ARTICLE
Polymorphic light eruption (PLE) is a common acquired photodermatosis
of unknown etiology [1-7]. Its diagnosis is usually based on a typical
clinical history and typical skin manifestations. In cases of ambiguous
presentation, histopathological examination and provocative phototesting
[2] can help to establish the diagnosis. To help improve these diagnostic
tools, we conducted a retrospective study, in which we characterized 133
patients with PLE seen over 14 years in Styria, Austria.
Materials and methods
Patients
In a retrospective study, we evaluated the clinical, laboratory, phototest,
and phototherapy data from 133 patients with PLE who presented at the
Phototherapy Unit, Department of Dermatology, University of Graz, between
May 1982 and June 1996. There were 109 females (82%) with a median age
of 33 years (range, 4-62 years) and 24 males (18%) with a median age of
35 years (range, 9-56 years). To provide specific data on their disease,
all patients had been asked to fill out a specific photodermatosis questionnaire
at first presentation. The completed questionnaires revealed data on the
patients' histories, characteristics, and the clinical features of their
PLE.
Laboratory studies
In 33 patients, antinuclear antibody (ANA) screening was performed using
an enzyme-linked immunosorbent assay (VARELISA; elias Medizintechnik Gmbh,
Freiburg i. Bv., Germany) and/or a commercial assay based on a human epithelial
cell substrate (HEp-2) (Kallestad Diagnostics, South Austin, TX, USA).
Antibodies to Ro (SSA), La (SSB), Sm, Scl-70, and Jo-1, were assayed using
the standard double-immunodiffusion technique on Ouchtalony plates, with
rabbit thymus and spleen extract as the antigen sources or by ELISA (Synelisa;
elias Medizintechnik GmbH, Freiburg i. Bv., Germany). Antibodies to native
DNA were detected by the Crithidia luciliae indirect immunofluorescence
assay (BIOS, GmbH Munich, Germany). In certain cases, punch biopsies were
taken from skin lesions, fixed in formaldehyde, processed routinely, and
stained with hematoxylin-eosin. For direct immunofluorescence studies,
skin specimens were incubated with fluorescein isothiocyanate (FITC)-labeled
goat antihuman IgG, IgA, IgM, C3, or fibrin antibodies.
Radiation sources for phototesting and phototherapy
UVA phototesting was performed with a high-intensity UVA source (Sellas
Sunlight 2001; Sellas Medizinische Geräte, Gevelsberg, Germany) equipped
with a metal halogenide lamp and a 5 mm-thick glass filter, as previously
described [2]. UVB phototesting was performed with a Sellas 1200 UVB source
(Sellas Medizinische Geräte). UVB and UVA treatments were administered
with either a Waldmann 8001, Waldmann 7001K, or Waldmann 4001 system (Waldmann
Medizintechnik, Schwenningen, Germany) equipped with Sylvania FR 90 T12
or Philips TL 09 tubes. UVA treatment was administered with a broad-band
UVA system (Sellas 12000; Sellas Medizinische Geräte) at an intensity
of 30-60 mW/cm2.
MED Phototesting
The minimal erythema doses of UVA (MED-UVA) and UVB (MED-UVB) were determined
by exposing six template test areas on the flexor aspect of each patient's
left and right forearm respectively, to a ladder of UV doses. The routine
test doses used were 28, 40, 56, 80, 112, and 160 J/cm2 of
UVA radiation and 17, 24, 34, 48, 68, and 96 mJ/cm2 of UVB
radiation.
Photoprovocative testing
Photoprovocative testing was performed on two symmetrically located
6 x 6 cm test areas on previously affected skin (mainly the upper chest
area). The test areas were exposed daily to either UVA or UVB radiation
on 3-5 consecutive days. The first UV dose given was 70% of the MED-UVA
or MED-UVB. The daily radiation dose was then increased 20%, until erythema
appeared. If the resulting erythema was slight, the next radiation dose
was not increased; if the erythema was strong, the next radiation dose
was decreased by 20%. A positive test result was defined as the reproduction
of PLE-like lesions in an exposed test area.
