- Email: [email protected]
Drug-induced photo-onycholysis
Journal of the
American Academy of
Watanabe e~ al
9. 10. 11. 12.
lateral sclerosis. A histochemical study. Neurology (Minneap) 1960;10:717-24. Ross R, Benditt EP. Wound healing and collagen formation. II. Fine structure in experimental scurvy. J Cell Biol 1962;12:533-54. Fleischmajer R, Damiano V, Nedwish A, Scleroderma and subcutaneous tissue. Science 1971 ; 171 : 1019-21. Kobayashi S, Asboe-Hansen G. Ultrastructural changes in the inflammatory zone of localized scleroderma. Acta Derm Venereol (Stockh) 1974;54:105-12. MitchelI RE. Chronic solar dermatosis. A light and elec-
Dermatology
tron microscopic study of the dermis. J Invest Dermatol 1967;48:203-19. 13. Kajikawa K. Connective tissue. Tokyo: Kanehara, 1984: 198. 14. Gunson DE, Halliwell REW, Minor RR. Dermal collagen degradation and phagocytosis. Occurrence in a horse with hyperextensible fragile skin. Arch Dermatol 1984; 120:599-604. 15. Kornberg RL, Hendler SS, Oikarinen AI, Matsuoka LY, Uitto J. Elastoderma--disease of elastin accumulation within the skin. N Engl J Med 1985;312:771-4.
Drug-induced photo-onycholysis Three subtypes identified in a study of 15 cases Robert Baran, M . D . , and Lennart Juhlin, M.D. Cannes, France, and Uppsala, Sweden We have studied 15 patients with photo-onycholysis induced by tetracyclines, psoralens, or fluoroquinolones. Three distinct clinical subtypes of onycholysis were seen. Type I showed a half-moon-shaped separation that was concave distally. Type II had a circular notch opened distally and shaped as if the distal nail plate had acted as a convex lens. In type III the changes were located in the central part of the nail bed with no connection to the margins. Ultraviolet (UV) irradiation of normal fingerriails with various wavelengths showed that 3% to 20% of the irradiation could penetrate the nail. The different patterns of photodamage might be caused by the nail acting as a lens. Less protection by lack of melanin and absence of sebum and stratum granulosum may favor penetration of UV irradiation and explain why the skin was not always affected. (J AM AcAD DERMATOL 1987;17:1012-6.)
Photo-onycholysis is a possible sequela to cutaneous photosensitization that occasionally complicates therapy with tetracycline derivatives, psoralens, and, more recently, fluoroquinolones. ~ It usually appears after more than 2 weeks of exposure to the drug and often follows a photosensitivity reaction in the skin. It m a y be seen initially as part of Segat's triad--photosensitivity followed by discoloration of the nails and onycholysis.2 Un-
From the Dermatology Unit, Cannes General Hospital, Cannes, and Department of Demlatology, University Hospital, Uppsala. Accepted for publication May 27, 1987, No reprints available.
1012
commonly, photo-onycholysis may be seen with cutaneous porphyria-like symptoms but without detectable elevations of porphyrins? 4 A case of onycholysis following doxycycline treatment but without clinical skin manifestation of photosensitivity was reported by Jeanmougin et al? In that study direct immunofluorescence of the clinically normal sun-exposed skin showed porphyrialike fluorescence around papillary vessel walls with IgG and IgM deposition. Tetracycline- and minocycline-induced onycholysis without photosensitivity of the skin has also been described. 6'7 The purpose of this article is to define the different clinical features of drug-induced photoonycholysis and to study its mechanism.
