The Diagnosis of Photosensitivity FREE

When a patient is examined because of a photosensitive disorder, it can sometimes be difficult to make the right diagnosis. Most patients seek consultation after the lesions have disappeared or before they reappear. Therefore, the patient's history is of critical importance.

The diagnosis can often be made solely on the basis of history. This is usually the case for polymorphous light eruption (PLE), which can be confirmed by phototesting. In other cases the history alone is insufficient for an accurate diagnosis, as is the case for chronic actinic dermatitis and solar urticaria. In such cases, phototesting is necessary. Sophisticated and expensive light sources are needed to determine a detailed action spectrum, but this is not always necessary. Adequate information can often be obtained with relatively simple light sources.

Phototesting may be performed for a variety of reasons, such as to confirm the diagnosis by provoking lesions, to establish the degree of photosensitivity, to determine an action spectrum, to evaluate the effect of a treatment, or to follow the evolution of the photodermatosis over time. Nevertheless, in some photodermatoses, the results of phototesting will be normal and, therefore, of no great value. This is the case, for example, with erythropoietic protoporphyria, for which biochemical investigations are necessary to confirm the diagnosis.

The purpose of the present article is to provide some guidelines on the examination of a photosensitive patient by presenting the important symptoms and signs, when and how to perform phototests, and what complementary examinations may be useful. Only skin disorders induced by sun exposure are covered here, and not those aggravated by sun exposure.

For a patient with a skin disorder induced by sun exposure, it is important to know at what age the symptoms first started. Therefore, the investigation of a photodermatosis and the differential diagnosis as described herein will be based on the age of the patient when the symptoms first appeared (Table 1).

When the photodermatosis starts or has started in childhood, one should consider especially PLE, including its juvenile spring eruption form, erythropoietic protoporphyria, actinic prurigo, and hydroa vacciniforme in the differential diagnosis.

In the case of the juvenile spring eruption form of PLE, the main lesions are papules, blisters, and crusts especially on the ears, although other sun-exposed areas may also be involved (Figure 1). The localization on the ears is highly characteristic (Figure 2). The lesions usually appear in the spring with the first prolonged sun exposure, often after a cold or frosty morning (March, April, or May). They will then heal spontaneously in a few days without scar formation, although this can take up to 2 weeks. Relapses are frequent every year in the springtime. In many cases, the photodermatosis will disappear after 3 or 4 years of relapses. Most of the cases have been reported in Europe.^{1 - 3} The diagnosis of this condition is mainly clinical. Histologically, subepidermal blisters are found with necrosis of the epidermis and a perivascular lymphohistiocytic infiltrate similar to that of PLE. Because the lesions usually disappear in a few days, a biopsy is not necessary. There are no other investigations available to confirm this diagnosis. Juvenile spring eruption may be considered to be an unusual form of PLE. Both conditions have similar histological and immunohistological features and may occur in the same family. Some children with juvenile spring eruption later develop a PLE.

Polymorphous light eruption of the conventional type is mainly a photodermatosis of adults but may also occur in children. It is characterized by itchy papules and sometimes vesicles, mainly in the V area of the neck and on the arms but sometimes also on the face. The face is more often involved in children than in adults. The diagnosis of a PLE is not difficult because of the typical locations and lesions, although it can be more difficult to diagnose in children than in adults. Although the diagnosis is mainly clinical, provocative phototesting may be valuable, especially in children, to confirm the diagnosis. The best way to do this is by using repetitive irradiations on the V area of the neck or forearms for 1 to 4 consecutive days.⁴ This can be done with high-intensity UV-A or with polychromatic irradiation (UV-A plus UV-B). Moderately high UV-A doses (between 60 and 100 J/cm² daily) are necessary.⁴ Quite a few UV-A fluorescent lamps are needed to administer such UV-A doses within an acceptable period. Therefore, the most appropriate light source is a high-intensity UV-A source. The polychromatic irradiation (1.5 minimal erythema dose [MED] daily) can be administered with a solar simulator (a xenon lamp with a WG 305 filter [Scott, Mainz, Germany], which eliminates the UV-C) or with a bank of broadband UV-B (Philips TL12 [NV Philips, Eindhoven, the Netherlands], Waldmann UV21 [Waldmann, GmbH, Schwenningen, Germany], Toshiba FL20S-E30 [Toshiba, Tokyo, Japan]) and UV-A fluorescent tubes (Philips Cleo Performance [NV Philips], Sylvania F85 [Sylvania, Danvers, Mass], Toshiba FL20BLB [Toshiba]), which is a less expensive method. In all these cases, readings are made immediately and up to 24 hours after the last irradiation. If the clinical lesions persist for more than 2 weeks after sun avoidance, lupus is more likely and should always be considered. Assessment of the circulating antinuclear antibody and extractable nuclear antigen titers is always mandatory. In this case, the phototesting is done in the same way as for PLE, but additional readings are necessary after 72 hours and after 1, 2, and 3 weeks. These late readings are necessary because it may take a considerable amount of time before a patient with lupus reacts to the UV exposure.

