Therapies (2019) 74, 527—530

Available online at

ScienceDirect
www.sciencedirect.com

PHARMACOVIGILANCE

Bullous fixed drug eruption: A potential diagnostic

pitfall: a study of 18 cases
Anissa Zaouak a,∗, Fatma Ben Salem b,
Sélima Ben Jannet a, Houda Hammami a,
Samy Fenniche a

a
Dermatology Department, Habib Thameur Hospital, 1089 Tunis, Tunisia
b
National Center of Pharmacovigilance, 1002 Tunis, Tunisia

Received 15 September 2018; accepted 21 January 2019

Available online 31 March 2019

KEYWORDS Summary
Adverse drug Background. — Bullous fixed drug eruption (BFDE) is a rare and particular adverse drug reac-
reaction; tion characterized by localized or generalized blisters and erosions, which can be confused
Bullous fixed drug with Stevens-Johnson syndrome, toxic epidermal necrolysis, major erythema multiforme and
eruption autoimmune bullous dermatosis.
Objective. — The aim of our study was to assess the epidemiological, clinical and therapeutic
features and outcome of BFDE.
Methods. — A retrospective and descriptive study collecting all observations of BFDE was con-
ducted in the dermatology department of Habib Thameur Hospital in Tunisia, over an 18-year
period (2000—2017). The diagnosis of BFDE was confirmed by histopathological examination and
all the patients underwent pharmacovigilance investigation.
Results. — Totally, 18 cases were enrolled in our study with BFDE. The mean age was 57.9 years
with a sex ratio M/F of 1. BFDE was localized in 8 cases and generalized in 10 cases. It was the
first episode of BFDE in 11 patients and a recurrence in 7 patients. Drugs involved in the genesis
of BFDE in our study were mainly non-steroidal anti-inflammatory drugs in 10 patients and
antibiotics in 5 cases. Drug patch tests were performed in four patients on the residual plaques
of FDE (fixed drug eruption) and were positive to the suspected drug. A favorable outcome was
observed in all our patients under treatment and after suspected drug withdrawal.
Conclusion. — BFDE is a rare adverse drug reaction and could be severe especially when it
presents as a generalized eruption. Drugs involved are mainly non-steroidal anti-inflammatory
drugs followed by antibiotics.
© 2019 Société française de pharmacologie et de thérapeutique. Published by Elsevier Masson
SAS. All rights reserved.

∗ Corresponding author at: Dermatology Department, Habib Thameur Hospital, 8 street Ali Ben Ayed, Montfleury, 1089 Tunis, Tunisia.
E-mail address: anissa zaouak@yahoo.fr (A. Zaouak).
https://doi.org/10.1016/j.therap.2019.01.009
0040-5957/© 2019 Société française de pharmacologie et de thérapeutique. Published by Elsevier Masson SAS. All rights reserved.
528 A. Zaouak et al.

Abbreviations

BFDE bullous fixed drug eruption

FDE fixed drug eruption
NSAIDs non-steroidal anti-inflammatory drugs
SJS Stevens-Johnson syndrome
TEN toxic epidermal necrolysis

Introduction
Fixed drug eruption (FDE) is a common adverse drug
reaction characterized by single or multiple round ery- Figure 1. Round erythematous plaque with a blister on the center
thematous plaques leaving a residual pigmented scar. located on the trunk.
It recurs at the same skin or mucosal sites, with the
possibility of appearance of other plaques, each time
the offending agent is taken [1,2]. The most common
offending agents include antimicrobials, non-steroidal anti-
inflammatory drugs (NSAIDs), and antiepileptic drugs [1,3].
Bullous fixed drug eruption (BFDE), a rare and severe
variant of FDE, is characterized by localized or generalized
blisters and erosions. Because of the extensive cutaneous
or mucosal involvement in the generalized form of BFDE,
clinically, it is sometimes difficult to differentiate it from
Stevens-Johnson syndrome (SJS), toxic epidermal necrolysis
(TEN), major erythema multiforme and autoimmune bullous
dermatosis [4]. In the literature, we found many case reports
that described BFDE but our study is, to our knowledge, the
first study that enrolled 18 cases of BFDE over an 18-year
period.