Volume 17 Number 6 December 1987
Drug-induced photo-onycholysis
1013
Fig. 1. Photo-onyeholysis, type I. MATERIAL AND M E T H O D S
Patients Patient 1 was a 10-year-old boy treated daily 10 days with 200 mg of doxycycline for a bronchial infection. The following week he developed a phototoxic reaction with erythema of the face and dorsal part of the hands. Onycholysis was observed a few days later. Patients 2 through 7 were women aged 27 to 55 years. Five had acne rosacea treated with 250 mg of tetracycline two to three times daily for 3 to 8 weeks. One had been treated with doxycycline for a sinusitis condition for 1 week. None of these patients showed phototoxic reactions of the skin. They were exposed to the sun during the period of treatment. Onycholysis was observed after 1 to 6 weeks of treatment. Patients 8 through 13 were women aged 30 to 48 years treated for vitiligo with 10 mg of trioxsalen before sun exposure during a period of 3 to 8 weeks. Onycholysis was noted after 2 to 8 weeks of treatment. Patient 14 was an 82-year-old man with skin infections treated with pefloxacin for 10 to 14 days and then exposed to the sun. After 4 days he developed erythema on the hands followed 2 days later by onycholysis on all fingers. Patient 15 was a 74-year-old man treated with ofloxacin for 10 days; 1 month later he received oral 8-methoxypsoralen and ultraviolet A (UVA) irradiation for 2 weeks. Onycholysis and subungual hemorrha~es were observed on all fingers 2 weeks later.
Nail changes Onycholysis. Three distinct types of onycholysis were noted after both antibiotics and trioxsalen were administered; one common sign was prevalent in all three types: the lateral margins of the nails were unaffected. We did not find any evident relationship between the responsible drugs and the different types of photoonycholysis. Type I. In nine patients several fingers were involved.
Fig. 2. Photo-onycholysis, type II.
The separating part of the nail plate was half-moonshaped and concave distally with a pigmentation of variable intensity and showed a well-demarcated proximal border (Fig. 1). Type II. In three cases on one finger only, a welldefined circular notch was present, which opened distally and had a brownish hue proximally (Fig. 2). Type III. In the central part of the pink nail bed on several fingers of three patients, there was initially a round yellow staining that turned reddish after 5-10 days (Fig. 3). Histology. Biopsy was done in five patients at the first visit. Two each from types I and II, and one from type Ill showed slightly dilated blood vessels but no inflammatory changes. In two patients there was a discrete hyperkeratosis. Staining for iron was negative.
Penetration of nails by UV irradiation The penetration of radiation of varius wavelengths was measured through normal and psoriatic nails removed for various reasons. A monochromatic photoirradiator* emitting wavelengths with narrow bands was used. The nails were placed against the 20 mm 2 aperture of the irradiator and the transmitted radiation measured with a thermopile placed hard against the other side of the nails. The percentage of transmitted radiation was obtained by measuring the radiation detected without nails at the same distance. Readings were made at 313, 340, 360, 400, 440, and 500 nm. The thickness *Applied PhotophysicsLtd., London, England.
1014
Journal of the American Academy of Dermatology
Baran and Juhlin
% 20-
Transmission o---o Normal
O,7mm
© e, Psoriatic 1,0mm
15.
10.
.
1,5 mm !
300 Fig. 3. Photo-onycholysis, type III,
of the nail was measured b y calipers. The results are shown in Fig. 4. It is evident that 3% to 20% of the radiation can pass through normal nails. In psoriatic nails the penetration was less than 4% for the wavelengths tested.
DISCUSSION A classification of onycholysis into three subtypes is justified by differences in appearance. The round or oval spot of the central nail bed of type III shows the link between photo-onycholysis and subungual photohemorrhage. The latter represents an aspect or sequela of photo-onycholysis, depending on the timing. There are several possible explanations why the nails are the focus of predilection for photodamage. 1. The nail acts as a convex lens, which would explain the pattern of the lesion. Such a mechanism has been discussed recently by Logan and Hawk 8 who reported on four cases of spontaneously appearing photo-onycholysis not associated with any drugs. 2. Melanin is sparse in the nail bed and therefore offers little protection except in black or Mongoloid persons. We have not found any reports of drug-induced photo-onycholysis in black patients taking tetracyclines, psoralens, or fluoroquinolones. In white subjects the melanocytes contain
350
•
400
|
|
450
500
nm
Wavelength
Fig. 4. Transmission of radiation (313 to 500 nm) through normal (oo) and opaque psoriatic nails (o--e). Thickness of nails ranges from 0.7 to 1.5 ram.