Phototherapy
To prevent PLE, patients received either broad-band UVB phototherapy,
UVA phototherapy, or psoralen and UVA (PUVA) photochemotherapy between
March and May of a year. The standard UVB treatment lasted 4-6 weeks,
its schedule depending on skin phototype: in patients with skin phototype
I/II, the first three treatments lasted 10, 20, and 30 sec each (20, 40,
and 60 mJ/cm2, respectively); in patients with skin phototype
III/IV, 15, 30, and 45 sec each (30, 60, and 90 mJ/cm2, respectively).
Thereafter, the exposure time was increased approximately 10-20% weekly.
The standard UVA treatment lasted 4-6 weeks at a dose of 10 J/cm2
3 to 4 times weekly. In most patients, the treatment dose was increased
5 J/cm2 weekly. PUVA therapy in most cases was given three
times a week for 4 weeks with 5-methoxypsoralen (Geralen®;
Gerot Pharmazeutika, Vienna, Austria). The PUVA treatment dose schedule
for patients with skin phototype III/IV was 3, 4, 5, and 7 J/cm2
and for patients with skin phototype I/II 2, 3, 4, 5 J/cm2
in weeks 1, 2, 3, and 4, respectively. In certain cases, 8-methoxypsoralen
(Oxsoralen®; Gerot Pharmazeutika, Vienna, Austria) was
applied instead of 5-methoxypsoralen. The starting dose was then 50% of
the minimal phototoxic dose (MPD) as determined by phototesting. The UVA
dose in the PUVA regimens was increased approximately 10-20% weekly, depending
on the individual erythema response, and therapy was given for a total
of 4 weeks.
Statistical analysis
For statistical analysis, comparisons between different groups of patients
were made using the khi2-test. A p-value < 0.05 was
considered to indicate a statistically significant difference between
two groups of patients.
Results
Clinical features
Questionnaires completed by 133 subjects with PLE were evaluated. The
response rates for the specific questions were as follows: age at onset,
78% (104/133); duration of disease, 87% (116/133); skin phototype, 95%
(126/133); seasonal occurrence, 74% (98/133); latent interval, 63% (84/133);
persistence of skin lesions, 67% (89/133); body site distribution, 92%
(122/133); morphological type, 77% (103/133); effect of sunscreen, 65%
(87/133); and result of preventive phototherapy 86% (79/92).
The age of patients at onset of PLE ranged from 3 to 62 years (median
age, 26 years). In most cases, onset occurred in the second or third decade
(Fig. 1). The duration
of PLE at first presentation ranged from 1 week to 25 years (median duration,
6.5 years). The skin phototype distribution was as follows: 4% (5/126)
skin phototype I, 47% (59/126) skin phototype II, 49% (62/126) skin phototype
III, and 0% (0/126) skin phototype IV. The seasonal distribution of skin
manifestations in 98 patients was as follows: 8 patients (8%) reported
PLE manifestation only in spring, 25 patients (25%), in spring and summer;
49 patients (50%), only in summer. In 19 patients (17%), there was no
seasonal association. The body site distribution of skin lesions is shown
in Table I. The most commonly
affected site was the V of the neck (80%). The morphological types of
PLE and their frequency are given in Table
II. The most common morphological type was papulovesicular (31%),
followed by plaque (26%), papular (25%), vesiculobullous (16%), and erythema
multiforme-like (2%).
The latent interval between light exposure and appearance of skin lesions
ranged from several minutes to 120 hrs (Fig.
2). There were two peaks in the distribution curve of the individual
latent intervals. These peaks in the distribution were evident for the
lower limit (Fig. 2A),
the upper limit (Fig. 2B),
and the mean of the latent interval (Fig.
2C). The peaks in the distribution curve of the mean latent interval
occurred at 1-1.5 and 24 hrs, respectively. For statistical analysis,
the patients were divided into two groups according to these peaks. The
first group (n = 58) contained those patients with a mean latent interval
of 12 hrs or less while the second group (n = 26) contained those patients
with a mean latent interval of more than 12 hrs. The following differences
between these two groups were statistically significant: Six percent of
patients in the first group developed their PLE-lesions only in more southern
countries but not in Austria versus 25% of patients in the second
group (khi2 = 6.155; p = 0.013). Forty-four percent
of patients in the first group noticed their PLE lesions also in the winter
period versus 13% of patients in the second group (khi2
= 7.031; p = 0.008). With regard to body site distribution the
analysis revealed that there was no statistically significant difference
in the involvement of the V of the neck (48/58, 82.8% versus 22/26,
84.6%) between the group with the latent interval of 12 hrs or less versus
that with more than 12 hrs, respectively, but the face was significantly
more often affected in the former than in later group (26/58, 44.8% versus
4/26, 15.4%) (khi2 = 6.778; p = 0.009). There was no
significant difference between the two groups in terms of sex, age, duration
of disease, skin phototype, persistence of skin lesions, morphological
type, ANA positivity, effect of sunscreen, or phototest and phototherapy
results.
