to drugs following a viral
infection: a change in the antigen expression of the drug or an
alteration of immune response [19, 20]. In a recent series of 8
cases patch tests were either positive (5 cases) or negative (3
cases) [20]. Challenge with amoxicillin was done in 3 cases: 2
cases were negative, one case was positive. Some authors have
already reported this phenomenon during DRESS as a drug
neo-sensitization [21]. They suggested that a transient state of
immunosuppression was induced during DRESS with latent virus
reactivation and a massive non-specific immune system response
leading to a breakdown of tolerance to other drugs [21].
Interpretation of our patch tests with amoxicillin in two patients
is limited. Moreover the patient with positive patch test had a
negative prick test.
A rethinking of this reaction is necessary. Another explanation
may be given. Alternatively a direct effect of drugs on viral
replication may have occurred in a state of viral reactivation.
Recently Saito-Katsuragi et al. reported an illustrative case
of an ampicillin-induced cutaneous eruption associated with an
Epstein-Barr virus reactivation [22]. A 23-year-old woman who
was diagnosed with adult onset Still’s disease developed a
cutaneous rash, fever and liver dysfunction when she was treated
with ampicillin. A high level of EBV-DNA was detected at this
time. A diagnosis of infectious mononucleosis was made and a
challenge test with ampicillin was performed at day 39, day 90, and
day 165. A cutaneous rash was observed after every challenge.
Interestingly, a high level of EBV-DNA was observed 24 to
48 hours after every challenge test, preceding the cutaneous
eruption. This observation suggests that ampicillin may induce or
increase herpesvirus reactivation (in this case EBV) with a
concomitant rash. This reaction may occur when the virus itself is
already in a state of reactivation. This is the case in infectious
mononucleosis, some transplant recipients, some patients admitted
in intensive care units, and DRESS [9].
A direct action of drugs, and especially amoxicillin, on
herpesvirus replication has not been much studied. We demonstrated
here that amoxicillin increases HHV-6 replication in vitro. In a
previous report we have already demonstrated that valporate acid
increases HHV-6 replication in the same way [12]. This observation
was also reported for another herpesvirus, the cytomegalovirus
[23]. Recently, Michaelis et al. demonstrated that therapeutic
valproic acid concentrations increased CMV replication and impaired
the antiviral activity of the anti CMV drugs ganciclovir, cidofovir
and foscarnet. The link between valproic acid and herpesviruses may
be by the way of histone desacetylase (HD) inhibition. Valproate
acid inhibits HD and induces HD-independent extracellular
signal-regulated kinases1/2 phosphorylation in endothelial cells
[24]. In the same way it was demonstrated that valproic acid
advanced the replication timing of Epstein-Barr virus [25]. These
HD inhibitors seem to be critical in switching latent forms into
lytic forms of viral infection. The mechanism of amoxicillin action
on virus replication remains unknown.
This amoxicillin-induced reaction is not constant in DRESS. When
we decided to undertake this retrospective study we examined many
cases of DRESS with amoxicillin intake. In most cases
administration of amoxicillin after the two first weeks of
diagnosing DRESS did not induce a significant flare. But an
amoxicillin-induced flare could be detected in some severe cases
with systemic manifestations. Some interfering factors (bacterial
infection, intake of other drugs including corticosteroids) did not
enable us to conclude with any certainty. It is probable that this
reaction is dependant of the state of virus reactivation. This
amoxicillin-induced flare was mainly observed either at the
beginning of DRESS or in cases of severe manifestations, two
periods that are associated with reactivation of herpesviruses.
We propose that DRESS has at its beginning a clinical and
biological picture close to severe infectious mononucleosis and
amoxicillin may increase the clinical and biological
manifestations, as observed in infectious mononucleosis.
We consider that amoxicillin is only a trigger in the
development of DRESS. We demonstrated that the replication of HHV-6
is increased by amoxicillin in vitro. We believe that amoxicillin
may increase HHV-6 reactivation, as was recently suggested in one
clinical case with EBV.
Dermatologists must recognise this possibility to avoid a false
interpretation of an amoxicillin-induced rash as allergic in case
of DRESS and to look for another potential culprit drug. This
hypothesis needs to be confirmed by sequential measurement of HHV-6
viremia in patients. We did not have the opportunity to study
serial blood samples in a DRESS patient taking amoxicillin. The
mechanism of action of amoxicillin in HHV-6 or EBV replication
remains unknown and warrants further study.
Acknowledgements
We thank Peter Höller for English proof-reading. Conflict of
interest: none declared. Financial support: none.
References
1 Roujeau JC, Stern RS. Severe adverse cutaneous
reactions to drugs. N Engl J Med 1994; 331: 1272-85.
2 Eshki M, Allanore L, Musette P, et al.
Twelve-year analysis of severe cases of drug reaction with
eosinophilia and systemic symptoms: a cause of unpredictable
multiorgan failure. Arch Dermatol 2009; 145: 67-72.