only a few melanized stage IV melanosomes, which are abundant in other races before irradiation. Because of this difference, melanized, stage IV melanosomes could function as a protection against photo-onycholysis. 3. The absence of sebaceous glands in the subungual area eventually may play a role since lipids on the skin can reduce UV transmission. 9 Scanty, horny cells and absence of stratum granulosum in the nail bed may also favor penetration of UV irradiation. 4. Psoralens act in combination with irradiation of wavelengths between 340 and 360 nm. Tetracycline derivatives cause photosensitivity with shorter wavelengths, but an influence of UVA and visible light has also been demonstrated, io-12Sams and Epstein t3 and Maibach and associates,14 using a plastic filter that blocks radiation below 310 nm, found much greater skin abnormalities than they did with window glass, which allowed passage of radiation above 315 nm. 13"a4 This would suggest that UV irradiation between 310 and 315 nm is critical in modifying the reaction. We have found that 313 nm can penetrate the normal nails. Thus, one could explain the appearance of photo-
Volume 17 Number 6 December 1987
Fig. 5.
Drug-induced photo-onycholysis
Photo-onycholysis,type I.
onycholysis without photosensitivity elsewhere, the thin normal nail being considered as a plastic plate whereas the skin is more like a window glass filter. We have found that irradiation with wavelengths 313 to 500 nm, 3% to 20%, can penetrate a 0.7-mm thick fingernail plate. Similar findings have been reported previously on toenails. ~5With long-term exposure to UV irradiation, the penetrating dose could be sufficient to cause damage. However, in three patients treated with tetracycline (1 gm daily) for 1 to 4 weeks, we were unable to induce photo-onycholysis experimentally by exposing the nails to single doses of 313 nm (2 joules) or 360 nm (10 joules). The phototoxic reaction also depends on the concentration of the drug in the nail bed. Drugs like tetracycline can remain fixed for a considerable time in the skin 4 and can thereby cause a phototoxic reaction. Sartoris et aF 6 observed pboto-onycholysis 1 month after a patient had taken 100 mg of doxycycline daily for 4 days. One of our patients stopped taking doxycycline when photo-onycholysis appeared on the three middle fingernails on both hands. One month later, she had photo-onycholysis on the left fifth finger. The low penetration of UV irradiation in psoriatic nails explains the poor effect of psoralens in such patients. The origin of the discoloration observed in most patients is not clear. It could be caused by extravasation of blood. Thus, splinter hemorrhages have been reported with onycholysis after tetracycline, and hemosiderin deposits should be considered in cases with a brown-yellow discoloration, v Using light and electron microscopy, Zala et al ~7 studied
1015
Fig. 6. Nails of same patient as shown in Fig. 51 Nails have been trimmed showing that the discoloration was caused by keratin dust.
a patient who developed photo-onycholysis following treatment with 8-methoxypsoralen. They observed numerous erythrocytes on the surface of the toenail bed and generation of numerous multinucleated epithelial cells and fibroblasts. Histologic examination in five of our patients showed only slight vasodilation because the biopsy specimens were taken at the first visit and a purpuric reaction might have appeared later, as we have seen in some of our patients. Keratin dust also can cause dyschromia as a result of discrete subungual hyperkeratosis or exogenous material. This was evident in one of our patients in whom such material could be removed at a later visit (Figs. 5 and 6). " Most of our patients became aware of their photo-onycholysis because they felt pain. Pain may also precede onycholysis. Several drugs can cause photo-onycholysis but as yet pain has been described only with tetracycline derivatives and psoralens plus ultraviolet A (PUVA). ~8.,~The cause of the pain is unknown, but terminal vessels, capillaries, or glomus bodies may play an important role. REFERENCES
1. Baran R, Brun P. Photoonycholysis induced by the fluoroquinolones pefloxacine and ofloxaeine. Dermatologica 1986;173:185-8. 2. Segal BN. Photosensitivity, nail discoloration and onycholysis (side effects of tetracycline therapy). Arch Intern Med 1963;112:165-7. 3. Epstein JH, Tuffanelli DL, Seibert JS, Epstein WL. Porphyria-like cutaneous changes induced by tetra-
Journal of the American Academy of Dermatology
Baran and Juhlin
4. 5. 6. 7.