The persistence of PLE lesions after light exposure ranged from 30 min
to 72 days (Fig. 3). The
main peak in the distribution curve of the individual persistence time
was at 2.5 days. For statistical analysis, patients were divided into
two groups according to the persistence of their skin lesions. The first
group (n = 67) contained those patients with a mean persistence time of
5 days or less, while the second group (n = 22) comprised those patients
with a mean persistence time of more than 5 days. The following differences
between the two groups were statistically significant: 56% of patients
in the first group found their skin manifestations diminished during summer
versus 24% of patients in the second group (khi2 = 5.546;
p = 0.019). There was no significant difference between the two
groups in terms of sex, age, duration of disease, skin phototype, latent
interval, morphological type, body site distribution, ANA positivity,
effect of sunscreen, or phototest and phototherapy results.
Results of laboratory testing
ANA testing was positive in three males and three females of 33 subjects
tested (Table III). Four
patients were ANA-positive (titer > 1:80) as determined by the HEp-2
epithelial cell culture assay and two patients were ANA-positive by ELISA.
Histopathological and direct immunofluorescence studies were carried out
in three of six ANA-positive patients. One patient (SR) showed immunoglobulin
A deposits at the dermoepidermal junction (Table
III). The follow-up of the ANA-positive patients ranged from 1
to 5 years. According to American Rheumatism Association (ARA) criteria,
no patient had or developed systemic lupus erythematosus (SLE) (ARA criteria
> 4) during follow-up.
Results of phototesting
Phototesting was performed in 30 patients between February and April.
The mean MED-UVA was 124 J/cm2 (range, 28-215 J/cm2),
and the mean MED-UVB was 51 mJ/cm2 (range, 18-129 mJ/cm2).
These values fell within the normal limits used in our laboratory. Provocative
phototesting was positive in 17 patients (57%) (Table
IV). These patients had developed PLE-like lesions in the test
area after 1-3 UV exposures. The action spectrum fell within the UVA range
in 10 patients (59%), the UVB range in 4 (23%), and both ranges in 3 (18%).
Results of phototherapy and photoprotection
Follow-up data were available for 79 of 92 patients who had received
preventive phototherapy. Fifty-six of those patients were treated with
broad-band UVB, 6 patients with broad-band UVA, and 17 patients with PUVA.
The protection rates in the first summer season after therapy are given
in Table V. The complete
protection rate in the first summer season after therapy was 27% for UVB,
0% for UVA, and 53% for PUVA therapy. The overall protection rate (including
partial and complete responders) was 83% for UVA, 82% for UVB and 65%
for PUVA.
In contrast, the patients' histories revealed that the use of a sunscreen
with a mean sun protection factor (SPF) of 14 did not prevent skin lesions
in 88% of PLE patients.
Discussion
We report here the clinical, laboratory, phototest, phototherapy findings
in 133 patients with PLE over a 14-year period from 1982 to 1996. The
sex distribution of the patients, age at onset of skin manifestations,
seasonal occurrence, body site distribution, and morphological type of
skin lesions were consistent with those given in previous reports on PLE
[1, 3, 7]. PLE was approximately four times more common in females than
males. In most patients, the onset of skin manifestations came in the
second or third decade. The most commonly affected body site was the V
of the neck. The most common morphological types of skin lesion were papular,
papulovesicular, vesiculobullous, and plaque.
Interestingly, we found two peaks in the distribution curve of the individual
latent interval, i.e., the interval between sun exposure to the
appearance of skin lesions. The first peak was at 1-1.5 hrs and the second
peak at 24 hrs. Importantly, these peaks may be of significance with regard
to the possible heterogeneity of PLE.