Materials and methods

A retrospective, descriptive and monocentric study collect-
ing all observations of BFDE in the dermatology department
of Habib Thameur hospital in Tunisia, during 18 years from
January 2000 to December 2017 was undertaken. We studied
56 cases of FDE among them 18 cases of BFDE were included.
Diagnosis of BFDE was based on clinical and histopathological
findings as well as pharmacovigilance investigations. Data
collection was based on patients’ medical records informa-
tion and included age at presentation, gender, morphology
topography and number of lesions, offending drug, allergo-
logic tests and histopathological findings. Drug imputability
was assessed according to the French method of Miremont- Figure 2. a: a blister overlying a violaceous plaque located on the
Salamé G et al [5]. Drug patch test was performed when the elbow; b: multiple round erythematous plaques on the face with a
patient was polymedicated. It was done, 6 weeks after the blister overlying a violacous plaque on the upper lip.
appearance of the eruption, on residual pigmented lesion,
with exposure for 48 hours according to European Society
of Contact Dermatitis guidelines. It was performed with the The male to female ratio recorded was 1. Clinically, BFDE
drug in its commercial form diluted to 30% in petrolatum presented as blisters with hemorrhagic content on erythe-
[6]. matous and violaceous skin (Figs. 1 and 2a, b). For some
patients, blisters were eroded leaving extensive cutaneous
erosions. BFDE was localized in 8 patients and generalized
Results in 8 patients. An involvement of both skin and mucosa was
present in 7 patients, the skin in 8 patients and affecting
The incidence of FDE was evaluated to 0.9 cases/10,000 only the mucosa in 3 patients. The number of bullous lesions
consultants/year. The ages of our patients varied from 2 varied from one plaque with bullous center or post-bullous
to 80 years old (median: 51.5 years). Our patients didn’t erosion to multiple blisters and post-bullous erosions. The
have a drug hypersensitivity history before BFDE occurred. topography of lesions predominated on the upper limbs
Bullous fixed drug eruption: Potential diagnostic pitfall 529