The diagnosis of erythropoietic protoporphyria is based on the typical subjective complaints of a burning pain within minutes to hours after exposure to the sun. These complaints are not necessarily accompanied by skin lesions, which occur on exposed areas and disappear when the sun is avoided. The burning pain can be even more pronounced when it is windy. In many patients, careful clinical examination will disclose discrete scar formation, which will be visible around the mouth and on the cheeks. In such patients, plasma and erythrocyte porphyrin determinations should always be done to confirm the diagnosis. Increased protoporphyrin concentrations are found in the red blood cells. In addition, about 10% to 30% of the red blood cells show an orange-red fluorescence on fluorescence microscopy, although this technique is now mostly of theoretical interest.⁵ The results of phototesting are often normal, although some patients may complain of a stinging sensation or may develop lesions.

Actinic prurigo usually manifests with itchy, often excoriated papules in most but not all patients on the bridge of the nose (Figure 3). Other areas of the face, upper part of the chest and back, and lower part of the arms and legs may also be involved. The lesions heal without scar formation. Personal or family atopy is found in many but fewer than half of the patients.^{6 - 7} Some patients with actinic prurigo even have lesions in the winter, which is not the case for the other photodermatoses. However, it is not possible to make a diagnosis of an actinic prurigo with certainty on a clinical basis. The histological findings are not typical. The diagnosis can be strongly supported by HLA-DR4 typing.^{8 - 9} Thus, in about 60% of the patients with an actinic prurigo, DRB1*0407 is present. However, this is not the case with some Colombian (HLA-C4)¹⁰ and North American (HLA-A24, C4) Indians.^{11 - 12}

On the other hand, hydroa vacciniforme is very rare, and vesicular lesions are present with marked scar formation. Lesions similar to those provoked by sun exposure can sometimes be induced by irradiating the back or forearms of the patient for 1 to 4 days with 60 to 100 J/cm² of UV-A daily with a high-intensity UV-A source.^{13 - 14} Because of the scar formation in hydroa vacciniforme, this condition must be differentiated from erythropoietic protoporphyria by means of negative results of plasma protoporphyrin determinations. In addition, erythropoietic protoporphyria usually causes early pruritus and pain on exposure to the sun.

When the photodermatosis starts in adulthood, the differential diagnosis must distinguish mainly between PLE, a drug-induced photosensitivity, and solar urticaria. Lupus erythematosus and porphyria cutanea tarda have to be excluded.

Polymorphous light eruption is by far the most common photodermatosis in adults and is also one of the easiest photodermatoses to diagnose on clinical grounds. The majority of the patients are young females. They nearly always manifest itchy papules on the sun-exposed areas that appear within hours of sun exposure, sometimes for a few consecutive days, although erythematous lesions or sometimes vesicles may also be present (Figure 4). The lesions fade within 2 to 4 days but may sometimes last longer, up to about 4 weeks. The distribution of the lesions is generally symmetrical, and, although sun-induced, they may not occur on some exposed areas of the body. In addition, covered sites may be affected if the patient wears thin clothing. The main locations are the V area of the neck and the arms (Figure 4), while the face is often free of lesions. However, when the face is involved, this may suggest a concomitant (photo) contact dermatitis to sunscreens or an additional rosacea. Phototesting is not necessary to make the diagnosis but can be useful as a confirmation or to establish the action spectrum. Phototesting can be done to provoke lesions with polychromatic irradiation or with high-intensity UV-A. Polychromatic irradiation can be administered with a solar simulator¹⁵ or with a bank of broadband UV-B¹⁶ (Philips TL12 [NV Philips], Waldmann UV21 [Waldmann, GmbH], Toshiba FL20S-E30 [Toshiba]) and UV-A (Philips Cleo Performance [NV Philips], Sylvania F85 [Sylvania], Toshiba FL20BLB [Toshiba]) fluorescent tubes.¹⁷ For up to 4 days, a daily dose of 1 to 2 MED is usually applied on the V area of the neck, the lateral part of the forearms, or, less optimally, on the back. When fluorescent tubes are used, it is important to apply the UV-B and the UV-A irradiation at the same time on the same area. Indeed, it is very difficult to provoke lesions with UV-A tubes only, unless whole-body irradiations are given.^{18 - 19} A daily dose of 20 J/cm² of UV-A for up to 5 days is usually needed to provoke lesions. Another method is to use a high-intensity UV-A source for up to 5 days with a daily dose of 60 to 100 J/cm² of UV-A on the same sun-exposed area.^{4 ,18 ,20} Lesions of PLE can also be provoked by repeated exposure to narrowband UV-B fluorescent lamps (Philips TL01 [NV Philips], Waldmann TL01 [Waldmann, GmbH]). The readings are made immediately and up to 24 hours after the last irradiation.