Benign summer light eruption has been differentiated from common PLE
particularly by French authors [8-12]. One of the major criteria to differentiate
BSLE from PLE was reported to be latency shorter than 12 hrs for the former
disease when compared to the latter [11]. Thus, our observation of two
peaks in the latent interval somehow supports the concept of BSLE. However,
with regard to body site distribution, a major criterion for BSLE is that
the V of the neck is affected most often and the face should always be
spared [11]. In our study, there was no statistically significant difference
in the involvement of the V of the neck (82.8% versus 84.6%) between
the group with latent interval of 12 hrs or less versus that with
more than 12 hrs, respectively, but the face was significantly more often
affected in the former than in latter group (44.8% versus 15.4%).
Thus, this observation rather contradicts the concept of BSLE. Indeed,
the existence of the entity of BSLE was recently questioned by other authors
because they did not find a difference in certain immunohistochemical
stainings of skin biopsies and phototest results between BSLE and PLE
[13, 14]. Alternatively, the two PLE latency peaks observed in our study
may have been due to different threshold levels of light intensity and/or
dose necessary to produce the skin lesions. Thus, for the group with the
long latent interval (> 12 hrs) it may simply have taken longer in
general to receive the dose necessary for skin manifestations to develop.
In addition, we observed large variance in the distribution curve of the
persistence of skin lesions. Patients with a long persistence of skin
lesions (> 5 days) became less tolerant of sunlight during summer than
patients with a short persistence (¾ 5 days). The variance in the
distribution curve of the persistence of skin lesions again suggested
the possible involvement of different pathogenic mechanisms in PLE.
Of 33 patients tested, 6 (18%) were ANA-positive. None of the 6 patients,
however, fulfilled the ARA criteria for diagnosis of SLE at presentation
or on follow-up (1 to 5 years). Other authors have also reported the presence
of ANA in patients with PLE. For instance, Murphy et al. [5] and
Petzelbauer et al. [6] found that approximately 8-12% of patients
with PLE were ANA-positive. More recently, Nyberg et al. [15] reported
that clinical symptoms of PLE seem to be common in patients with both
systemic and cutaneous lupus erythematosus. They suggested that the two
conditions may often coexist and in about half of the cases PLE may precede
lupus erythematosus. Furthermore, the two diseases may share pathogenic
factors and PLE might predispose to lupus erythematosus in a subgroup
of PLE patients [15]. However, longer follow-up may be necessary to exclude
the possibility that lupus erythematosus develops in patients with PLE
since in the study by Nyberg et al. in about half of the patients
with PLE, lupus erythematosus was diagnosed more than 7 years after the
onset of PLE [15].
PLE lesions were reproducible in 57% of patients tested. In addition,
the action spectrum fell within the UVA range in 59% of patients; the
UVB range in 23%; and both ranges in 18%. Phototherapy in early spring
was effective in preventing PLE in most of our patients. The highest complete
protection rates were observed after PUVA (53%) and after UVB therapy
(27%). The best overall protection rate including partial and complete
responders was observed after UVA (83%), followed by UVB (82%) and PUVA
(65%). However, these rates are possibly biased because PUVA was given
to patients with severe symptoms of PLE while UVA and UVB therapy were
given to patients with minor symptoms. High photoprotection rates have
been previously reported for the preventive treatment in PLE with both
UVB [16-18] and PUVA [2, 12, 17, 18], although broad band UVB seemed to
be slightly less effective than PUVA [17, 18]. However, Bilsland et
al. [19] recently demonstrated that the treatment with the novel 311
nm narrow band UVB therapy was as effective as PUVA in the prevention
of PLE. Our study confirmed the results of some previous reports suggesting
that UVA therapy may be clinically useful in preventing PLE [20, 21].
In contrast, the patients' histories revealed that sunscreens could not
prevent skin manifestations in 88% of our patients with PLE.
In short, the most important findings of the present study are the different
peaks in the distribution curve of the individual latent interval between
sunlight exposure to the appearance of skin lesions and the variance of
persistence of sunlight-induced skin lesions. Further studies are necessary
to better characterize these factors, which may be of significance in
the etiology and pathogenesis of PLE.
CONCLUSION
Acknowledgements
We want to thank Dr. A. Hofer, Dr. E. Rieger, and Dr. W. Salmhofer,
Department of Dermatology, Karl-Franzens-University, Graz, Austria, for
their help in the collection of clinical data.
REFERENCES
1. Hölzle E, Plewig G, Hofmann C, Roser-Maass E. Polymorphous light
eruption. J Am Acad Dermatol 1982; 7: 111-25.