lesions and a generalized form in 2 patients with multiple

Table 1 Offending drugs of bullous fixed drug eruption
erosive plaques affecting the skin and mucosa. In fact,
in our patients.
recurrences in our patients occurred after unconscious
Offending drugs Number of cases intake of the offending drug. In those patients, the first
Non steroidal anti-inflammatory 10 episode of BFDE was initially misdiagnosed and the patients
Mefenamic acid 4 continued to take the same offending drug, which resulted
Ibuprofene 4 in recurrences. Concerning allergological investigations, 4
Diclofenac 1 patients, who were polymedicated, had drug patch tests
Oxicam 1 to identify the offending drug. They were positive to
Antibiotics 5 doxycycline in 2 cases and to mefenamic acid in 2 cases.
Doxycycline 3
Cotrimoxazole 2
Others 3 Discussion
Tenoic acid 2
Paracetamol 1 Our study reported the main clinical features of BFDE,
which is a severe, and a rare adverse drug reaction. Many
case reports had described this entity but this is, to our
knowledge, the first study of BFDE reporting 18 cases. The
followed by the genital organs, the trunk, the lower incidence of FDE was 0.9 cases/10,000 consultants/year.
limbs and the face. Ten patients had mucosal involve- The frequency of the bullous form in our study was 29.62%.
ment. Erosions of the genital mucosa occurred in 8 male Through some series published in the literature, it var-
patients affecting the glans and 2 female patients affecting ied from 6.7% in Iranian series of 30 cases of FDE [7] to
the vulva. On histopathological examination, we found 50.8% in French series of 59 cases of FDE [8] revealing
keratinocytic necrosis, which was sometimes confluent, of the heterogeneousness of the series. Hence, it seems
responsible for epidermal detachment. There was also that the frequency of fixed drug eruption varies consid-
superficial and deep perivascular and interstitial infiltrate erably from one continent to another and on the same
of lymphocytes, neutrophils and eosinophils. Sometimes continent from one country to another [8,9]. BFDE affects
there was an accumulation of melanophages in the super- equally male and female without gender predominance [10]
ficial dermis. Direct immunofluorescence, performed in 5 such as our study and can affect all ages both children
patients who had generalized BFDE, was negative, which and the elderly with predominance for young adult [11].
allowed us to rule out the diagnosis of autoimmune bullous This can be explained by the automedication in this age
dermatosis. All the patients underwent pharmacovigilance group as it showed the highest percentage of NSAIDs, as
investigation, which allowed us to incriminate NSAIDs in 10 the most offending drug in BFDE, which is used frequently
cases (4 cases respectively with mefenamic acid and ibupro- as antipyretic, anti-inflammatory and analgesic. Clinically,
fen, one case respectively with diclofenac and oxicam), BFDE is characterized by violaceous plaques with blisters or
antibiotics in 5 cases (3 cases with doxycycline and 2 cases erosions in the center. It can be localized or generalized.
with cotrimoxazole), tenoic acid in 2 cases and paracetamol Generalized form is characterized by many well-defined
only in one case (Table 1). The offending drugs were vesiculo-bullous and edematous plaques (>10) on erythema-
contraindicated in our patients. The delay of appearance of tous and violaceous skin and interesting at least 3 locations.
BFDE after drug ingestion varied from few hours to few days. In our study, the most common location was the upper limbs
In fact, initially the diagnosis of BFDE was misdiagnosed in followed by the external genital organs and then the lower
the patients who had a NSAIDs intake and they consulted limbs. This has been confirmed in the different series of the
after the second or the third recurrence of BFDE. Indeed, literature. The mucosal form may have a bullous, erosive,
after a careful medical history, the patients reported that aphthoid or erythematous aspect and usually does not leave
the appearance of BFDE after drug intake of NSAIDs was pigmented scars [1,3]. It affected men more than women
shorter since they have already been exposed before to the especially in the genital location. In fact, genital lesions
same drug. Hence, the time between BFDE occurrence and are frequent in the BFDE mainly affecting the glans in male
medical consultation varied between 24 hours and 3 days. patients. Genital mucosal form is rare in female patients
The outcome was favorable after drug discontinuation. and was present in 2 cases in our patients. It is important to
Ten patients were treated with dermocorticoids and oral know that genital form regresses without residual scar and
corticosteroids 0.5 mg/kg/day. Eight patients were treated do not affect the fertility. This study highlights the fact that
with only dermocorticoids. Healing of cutaneous erosions BFDE could be easily misdiagnosed with the other bullous
was total and progressive with a delay of 7 to 15 days. No dermatosis, such as pemphigus, bullous pemphigoid, major
cases of cutaneous superinfections were reported during erythema multiform but also other more severe adverse
the healing process. The healing was progressive leaving drug reaction like SJS or TEN. Hence, combining a careful
a residual pigmented scar. In patients with only mucosal medical history, clinical and histopathological findings are
lesions, healing was obtained without pigmented scars. In important to assess an accurate diagnosis. The diagnosis of
our study, 11 patients consulted after the first episode and BFDE is suspected clinically and confirmed histolopathog-
7 patients consulted after the 2nd or 3rd episode of BFDE. ically. The histolopathological examination in acute BFDE
Patients who had pharmacovigilance investigation didn’t highlights numerous, isolated or confluent and eosinophilic
have a recurrence. Seven patients had recurrences mani- keratinocyte necrosis often associated with basal vacuoliza-
festing as a localized form at the initial site of cutaneous tion, whose extension is responsible for bullous forms.
