|
|
| |||||||||||||||||||||||||||||||||||||||
Background and Methods Irregular disfiguring skin hyperpigmentation due to inflammation may develop in black persons. We investigated the treatment of this hyperpigmentation with topical tretinoin (0.1 percent retinoic acid cream). Fifty-four subjects completed a 40-week randomized, double-blind, vehicle-controlled study. Twenty-four subjects applied tretinoin daily to the face, arms, or both areas, and 30 subjects applied vehicle cream. At base line and after 40 weeks of treatment, each subject's postinflammatory hyperpigmented lesions and normal skin were assessed by clinical and colorimetric evaluations and by analysis of biopsy specimens.
Results The facial postinflammatory hyperpigmented lesions of the tretinoin-treated subjects were significantly lighter after the 40 weeks of therapy than those of the vehicle-treated subjects (P<0.001); overall improvement was first noted after 4 weeks of tretinoin treatment. At the end of treatment, colorimetry demonstrated a 40 percent lightening of the lesions toward normal skin color in the tretinoin-treated lesions, as compared with an 18 percent lightening in vehicle-treated lesions (P = 0.05). The epidermal melanin content in the lesions decreased by 23 percent with tretinoin and by 3 percent with vehicle (P = 0.24). Normal skin was minimally lightened by tretinoin as compared with vehicle, according to both clinical evaluation (0.1 vs. -0.1 unit change on an 8-point scale; P = 0.055) and colorimetry (P<0.001). Retinoid dermatitis developed in 12 of the 24 tretinoin-treated subjects who completed the study (50 percent) and in 1 tretinoin-treated subject who withdrew from the study, but diminished as the study progressed.
Conclusions Topical application of tretinoin significantly lightens postinflammatory hyperpigmentation; to a clinically minimal but statistically significant degree, it also lightens normal skin in black persons.
Recently, we and others noted that treatment with topical tretinoin (retinoic acid) lightened hyperpigmented spots in white subjects7,8,9,10,11,12. Because the deposition of melanin may be similar in hyperpigmented spots in both white and black persons, we studied the ability of tretinoin to lighten postinflammatory hyperpigmentation in black subjects. Tretinoin is thought to cause hyperpigmentation and a poorly tolerated dermatitis when applied to black skin. However, we found that topical tretinoin lightened postinflammatory hyperpigmented lesions in black subjects. An unexpected finding was a clinically minimal but statistically significant lightening of these subjects' normal skin.
Methods
Patients and Design
We enrolled 68 healthy black subjects (21 men and 47 women) with moderate to severe lesions of postinflammatory hyperpigmentation on their faces, arms, or both areas. The sample size was chosen to provide a statistical power of approximately 0.90 in detecting a difference in overall improvement of at least 0.5 unit on a 5-point scale (see "Clinical Evaluation") between study groups, with a type I error rate of 0.05 and a two-tailed hypothesis. A computer-generated randomization code was used to assign 33 subjects to tretinoin treatment (9 men and 24 women, 21 to 58 years old; mean age, 35) and 35 subjects to treatment with vehicle (12 men and 23 women, 20 to 55 years old; mean age, 34). No subject had used topical medications (including a hydroquinone or a corticosteroid) for at least two weeks, systemic corticosteroids for one month, topical retinoids for six months, or systemic retinoids for one year before the study.
All subjects gave written informed consent after the potential side effects of the agents had been explained to them. The protocol was approved by the institutional review board of the University of Michigan Medical Center.
Treatment
Topical tretinoin (a 0.1 percent retinoic acid cream, Retin-A) and a vehicle cream were supplied by the manufacturer of the tretinoin (Ortho Pharmaceutical, Raritan, N.J.). Both agents were in identical tubes; neither the investigators nor the subjects knew the treatment group to which any subject was assigned. The subjects applied their medication once nightly for 40 weeks to the entire face, arms, or both areas, depending on the sites of the lesions. The subjects were instructed to apply a pea-sized amount and to increase the amount of the cream gradually unless undue irritation developed.
All subjects were provided with sunscreen with a sun-protection factor of 15 that was to be worn during the day, and they were told to avoid excessive exposure to sunlight, wind, and cold. Cosmetics were to be used minimally, and not at all during evaluation or photographing of the skin.