8. 9. 10. 1l.
cycline hydrochloride photosensitization. Arch Dermatol 1976;I 12:661-6. Hawk JLM. Skin changes resembling hepatic cutaneous porphyria induced by oxytetracycline photosensitization. Clin Exp Dermatol 1980;5:321-5. Jeanmougin M, Morel P, Civatte J. Photo-onycholyse induite par la doxycycline. Ann Dermatol Vcnereol 1982;109:165-6. Kestel JL. Phnto-onyeholysis from minocycline. Sideeffects of minocycline therapy. Cutis 1981;28:53-4. Sanders CV, Saenz RE, Lopez M. Splinter hemorrhages and onycholysis: unusual reactions associated with tetracycline hydrochloride therapy. South Med J 1976;69: 1090-2. Logan RH, Hawk JLM. Spontaneous photo-onycholysis, Br J Dermatol 1981;113:605-10. BeadlePC, Burton JL. Absorption ofultraviolet radiation by skin surface lipid. Br J Dermatol 198 [; 104:549-5 l. Orentreich N, Harber LC, Tromovitch TA. Photosensitivity and photo-onycholysis, due to demethylchlortetracycline. Arch Dermatol 1961;83:730-7. Cahn MM, Levy EJ, McMillan JA. Nature and incidence of photosensitivity reactions to demethylchlortetracycline. Arch Dermatol 1961 ;84:485-9.
12. Hasan T, Kochevar IE, McAuliffe DJ, Cooperman BS, Ahdulah D. Mechanism of tetracycline phototoxicity. J Invest Dermatol 1984;83:179-83. 13. Sams WM, Epstein JH. The experimental production of drug phototoxicity in guinea pigs. J Invest Dcrmatol 1967;48:89-94. 14. Maibach HI, Sams WM, Epstein JH. Screening for drug toxicity by wavelengths greater than 3,100 A. Arch Der• matol 1967;95:12-5. 15. Parker SG, Diffey BL. The transmission of optical radiation through human nails. Br J Dermatol 1983;108: 11-6. 16. Sartoris S, Goitre M, Cervetti O. Onicopatia fotocondizionata da tetracicline (pigmentazione delle lamine e onicolisi). Giornale Italiano di Dermatologia, Minerva Dermatol 1976; 111:557-65. 17. Zala L, Omar A, Krebs A. Photo-onycholysis induced by 8-methoxypsoralen. Dermatologica 1977; 154:203-15. 18. Dawber RPR. Photo-onycholysis. Arch Dermatol 1978; 114:1715. 19. Mackie RM. Onycholysis occurring during PUVA therapy. Clin Exp Dermatol 1979;4:111-3.
I
I
I
Bacitracin: A unique topical antibiotic sensitizer B r u c e E. K a t z , M . D . , * and A l e x a n d e r A. Fisher, M . D . * * New York, NY Topical bacitracin appears to be unique for the following reasons: (1) It is becoming a frequent sensitizer, particularly when used after surgery. (2) It m a y cause not only a delayed, eczematous contact dermatitis but an immediate urticarial reaction and, rarely, anaphylactic shock. (3) Positive patch test reactions to bacitracin often do not appear at the usual 48-hour test reading but may be positive when read at 96 hours. (4) Zinc bacitracin and "plain" bacitracin are two forms of topical bacitracin available, with possibly different degrees of sensitizing potential. (5) Bacitracin injected intradermally has been reported to be a histamine-releasing agent. (6) Bacitracin often coreacts but does not cross-react with neomycin. In this presentation we document the clinical and patch test findings of nine patients seen in the past year with allergic contact dermatitis to bacitracin ointment. (J AM ACAD DERMATOL 1987; 17:1016-24.)
From the Department of Dermatology, College of Physicians and Surgeons, Columbia University,* and the Departmentof Dermatology, New York University Post-Graduate Medical School•** Accepted for publication May 13, 1987. Reprint requests to: Dr. Bruce E. Katz, 14 E. 82nd St., New York, NY 10028.