With drug-induced photosensitivity, the patient will have lesions on the face and other sun-exposed areas, whereas with PLE the face is often free of lesions. In addition, there will be a history of photosensitizing drug intake. Phototoxic drugs usually give an exaggerated sunburn-type reaction with or without a burning feeling, while the much rarer photoallergic drugs may give a more eczematous type of reaction with itching. Phototesting may be used to confirm the diagnosis. The phototesting has to be done twice, during the drug intake and 1 month after stopping the drug, thus showing a normalization of the photosensitivity. The optimal light source is a xenon lamp in combination with a monochromator to determine the degree of photosensitivity as well as the action spectrum, but a xenon lamp in combination with filters or a bank of UV-B and UV-A tubes can also be used. Drug-induced photosensitivity may also occur after topical use. However, in these cases, the photosensitivity is mostly confined to the area of application. Photopatch tests are indicated for photoallergic reactions, and testing with the agent used will be positive. It makes no sense to test phototoxic substances because they will react in most patients, depending on the UV dose applied, and because they could induce blisters and cause residual pigmentation.

The test materials for photopatch testing or a photopatch test series or both are applied to the back in a duplicate set, one for irradiation and one for a control. After 24 hours, one of the series of patches is removed along with the excess test material. The rest of the body is covered and the site irradiated. The irradiation source is usually a bank of UV-A fluorescent tubes. The UV-A dose applied may vary between 5 J/cm² and 10 J/cm².^{21 - 22} After irradiation, the site is covered and examined 24 hours later. At that time, the control set is also removed and examined. If the precise action spectrum of the test materials is not known, it is advisable to apply the series in triplicate: one set for irradiation with UV-A, one set for irradiation with UV-B, and the other as the control. The UV-B irradiation with 0.75 of the MED can be applied with a bank of broadband UV-B fluorescent tubes or with a filtered solar simulator.

Solar urticaria is a rare condition. The differential diagnosis with PLE, which has about the same localization, is based on the urticarial aspect of the solar urticaria lesions, which appear within 5 to 10 minutes and soon disappear, always within the subsequent 1 to 2 hours. This also allows differentiation from a drug-induced photosensitivity. Phototesting is necessary to confirm the diagnosis: the lesions will be provoked almost immediately or within a few minutes. Therefore, the immediate reading is the most important one, and repeated phototesting is not necessary. The phototesting is done on the upper part of the back of the patient to provoke the urticarial lesions and/or to determine the action spectrum (Figure 5). Not all patients develop lesions during phototesting.²³ The best light source is a xenon lamp in combination with a monochromator, but less expensive light sources can be used, such as a bank of UV-B and UV-A tubes and a slide projector with a water filter (to eliminate induction of heat urticaria) as the visible light source. For each type of irradiation, the minimal urticarial dose is determined with dose increases of 25% to 40%. If a negative result is obtained with a particular light source, other light sources should be tried, and if the results of phototesting remain negative, provocative phototesting should be done with natural sunlight.²³