530
Inflammatory infiltrates are often important polymorphic
and associated sometimes with marked edema of the pap-
illary dermis extending to the middle dermis. They are
composed of lymphocytes associated with many polynuclear
neutrophils and often polynuclear eosinophils. In the qui-
escent phase, the epidermis returns to a normal aspect,
polynuclear cells disappear and melanophages settle in the
papillary dermis with some lymphocytes. Recurrent forms
show acute phase lesions associated with melanophages
[1,12—14]. In generalized bullous forms that can mimic
histopathologically autoimmune bullous dermatosis, a direct
immunofluorescence is essential to assess the diagnosis. It
was perfomed in five of our patients and was negative. FDE
is a hypersensitivity reaction most often to a drug intake.
In rare cases, FDE may be due to a food intake such as
liquorice, propolis or even at a solar exposure [15]. Accord-
ing to Mahboob et al [9], which reported the largest series
of FDE in the literature including 450 cases, it appears that
drugs inducing BFDE are the same as drugs inducing FDE
and according to Brahimi et al [8], there is no relationship
between FDE severity and drug class. In our study, com-
paring our 18 patients with BFDE and the other 38 cases
of FDE without bullae, it appears that the main offend-
ing drugs were NSAIDs and antibiotics such as doxycycline
and cotrimoxazole in the two groups. Pharmacovigilance
investigation using drug imputability scores was crucial to
determine the offending drug especially if the patient is
polymedicated. The definitive diagnosis of drug imputabil-
ity is the drug patch test. In case of negativity of drug patch
tests, we can use the oral provocation test but starting with
doses lower than usual doses [16]. Oral provocation test
remains the most objective examination in determining the
offending drug but which may induce severe reactions and
is contraindicated in patients who have had generalized bul-
lous forms. The prognosis of the BFDE depends on the extent
of the erosive lesions surface and past medical history of the
patient [17]. Usually, erosions heal in 1 to 2 weeks leaving
pigmented scars that can last for months, which may cause
cosmetic discomfort in these patients. Unlike the cutaneous
lesion that leaves after healing a pigmented scar, the ero-
sions of the oral and genital mucosa don’t leave pigmented
scars. Recurrences may be localized or generalized with new
patches in old and new locations. Cutaneous and genital
post-bullous erosions can be an infectious gateway, hence
the importance of skin care with antiseptics.
Conclusion
In summary, this study tried to highlight this rare BFDE vari-
ety that can be confused with other autoimmune bullous
dermatosis or other more severe adverse drug reactions like
toxic epidermal necrolysis. The purpose of this study was
to describe the clinical features of this adverse drug reac-
tion that frequently affects the mucous membranes and to
emphasize the contribution of the pharmacovigilance inves-
tigations in the identification of suspected drug and thus
prevent further recurrence.
A. Zaouak et al.
Disclosure of interest
The authors declare that they have no competing interest.
References
[1] Ozkaya E. Fixed drug eruption: state of the art. J Dtsch Der-
matol Ges 2008;6(3):181—8.
[2] Sehgal VN, Srivastava G. Fixed drug eruption (FDE):
changing scenario of incriminating drugs. Int J Dermatol
2006;45(8):897—908.
[3] Valeyrie-Allanore L, Lebrun-Vignes B, Bensaid B, Sassolas
B, Barbaub A. Fixed pigmented erythema: epidemiol-
ogy, physiopathology, clinical features, differential diagno-
sis and therapeutic management. Ann Dermatol Venereol
2015;142(11):701—6.
[4] Cho YT, Lin JW, Chen YC, Chang CY, Hsiao CH, Chung WH,
et al. Generalized bullous fixed drug eruption is distinct
from Stevens-Johnson syndrome/toxic epidermal necrolysis
by immunohistopathological features. J Am Acad Dermatol
2014;70(3):539—48.
[5] Miremont Salamé G, Theophile H, Haramburu F, Bégaud B.
Causality assessment in pharmacovigilance: the French method
and its successive updates. Therapie 2016;71:179—86.
[6] Barbaud A, Goncalo M, Bruynzeel D, Bircher A. Guidelines
for performing skin tests with drugs in the investigation
of cutaneous adverse drug reactions. Contact Dermatitis
2001;45(6):321—8.
[7] Kavoussi H, Rezaei M, Derakhshandeh K, Moradi A, Ebrahimi A,
Rashidian H, et al. Clinical features and drug characteristics
of patients with generalized fixed drug eruption in the West of
Iran (2005—2014). Dermatol Res Pract 2015;2015:236703.
[8] Brahimi N, Routier E, Raison-Peyron N, Tronquoy AF, Pouget-
Jasson C, Amarger S, et al. A three-year-analysis of fixed
drug eruptions in hospital settings in France. Eur J Dermatol
2010;20(4):461—4.
[9] Mahboob A, Haroon TS. Drugs causing fixed eruptions: a study
of 450 cases. Int J Dermatol 1998;37(11):833—8.
[10] Sharma VK, Dhar S. Clinical pattern of cutaneous drug erup-
tion among children and adolescents in north India. Pediatr
Dermatol 1995;12(2):178—83.
[11] Ramam M, Pandhi RK. Fixed drug eruption in infancy. Indian
Pediatr 1993;30(1):110—1.
[12] Patell RD, Dosi RV, Shah PC, Joshi HS. Widespread bullous fixed
drug eruption. BMJ Case Rep 2014 [pii: bcr2013200584].
[13] Bataille M, Vonarx M, Vermersch-Langlin A. Illustration of
diagnostic and prognostic difficulties during the early stages
of generalized bullous fixed drug eruptions. Eur J Dermatol
2017;27(1):86—8.
[14] Özkaya E. Oral mucosal fixed drug eruption: characteristics and
differential diagnosis. J Am Acad Dermatol 2013;69(2):e51—8.
[15] Benomar S, Ismaili N, Koufane J, Senouci K, Hassam B. Fixed
food eruption caused by liquorice. Ann Dermatol Venereol
2010;137(2):121—3.
[16] Barbaud A. Drug patch tests in the investigation of cuta-
neous adverse drug reactions. Ann Dermatol Venereol
2009;136(8—9):635—44.
[17] Lipowicz S, Sekula P, Ingen-Housz-Oro S, Liss Y, Sassolas B,
Dunant A, et al. Prognosis of generalized bullous fixed drug
eruption: comparison with Stevens-Johnson syndrome and
toxic epidermal necrolysis. Br J Dermatol 2013;168(4):726—32.