1016
DERIVATION OF T H E N A M E BACITRACIN Bacitracin is an antibiotic p r o d u c e d by the Trac e y I strain of Bacillus subtilis, discovered in 1943 in the L a b o r a t o r y o f Bacteriological R e s e a r c h of the D e p a r t m e n t o f Surgery, College o f Physicians
American Academy of
Watanabe e~ al
9. 10. 11. 12.
lateral sclerosis. A histochemical study. Neurology (Minneap) 1960;10:717-24. Ross R, Benditt EP. Wound healing and collagen formation. II. Fine structure in experimental scurvy. J Cell Biol 1962;12:533-54. Fleischmajer R, Damiano V, Nedwish A, Scleroderma and subcutaneous tissue. Science 1971 ; 171 : 1019-21. Kobayashi S, Asboe-Hansen G. Ultrastructural changes in the inflammatory zone of localized scleroderma. Acta Derm Venereol (Stockh) 1974;54:105-12. MitchelI RE. Chronic solar dermatosis. A light and elec-
Dermatology
tron microscopic study of the dermis. J Invest Dermatol 1967;48:203-19. 13. Kajikawa K. Connective tissue. Tokyo: Kanehara, 1984: 198. 14. Gunson DE, Halliwell REW, Minor RR. Dermal collagen degradation and phagocytosis. Occurrence in a horse with hyperextensible fragile skin. Arch Dermatol 1984; 120:599-604. 15. Kornberg RL, Hendler SS, Oikarinen AI, Matsuoka LY, Uitto J. Elastoderma--disease of elastin accumulation within the skin. N Engl J Med 1985;312:771-4.
Drug-induced photo-onycholysis Three subtypes identified in a study of 15 cases Robert Baran, M . D . , and Lennart Juhlin, M.D. Cannes, France, and Uppsala, Sweden We have studied 15 patients with photo-onycholysis induced by tetracyclines, psoralens, or fluoroquinolones. Three distinct clinical subtypes of onycholysis were seen. Type I showed a half-moon-shaped separation that was concave distally. Type II had a circular notch opened distally and shaped as if the distal nail plate had acted as a convex lens. In type III the changes were located in the central part of the nail bed with no connection to the margins. Ultraviolet (UV) irradiation of normal fingerriails with various wavelengths showed that 3% to 20% of the irradiation could penetrate the nail. The different patterns of photodamage might be caused by the nail acting as a lens. Less protection by lack of melanin and absence of sebum and stratum granulosum may favor penetration of UV irradiation and explain why the skin was not always affected. (J AM AcAD DERMATOL 1987;17:1012-6.)
Photo-onycholysis is a possible sequela to cutaneous photosensitization that occasionally complicates therapy with tetracycline derivatives, psoralens, and, more recently, fluoroquinolones. ~ It usually appears after more than 2 weeks of exposure to the drug and often follows a photosensitivity reaction in the skin. It m a y be seen initially as part of Segat's triad--photosensitivity followed by discoloration of the nails and onycholysis.2 Un-
From the Dermatology Unit, Cannes General Hospital, Cannes, and Department of Demlatology, University Hospital, Uppsala. Accepted for publication May 27, 1987, No reprints available.
1012
commonly, photo-onycholysis may be seen with cutaneous porphyria-like symptoms but without detectable elevations of porphyrins? 4 A case of onycholysis following doxycycline treatment but without clinical skin manifestation of photosensitivity was reported by Jeanmougin et al? In that study direct immunofluorescence of the clinically normal sun-exposed skin showed porphyrialike fluorescence around papillary vessel walls with IgG and IgM deposition. Tetracycline- and minocycline-induced onycholysis without photosensitivity of the skin has also been described. 6'7 The purpose of this article is to define the different clinical features of drug-induced photoonycholysis and to study its mechanism.