Clinical Evaluation
Overall evaluations of the treated areas were performed at base line, after 2 and 4 weeks of treatment, and then every 4 weeks for 40 weeks. The clinical response (skin color) of these lesions was graded by the same investigator at virtually all visits, according to the following scale: -2, much darker; -1, darker; 0, unchanged; 1, lighter; and 2, much lighter. Because only seven subjects had postinflammatory hyperpigmented lesions on their arms, the overall scores for all subjects' arms were not analyzed.
A separate evaluation of normal areas of skin and four individual lesions of postinflammatory hyperpigmentation (selected at base line for the ease with which they could be accurately located) was performed in each subject. At each follow-up visit, changes in the color of the normal skin and each of the designated lesions as compared with the color at base line were graded according to the following scale: -3, much darker; -2, darker; -1, slightly darker; 0, unchanged; 1, slightly lighter; 2, lighter; 3, much lighter; and 4, absent (undetectable).
The side effects of erythema and desquamation were rated at all visits on a scale of 0 (side effects absent) to 4 (side effects most severe); we classified subjects as having a retinoid reaction if they had a score of 2 or more for erythema or desquamation at two or more visits.
Colorimetry
The designated lesions of postinflammatory hyperpigmentation and normal skin were assessed with a colorimeter (Chroma Meter CR-200, Minolta Camera, Osaka, Japan) before treatment and after 12, 24, and 40 weeks of therapy. Colorimetric assessment produces measures of reflectance; we used the L* value of the L*a*b * color system defined by the Commission Internationale de L'Eclairage13. Although the L* value has a theoretical range of 0 (pure black) to 100 (pure white), its range in human skin is much narrower. Previous studies of sun-protected skin in white subjects demonstrated that their L* values ranged from 62 to 7114. Colorimetry was used to obtain an objective, quantifiable measure of tretinoin-induced lightening of lesions of postinflammatory hyperpigmentation and also to determine whether the colorimetric findings correlated with the clinical assessments of each subject.
During measurements the colorimeter was placed perpendicularly on the skin; blanching was minimized by applying little pressure. The value used was the average of five measurements of each site at each assessment.
Light Microscopy
A 2-mm punch-biopsy specimen of normal facial skin and another of a lesion on the face or an arm were obtained at base line and at the end of the study. When feasible, a lesion large enough to provide both biopsy specimens was selected so that the post-treatment specimen did not include scar tissue due to the base-line biopsy; when this was not possible, the second specimen was taken from a nearby lesion. The second biopsy specimen of normal skin was taken up to 2 cm away from the first. Post-treatment specimens from two subjects given the vehicle were not available because of laboratory errors.
The biopsy specimens were fixed in 10 percent neutral-buffered formalin and stained with hematoxylin and eosin. All histologic sections were examined by the same dermatopathologist, who was not involved in the clinical part of the study and was unaware of the clinical results, treatment assignments, and timing of the biopsy.
Photography
Color photographs for documentation of changes in the hyperpigmented lesions were taken by a professional photographer. Standardized positioning and broad, diffuse illumination with fixed flash units were used. A standard gray card (18 percent reflectance [Kodak, Eastman Kodak, Rochester, N.Y.]) was included in the first frame of each subject's photographic session, to detect extrinsic variation. All film was of the same emulsion lot and was processed in the same laboratory.
Statistical Analysis
The changes in the two study groups in clinical, histologic, and colorimetric values for the average of the four designated lesions and for normal skin from base line to week 40 were compared by a two-sample t-test. The effects of treatment on the overall response at week 40 were analyzed with the chi-square test. The strength of the relation between color changes observed clinically and changes detected colorimetrically was assessed by calculating Pearson's product-moment correlation coefficient.
The results were analyzed with the Michigan Interactive Data Analysis System (MIDAS, a statistical software package developed by the Center for Statistical Consultation and Research at the University of Michigan). All P values are two-sided. Summary statistics are expressed as means ±SE.
Results
The clinical diagnosis of the cause of postinflammatory hyperpigmentation was acne in 62 subjects, shaving irritation in 10, eczema in 5, ingrown facial hairs in 4, and folliculitis in 3; in 15 subjects the hyperpigmentation had more than one cause (two causes in 14 subjects and three causes in 1).
Fourteen subjects did not complete the study. Eleven were withdrawn because they did not attend scheduled visits (six in the tretinoin group and five in the vehicle group). Three subjects in the tretinoin group withdrew from the study, two because of an exacerbation of eczema and one because of a retinoid reaction. Therefore, results are presented for 54 subjects (24 in the tretinoin group and 30 in the vehicle group).