2. Ortel B, Tanew A, Wolff K, Hönigsmann H. Polymorphous light
eruption: action spectrum and photoprotection. J Am Acad Dermatol
1986; 14: 748-53.
3. Hölzle E, Plewig G, Kries R, Lehmann P. Polymorphous light eruption.
J Invest Dermatol 1987; 88 (suppl.): 32-8.
4. Hönigsmann H, Ortel B. Die Polymorphe Lichtdermatose-Photobiologische
Diagnostik und Therapie. Z Hautkr 1988; 63: 676-8.
5. Murphy GM, Hawk JLM. The prevalence of antinuclear antibodies in
patients with apparent polymorphic light eruption. Br J Dermatol
1991; 125: 448-51.
6. Petzelbauer P, Binder M, Nikolakis P, Ortel B, Hönigsmann H.
Severe sun sensitivity and the presence of antinuclear antibodies in patients
with polymorphous light eruption-like lesions. J Am Acad Dermatol
1992; 26: 68-74.
7. Bergner T, Przybilla B, Heppeler M. Polymorphe Lichtdermatose. Hautarzt
1993; 44: 215-20.
8. Bonvalet D, Baddoura R, Jeanmougin M. Intérêt de l'étude
histologique du phototest dans les photodermatoses. Ann Dermatol Venereol
1986; 113: 1205-12.
9. Jeanmougin M, Civatte J. Benign summer light eruption: a new entity?
[letter]. Arch Dermatol 1986; 122: 376.
10. Jeanmougin M, Civatte J. Benign summer light eruption: the most
common photodermatosis [letter]. J Am Acad Dermatol 1987; 17: 690-1.
11. Guarrera M, Micalizzi C, Rebora A. Heterogeneity of polymorphous
light eruption: a study of 105 patients [letter]. Arch Dermatol
1993; 129: 1060-1.
12. Leonard F, Morel M, Kalis B, Amblard P, Avenel Audran M, Beani JC,
Bonnetblanc JM, Leroy D, Marguery MC, Peyron JL, et al. Psoralen
plus ultraviolet A in the prophylactic treatment of benign summer light
eruption. Photodermatol Photoimmunol Photomed 1991; 8: 95-8.
13. Verheyen AM, Lambert JR, Van Marck EA, Dockx PF. Polymorphic light
eruption an immunopathological study of provoked lesions. Clin
Exp Dermatol 1995; 20: 297-303.
14. Lambert J, Verheyen A, Dockx P. Experimental reproduction of polymorphous
light eruption and benign summer light eruption by whole-body UVA irradiation.
Dermatology 1997; 194: 388-91.
15. Nyberg F, Hasan T, Puska P, Stephansson E, Hakkinen M, Ranki A,
Ros AM. Occurrence of polymorphous light eruption in lupus erythematosus.
Br J Dermatol 1997; 136: 217-21.
16. Morison WL, Momtaz K, Mosher DB, Parrish JA. UVB phototherapy in
the prophylaxis of polymorphous light eruption. Br J Dermatol 1982;
106: 231-3.
17. Addo HA, Sharma SC. UVB phototherapy and photochemotherapy (PUVA)
in the treatment of polymorphic light eruption and solar urticaria. Br
J Dermatol 1987; 116: 539-47.
18. Murphy GM, Logan RA, Lovell CR, Morris RW, Hawk JLM, Magnus IA.
Prophylactic PUVA and UVB therapy in polymorphic light eruption
a controlled trial. Br J Dermatol 1987; 116: 531-8.
19. Bilsland D, George SA, Gibbs NK, Aitchison T, Johnson BE, Ferguson
J. A comparison of narrow band phototherapy (TL-01) and photochemotherapy
(PUVA) in the management of polymorphic light eruption. Br J Dermatol
1993; 129: 708-12.
20. Rücker BU, Häberle M, Koch HU, Bocionek P, Schriever KH,
Hornstein OP. Ultraviolet light hardening in polymorphous light eruption
a controlled study comparing different emission spectra. Photodermatol
Photoimmunol Photomed 1991; 8: 73-8.
21. Berg M, Ros AM, Berne B. Ultraviolet A phototherapy and trimethylpsoralen
UVA photochemotherapy in polymorphous light eruption a controlled
study. Photodermatol Photoimmunol Photomed 1994; 10: 139-43.
|
Version imprimable