Volume 17 Number 6 December 1987
Drug-induced photo-onycholysis
1013
Fig. 1. Photo-onyeholysis, type I. MATERIAL AND M E T H O D S
Patients Patient 1 was a 10-year-old boy treated daily 10 days with 200 mg of doxycycline for a bronchial infection. The following week he developed a phototoxic reaction with erythema of the face and dorsal part of the hands. Onycholysis was observed a few days later. Patients 2 through 7 were women aged 27 to 55 years. Five had acne rosacea treated with 250 mg of tetracycline two to three times daily for 3 to 8 weeks. One had been treated with doxycycline for a sinusitis condition for 1 week. None of these patients showed phototoxic reactions of the skin. They were exposed to the sun during the period of treatment. Onycholysis was observed after 1 to 6 weeks of treatment. Patients 8 through 13 were women aged 30 to 48 years treated for vitiligo with 10 mg of trioxsalen before sun exposure during a period of 3 to 8 weeks. Onycholysis was noted after 2 to 8 weeks of treatment. Patient 14 was an 82-year-old man with skin infections treated with pefloxacin for 10 to 14 days and then exposed to the sun. After 4 days he developed erythema on the hands followed 2 days later by onycholysis on all fingers. Patient 15 was a 74-year-old man treated with ofloxacin for 10 days; 1 month later he received oral 8-methoxypsoralen and ultraviolet A (UVA) irradiation for 2 weeks. Onycholysis and subungual hemorrha~es were observed on all fingers 2 weeks later.
Nail changes Onycholysis. Three distinct types of onycholysis were noted after both antibiotics and trioxsalen were administered; one common sign was prevalent in all three types: the lateral margins of the nails were unaffected. We did not find any evident relationship between the responsible drugs and the different types of photoonycholysis. Type I. In nine patients several fingers were involved.
Fig. 2. Photo-onycholysis, type II.
The separating part of the nail plate was half-moonshaped and concave distally with a pigmentation of variable intensity and showed a well-demarcated proximal border (Fig. 1). Type II. In three cases on one finger only, a welldefined circular notch was present, which opened distally and had a brownish hue proximally (Fig. 2). Type III. In the central part of the pink nail bed on several fingers of three patients, there was initially a round yellow staining that turned reddish after 5-10 days (Fig. 3). Histology. Biopsy was done in five patients at the first visit. Two each from types I and II, and one from type Ill showed slightly dilated blood vessels but no inflammatory changes. In two patients there was a discrete hyperkeratosis. Staining for iron was negative.
Penetration of nails by UV irradiation The penetration of radiation of varius wavelengths was measured through normal and psoriatic nails removed for various reasons. A monochromatic photoirradiator* emitting wavelengths with narrow bands was used. The nails were placed against the 20 mm 2 aperture of the irradiator and the transmitted radiation measured with a thermopile placed hard against the other side of the nails. The percentage of transmitted radiation was obtained by measuring the radiation detected without nails at the same distance. Readings were made at 313, 340, 360, 400, 440, and 500 nm. The thickness *Applied PhotophysicsLtd., London, England.
1014
Journal of the American Academy of Dermatology
Baran and Juhlin
% 20-
Transmission o---o Normal
O,7mm
© e, Psoriatic 1,0mm
15.
10.
.
1,5 mm !
300 Fig. 3. Photo-onycholysis, type III,
of the nail was measured b y calipers. The results are shown in Fig. 4. It is evident that 3% to 20% of the radiation can pass through normal nails. In psoriatic nails the penetration was less than 4% for the wavelengths tested.
DISCUSSION A classification of onycholysis into three subtypes is justified by differences in appearance. The round or oval spot of the central nail bed of type III shows the link between photo-onycholysis and subungual photohemorrhage. The latter represents an aspect or sequela of photo-onycholysis, depending on the timing. There are several possible explanations why the nails are the focus of predilection for photodamage. 1. The nail acts as a convex lens, which would explain the pattern of the lesion. Such a mechanism has been discussed recently by Logan and Hawk 8 who reported on four cases of spontaneously appearing photo-onycholysis not associated with any drugs. 2. Melanin is sparse in the nail bed and therefore offers little protection except in black or Mongoloid persons. We have not found any reports of drug-induced photo-onycholysis in black patients taking tetracyclines, psoralens, or fluoroquinolones. In white subjects the melanocytes contain
350
•
400
|
|
450
500
nm
Wavelength
Fig. 4. Transmission of radiation (313 to 500 nm) through normal (oo) and opaque psoriatic nails (o--e). Thickness of nails ranges from 0.7 to 1.5 ram.