Clinical Findings
Overall Changes
Overall evaluation after 40 weeks of treatment demonstrated significant improvement in the clinical status of the tretinoin group as compared with the vehicle group (P<0.001) (Figure 1 and Figure 2). The facial lesions of 14 of the 24 tretinoin-treated subjects (58 percent) were classified as being much lighter, those of 8 (33 percent) as being lighter, and those of 2 (8 percent) as being unchanged as compared with their appearance at base line. In contrast, the facial lesions of 2 of the 30 vehicle-treated subjects (7 percent) were classified as much lighter, those of 15 (50 percent) were considered lighter and those of 13 (43 percent) were considered unchanged (Figure 1). No subject in either group had lesions that were evaluated as being worse or much worse. Overall lightening of the lesions was noted in the tretinoin group as early as four weeks after treatment was begun (P = 0.009) (Figure 3), and this lightening persisted throughout the study.
|
|
|
The clinical lightening of the designated postinflammatory hyperpigmented lesions after 40 weeks of treatment was significantly greater in the tretinoin group than in the vehicle group (clinical scores, 2.6 ±0.2 vs. 1.9 ±0.2; P = 0.03). Colorimetry demonstrated an increase of 2.7 ±0.7 units in the L* value (lightening) for lesions treated with tretinoin, as compared with an increase of 1.2 ±0.4 units for vehicle-treated lesions (P = 0.05) (Figure 4). The average base-line L* value for all lesions was 40.6 ±0.6 units, and the corresponding average for normal skin was 47.3 ±0.6 units -- a difference of approximately 7 units. Therefore, after 40 weeks of treatment with tretinoin, the L* value was 40 percent closer to the value for normal skin, as compared with 18 percent closer for vehicle-treated lesions. The change in the color of the lesions detected by colorimetry at week 40 was correlated significantly with the color change noted on clinical evaluation (r = 0.43, P = 0.001).
|
The difference between the tretinoin and vehicle groups in lightening of normal skin was barely discernible on clinical evaluation (clinical scores, 0.1 ±0.1 vs. -0.1 ±0.1; P = 0.055). Lightening was also detected by colorimetry, with an average lightening from base line of 1.6 ±0.7 units in the tretinoin group, as compared with a darkening of 1.9 ±0.4 units in the vehicle group (P<0.001) (Figure 4). Because there is no defined end point for lightening of normal skin, it was not possible to quantify the change as a percentage of lightening as in the evaluation of postinflammatory hyperpigmented lesions.
Side Effects
The adverse reactions to tretinoin among the subjects completing the study were limited to retinoid reactions in 12 of the 24 subjects (50 percent); 7 of these 12 subjects had moderate erythema and desquamation, and 5 had moderately severe reactions. The retinoid reactions occurred only in areas where medication had been applied. There was a gradual diminution in the severity, frequency, and duration of these reactions as the study progressed. The cutaneous reactions were well tolerated by all subjects except the three who withdrew from the study. The reactions responded to the application of emollients, the temporary reduction of the amount of medication, or discontinuation of the agent for one to three days. No subject in the tretinoin group had any hyperpigmentation or gross depigmentation as a side effect, and none in the vehicle group had retinoid reactions.
Histologic Findings
In all subjects at base line, the average score for epidermal melanin in the postinflammatory hyperpigmented lesions was 2.2 ±0.1, as compared with 1.7 ±0.1 for normal skin (P<0.002) (Table 1). In more than 95 percent of the subjects the epidermal melanin was located in the basal layer in both lesional and normal skin. There was no significant difference between the hyperpigmented lesions and normal skin in their dermal melanin content, most of which was intracellular. Thus, it appears that the postinflammatory hyperpigmentation resulted from epidermal melanin deposition.
|
Treatment for 40 weeks with topical tretinoin resulted in significant increases in the degree of compaction of the stratum corneum, the thickness of the granular-cell layer, and spongiosis (widening of the spaces between keratinocytes) in the lesions, as compared with treatment with vehicle (Table 1). The epidermal melanin content decreased 23 percent in the tretinoin group, as compared with 3 percent in the vehicle group (P = 0.24). The dermal melanin content increased 49 percent in the tretinoin group but decreased 7 percent in the vehicle group (P = 0.14).