3 Descamps V, Bouscarat F, Laglenne S,
et al. Human herpesvirus 6 infection associated with
anticonvulsant hypersensitivity syndrome and reactive
haemophagocytic syndrome. Br J Dermatol 1997; 137: 605-8.
4 Suzuki Y, Inagi R, Aono T, Yamanishi K,
Shiohara T. Human herpesvirus 6 infection as a risk factor for
the development of severe drug-induced hypersensitivity syndrome.
Arch Dermatol 1998; 134: 1108-12.
5 Tohyama M, Yahata Y, Yasukawa M, et al.
Severe hypersensitivity syndrome due to sulfasalazine associated
with reactivation of human herpesvirus 6. Arch Dermatol 1998; 134:
1113-7.
6 Descamps V, Valance A, Edlinger C, et al.
Association of human herpesvirus 6 infection with drug reaction
with eosinophilia and systemic symptoms. Arch Dermatol 2001; 137:
301-4.
7 Shiohara T, Iijima M, Ikezawa Z,
Hashimoto K. The diagnosis of a DRESS syndrome has been
sufficiently established on the basis of typical clinical features
and viral reactivations. Br J Dermatol 2007; 156: 1083-4.
8 Picard D, Janela B, Descamps V, et al.
T-cell repertoire, phenotype and functionality in drug
hypersensitivity syndrome. 2007 ESDR, Zurich, 2007. J Invest
Dermatol 2007; 127: S21; (abstract).
9 Shiohara T, Kano Y. A complex interaction between
drug allergy and viral infection. Clin Rev Allergy Immunol 2007;
33: 124-33.
10 Pullen H, Wright N, Murdoch JM.
Hypersensitivity reactions to antibacterial drugs in infectious
mononucleosis. Lancet 1967; 2: 1176-8.
11 Nazareth I, Mortimer P, McKendrick GD.
Ampicillin sensitivity in infectious mononucleosis temporary or
permanent? Scand J Infect Dis 1972; 4: 229-30s.
12 Mardivirin L, Lacroix A. Delebassée, Descamps V,
Ranger-Rogez S. Augmentation de la réplication in vitro de
l’herpèsvirus humain 6 en présence de valproate de sodium.
Virologie 2007; 11: 1-3.
13 Dhepakson P, Mori Y, Jiang YB, et al.
Human herpesvirus-6 rep/U94 gene product has single-stranded
DNA-binding activity. J Gen Virol 2002; 83: 847-54.
14 Mosmann T. Rapid colorimetric assay for cellular growth
and survival: application to proliferation and cytotoxicity assays.
J Immunol Methods 1983; 65: 55-63.
15 Sambrook J, Fritsh EF, Maniatis T. Molecular
Cloning: A Laboratory Manual. New York: Cold Spring Harbor
Laboratory, 1989.
16 Lacroix A, Jaccard A, Rouzioux C, et al.
HHV-6 and EBV DNA quantification in lymph nodes of 86 patients with
Hodgkin’s lymphoma. J Med Virol 2007; 79: 1349-56.
17 Jones AE, Barnes ND, Tasker TC, Horton R.
Pharmacokinetics of intravenous amoxicillin and potassium
clavulanate in seriously ill children. J Antimicrob Chemother 1990;
25: 269-74.
18 Descamps V, Mardivirin L, Ranger-Rogez S.
Chromosomal integration of human herpesvirus 6 DNA in
anticonvulsant hypersensitivity syndrome: a unique opportunity for
the understanding of their relation. Br J Dermatol 2008; 159:
759.
19 Renn CN, Straff W, Dorfmüller A,
Al-Masaoudi T, Merk HF, Sachs B. Amoxicillin-induced
exanthema in young adults with infectious mononucleosis:
demonstration of drug-specific lymphocyte reactivity. Br J Dermatol
2002; 147: 1166-70.
20 Jappe U. Amoxicillin-induced exanthema in patients with
infectious mononucleosis: allergy or transient immunostimulation?
Allergy 2007; 62: 1474-5.
21 Gaig P, García-Ortega P, Baltasar M,
Bartra J. Drug neosensitization during anticonvulsant
hypersensitivity syndrome. J Investig Allergol Clin Immunol 2006;
16: 321-6.
22 Saito-Katsuragi M, Asada H, Yokoi S, Niizeki H, Miyagawa S.
Ampicillin-induced cutaneous eruption associated with Epstein-Barr
virus reactivation. J Am Acad Dermatol 2005; 52 (5 Suppl 1):
S127-8.
23 Kuntz-Simon G, Obert G. Sodium valproate, an
anticonvulsant drug, stimulates human cytomegalovirus replication.
J Gen Virol 1995; 76: 1409-15.
24 Michaelis M, Ha TA, Doerr HW,
Cinatl Jr J. Valproic acid interferes with antiviral
treatment in human cytomegalovirus-infected endothelial cells.
Cardiovasc Res 2008; 77: 544-50.
25 Zhou J, Snyder AR, Lieberman PM. Epstein-Barr
virus episome stability is coupled to a delay in replication
timing. J Virol 2009; 83: 2154-62.
|
Version imprimable
Version PDF