only a few melanized stage IV melanosomes, which are abundant in other races before irradiation. Because of this difference, melanized, stage IV melanosomes could function as a protection against photo-onycholysis. 3. The absence of sebaceous glands in the subungual area eventually may play a role since lipids on the skin can reduce UV transmission. 9 Scanty, horny cells and absence of stratum granulosum in the nail bed may also favor penetration of UV irradiation. 4. Psoralens act in combination with irradiation of wavelengths between 340 and 360 nm. Tetracycline derivatives cause photosensitivity with shorter wavelengths, but an influence of UVA and visible light has also been demonstrated, io-12Sams and Epstein t3 and Maibach and associates,14 using a plastic filter that blocks radiation below 310 nm, found much greater skin abnormalities than they did with window glass, which allowed passage of radiation above 315 nm. 13"a4 This would suggest that UV irradiation between 310 and 315 nm is critical in modifying the reaction. We have found that 313 nm can penetrate the normal nails. Thus, one could explain the appearance of photo-
Volume 17 Number 6 December 1987
Fig. 5.
Drug-induced photo-onycholysis
Photo-onycholysis,type I.
onycholysis without photosensitivity elsewhere, the thin normal nail being considered as a plastic plate whereas the skin is more like a window glass filter. We have found that irradiation with wavelengths 313 to 500 nm, 3% to 20%, can penetrate a 0.7-mm thick fingernail plate. Similar findings have been reported previously on toenails. ~5With long-term exposure to UV irradiation, the penetrating dose could be sufficient to cause damage. However, in three patients treated with tetracycline (1 gm daily) for 1 to 4 weeks, we were unable to induce photo-onycholysis experimentally by exposing the nails to single doses of 313 nm (2 joules) or 360 nm (10 joules). The phototoxic reaction also depends on the concentration of the drug in the nail bed. Drugs like tetracycline can remain fixed for a considerable time in the skin 4 and can thereby cause a phototoxic reaction. Sartoris et aF 6 observed pboto-onycholysis 1 month after a patient had taken 100 mg of doxycycline daily for 4 days. One of our patients stopped taking doxycycline when photo-onycholysis appeared on the three middle fingernails on both hands. One month later, she had photo-onycholysis on the left fifth finger. The low penetration of UV irradiation in psoriatic nails explains the poor effect of psoralens in such patients. The origin of the discoloration observed in most patients is not clear. It could be caused by extravasation of blood. Thus, splinter hemorrhages have been reported with onycholysis after tetracycline, and hemosiderin deposits should be considered in cases with a brown-yellow discoloration, v Using light and electron microscopy, Zala et al ~7 studied
1015
Fig. 6. Nails of same patient as shown in Fig. 51 Nails have been trimmed showing that the discoloration was caused by keratin dust.
a patient who developed photo-onycholysis following treatment with 8-methoxypsoralen. They observed numerous erythrocytes on the surface of the toenail bed and generation of numerous multinucleated epithelial cells and fibroblasts. Histologic examination in five of our patients showed only slight vasodilation because the biopsy specimens were taken at the first visit and a purpuric reaction might have appeared later, as we have seen in some of our patients. Keratin dust also can cause dyschromia as a result of discrete subungual hyperkeratosis or exogenous material. This was evident in one of our patients in whom such material could be removed at a later visit (Figs. 5 and 6). " Most of our patients became aware of their photo-onycholysis because they felt pain. Pain may also precede onycholysis. Several drugs can cause photo-onycholysis but as yet pain has been described only with tetracycline derivatives and psoralens plus ultraviolet A (PUVA). ~8.,~The cause of the pain is unknown, but terminal vessels, capillaries, or glomus bodies may play an important role. REFERENCES
1. Baran R, Brun P. Photoonycholysis induced by the fluoroquinolones pefloxacine and ofloxaeine. Dermatologica 1986;173:185-8. 2. Segal BN. Photosensitivity, nail discoloration and onycholysis (side effects of tetracycline therapy). Arch Intern Med 1963;112:165-7. 3. Epstein JH, Tuffanelli DL, Seibert JS, Epstein WL. Porphyria-like cutaneous changes induced by tetra-
Journal of the American Academy of Dermatology
Baran and Juhlin
4. 5. 6. 7.