Lupus erythematosus may appear as a photodermatosis, although in most cases it is a skin disease aggravated by sun exposure. The lesions may sometimes mimic those of PLE. In such cases, the lesions may be present on the bridge of the nose and usually persist much longer than those of PLE, which are nearly always gone within 2 weeks after sun avoidance; however, they may also appear as a patchy scaling erythema. Phototesting can be done on the upper part of the back or on the extensor surface of the forearms with a high-intensity UV-A source or a solar simulator or with a bank of UV-B fluorescent tubes applying, respectively, 100 J/cm² of UV-A and 1.5 MED of UV-B daily on the same area for up to 3 days, in the same way as for PLE. However, readings should be done not only immediately and after 24 hours but also after 72 hours and 1, 2, and 3 weeks. In white patients, a positive reaction will be observed in 60% to 90% of the cases of subacute cutaneous lupus, in 40% of the cases of discoid lupus, and in 10% to 25% of the cases of systemic lupus.^{24 - 26} Apart from phototesting with late readings, lupus can also be differentiated from PLE by serological examination: the anti–Ro/SS-A antibodies are negative for PLE²⁷ and positive in 40% to 100% of the cases of subacute cutaneous lupus (70% to 100% with enzyme-linked immunosorbent assay) and in 25% to 60% of the cases of systemic lupus²⁸ ; they are unusual in chronic discoid lupus. The HLA-DR3 phenotype is present in 50% to 80% of the cases of subacute cutaneous lupus. Another way of differentiating the condition from PLE is by skin immunofluorescence, dustlike particles of IgG and IgM deposition being found in the epidermis, the subepidermal region, and the dermal cellular infiltrates in 32% of the patients with subacute cutaneous lupus²⁹ but not in those with chronic discoid lupus or systemic lupus. On the other hand, IgG, IgM, and IgA deposits are found at the dermoepidermal junction of patients with chronic discoid lupus but not of those with subacute cutaneous lupus. Results of skin immunofluorescence are negative in PLE.

Patients with porphyria cutanea tarda can also have a photodistribution of their lesions, which could be skin fragility, vesicles and blisters, superficial scars, milia, hyperpigmentation, hypertrichosis, premature aging, sclerodermalike changes, and alopecia. Porphyrin determinations will disclose normal levels in the red blood cells and elevated levels of uroporphyrins I to III in the urine and of isocoproporphyrin III in the feces.

When the photodermatosis starts at an older age, the differential diagnosis will mainly be between chronic actinic dermatitis and drug-induced photosensitivity, but dermatomyositis also has to be excluded.

Chronic actinic dermatitis occurs most frequently in elderly men and is characterized by a persistent erythema or eczema of the face, hands, and other exposed areas (Figure 6). Although areas receiving less sun exposure may be spared in the beginning (Figure 7), non–sun-exposed areas may also become involved in time (Figure 6). Phototesting is always necessary to confirm the diagnosis and will show a broad action spectrum with low threshold doses.³⁰ Phototesting can be done with a xenon lamp and a monochromator or with a bank of UV-B and UV-A tubes and a slide projector as the visible light source. For each wavelength, the MED will be determined. Readings are done immediately and after 24 hours. The differential diagnosis with a cutaneous T-cell lymphoma is based on the histological examination whereby CD8⁺ cells will predominate in cases of chronic actinic dermatitis and CD4⁺ cells in cases of cutaneous T-cell lymphoma. In many cases, results of phototesting will be normal in the case of a cutaneous T-cell lymphoma. However, some patients may show abnormal reactions after phototesting.^{31 - 32}

Drug-induced photosensitivity occurs as frequently in women as in men and is characterized by a sunburn type of reaction in cases of phototoxic reaction or a more eczematous pattern in cases of photoallergic reaction. Again, phototesting is necessary to confirm the diagnosis. The phototesting has to be done twice, during the drug intake and 1 month after stopping the drug intake, whereupon an improvement or a normalization of the photosensitivity should be demonstrated. The optimal light source is a xenon lamp in combination with a monochromator, but a xenon lamp in combination with filters or a bank of UV-B and UV-A tubes can also be used.

Dermatomyositis is not a real photodermatosis. However, because it can be aggravated by sun exposure,³³ because the lesions are mainly localized in the face, and because it is mainly a condition of elderly people, dermatomyositis sometimes has to be differentiated from chronic actinic dermatitis: the creatine level in the 24-hour urine collection will be high for dermatomyositis but not for chronic actinic dermatitis.

Accepted for publication January 20, 2000.

Reprints: Rik Roelandts, MD, PhD, Department of Dermatology, Photodermatology Unit, University Hospital, Kapucijnenvoer 33, B-3000 Leuven, Belgium (e-mail: Rik.Roelandts@uz.kuleuven.ac.be).