8. 9. 10. 1l.
cycline hydrochloride photosensitization. Arch Dermatol 1976;I 12:661-6. Hawk JLM. Skin changes resembling hepatic cutaneous porphyria induced by oxytetracycline photosensitization. Clin Exp Dermatol 1980;5:321-5. Jeanmougin M, Morel P, Civatte J. Photo-onycholyse induite par la doxycycline. Ann Dermatol Vcnereol 1982;109:165-6. Kestel JL. Phnto-onyeholysis from minocycline. Sideeffects of minocycline therapy. Cutis 1981;28:53-4. Sanders CV, Saenz RE, Lopez M. Splinter hemorrhages and onycholysis: unusual reactions associated with tetracycline hydrochloride therapy. South Med J 1976;69: 1090-2. Logan RH, Hawk JLM. Spontaneous photo-onycholysis, Br J Dermatol 1981;113:605-10. BeadlePC, Burton JL. Absorption ofultraviolet radiation by skin surface lipid. Br J Dermatol 198 [; 104:549-5 l. Orentreich N, Harber LC, Tromovitch TA. Photosensitivity and photo-onycholysis, due to demethylchlortetracycline. Arch Dermatol 1961;83:730-7. Cahn MM, Levy EJ, McMillan JA. Nature and incidence of photosensitivity reactions to demethylchlortetracycline. Arch Dermatol 1961 ;84:485-9.
12. Hasan T, Kochevar IE, McAuliffe DJ, Cooperman BS, Ahdulah D. Mechanism of tetracycline phototoxicity. J Invest Dermatol 1984;83:179-83. 13. Sams WM, Epstein JH. The experimental production of drug phototoxicity in guinea pigs. J Invest Dcrmatol 1967;48:89-94. 14. Maibach HI, Sams WM, Epstein JH. Screening for drug toxicity by wavelengths greater than 3,100 A. Arch Der• matol 1967;95:12-5. 15. Parker SG, Diffey BL. The transmission of optical radiation through human nails. Br J Dermatol 1983;108: 11-6. 16. Sartoris S, Goitre M, Cervetti O. Onicopatia fotocondizionata da tetracicline (pigmentazione delle lamine e onicolisi). Giornale Italiano di Dermatologia, Minerva Dermatol 1976; 111:557-65. 17. Zala L, Omar A, Krebs A. Photo-onycholysis induced by 8-methoxypsoralen. Dermatologica 1977; 154:203-15. 18. Dawber RPR. Photo-onycholysis. Arch Dermatol 1978; 114:1715. 19. Mackie RM. Onycholysis occurring during PUVA therapy. Clin Exp Dermatol 1979;4:111-3.
I
I
I
Bacitracin: A unique topical antibiotic sensitizer B r u c e E. K a t z , M . D . , * and A l e x a n d e r A. Fisher, M . D . * * New York, NY Topical bacitracin appears to be unique for the following reasons: (1) It is becoming a frequent sensitizer, particularly when used after surgery. (2) It m a y cause not only a delayed, eczematous contact dermatitis but an immediate urticarial reaction and, rarely, anaphylactic shock. (3) Positive patch test reactions to bacitracin often do not appear at the usual 48-hour test reading but may be positive when read at 96 hours. (4) Zinc bacitracin and "plain" bacitracin are two forms of topical bacitracin available, with possibly different degrees of sensitizing potential. (5) Bacitracin injected intradermally has been reported to be a histamine-releasing agent. (6) Bacitracin often coreacts but does not cross-react with neomycin. In this presentation we document the clinical and patch test findings of nine patients seen in the past year with allergic contact dermatitis to bacitracin ointment. (J AM ACAD DERMATOL 1987; 17:1016-24.)
From the Department of Dermatology, College of Physicians and Surgeons, Columbia University,* and the Departmentof Dermatology, New York University Post-Graduate Medical School•** Accepted for publication May 13, 1987. Reprint requests to: Dr. Bruce E. Katz, 14 E. 82nd St., New York, NY 10028.
1016
DERIVATION OF T H E N A M E BACITRACIN Bacitracin is an antibiotic p r o d u c e d by the Trac e y I strain of Bacillus subtilis, discovered in 1943 in the L a b o r a t o r y o f Bacteriological R e s e a r c h of the D e p a r t m e n t o f Surgery, College o f Physicians