Normal Skin
Topical tretinoin treatment resulted in significant increases in stratum corneum compaction, granular-cell-layer thickness, epidermal thickness, mitotic figures, and spongiosis in normal skin, as compared with treatment with vehicle (Table 1). After 40 weeks, epidermal melanin content decreased 11 percent in the tretinoin group but increased 22 percent in the vehicle group (P = 0.07). Dermal melanin content increased 40 percent in the tretinoin group but decreased 1 percent in the vehicle group (P = 0.06).
For both postinflammatory hyperpigmented lesions and normal skin, there was no correlation between the decrease in epidermal melanin content and the increase in dermal melanin content.
Discussion
Our results demonstrate the tolerability and efficacy of tretinoin (0.1 percent retinoic acid cream) in the treatment of postinflammatory hyperpigmentation (resulting mainly from acne) in black persons. Contrary to previous reports,3 the application of tretinoin did not result in excessive hyperpigmentation or depigmentation of the normal skin in any subject. Furthermore, side effects occurred less frequently in the black subjects who completed this study (50 percent) than in white subjects in another trial (82 percent)10.
Significant clinical improvement occurred earlier in the tretinoin group than in the vehicle group (4 vs. 24 weeks) (Figure 3). Indeed, by the fourth week of therapy the hyperpigmented lesions were rated as lighter in 23 percent of the subjects given tretinoin, as compared with only 3 percent of those given vehicle. However, the eventual partial lightening of the lesions of subjects given the vehicle (Figure 3) suggested that such lesions can undergo spontaneous improvement. Thus, although the vehicle increased lightening by 1.9 units (i.e., two thirds of the increase produced by tretinoin -- probably because of the use of sunscreens, avoidance of sunlight, and the natural course of the disorder), tretinoin clearly speeded resolution. In six subjects who had received tretinoin and who were followed without treatment for six months after medication was stopped, no designated postinflammatory hyperpigmented lesion that had cleared reappeared, although new lesions developed in three of these subjects.
The histologic conditions usually induced by tretinoin, such as epidermal hyperplasia and compaction of the stratum corneum, occurred in our subjects (Table 1)7,8,9,10,11,12. Epidermal melanin content decreased by 23 percent in lesions treated for 40 weeks with tretinoin, although this decrease was not significantly different from the 3 percent reduction during vehicle treatment, which was presumably due to spontaneous resolution. Tretinoin also decreased the epidermal melanin content of normal skin more than the vehicle did (an 11 percent decrease vs. a 22 percent increase; P = 0.07).
A source of potential bias in this study was the retinoid reactions that occurred in some subjects. Because these reactions occurred in only half the subjects given tretinoin, the possibility of such bias was lower than in most studies of this agent. Furthermore, clinical improvement and side effects were graded in reference to base line throughout the study, without reference to findings on previous visits, thus reducing the chance of bias. Moreover, the results of the colorimetric and histologic measurements corroborated the clinical observation of lightening of both postinflammatory hyperpigmented lesions and normal skin.
Although the mechanism of action of tretinoin is unknown, Orlow et al.15 have reported that it inhibits induced melanogenesis. Indeed, the rate of melanogenesis is high in the normal skin of black persons and thus could be inhibited by tretinoin16. In addition, tretinoin may cause epidermal melanin to be redistributed or dispersed, and thus induce clinical lightening. Such a mechanism might be responsible in part for the relatively greater clinical changes after 40 weeks of tretinoin therapy (e.g., a mean colorimetric value that was 40 percent closer to the values for normal skin color), as compared with the change in microscopically detectable melanin content (a 23 percent reduction).
In conclusion, topical tretinoin is an effective treatment for black patients with postinflammatory skin hyperpigmentation, and the severity of its side effects is similar to or less than that in white patients7,8,9,10,11,12. Physicians are reluctant to use topical tretinoin in the treatment of postinflammatory hyperpigmentation in black patients because of reports of unwanted hyperpigmentation, depigmentation, and drug intolerance2,3. In the present study no subject had either hyperpigmentation or excessive depigmentation of normal skin. Tretinoin therapy did lighten normal skin slightly, and it lightened hyperpigmented lesions toward background skin color. The mechanisms by which tretinoin produces clinical improvement in postinflammatory hyperpigmentation in the skin of black persons are not known.
Supported by a grant from the R.W. Johnson Pharmaceutical Research Institute (Raritan, N.J.), which had no part in the design or conduct of the study or in the analysis, interpretation, or reporting of the results, and by the Babcock Dermatologic Endowment (Ann Arbor, Mich.).
We are indebted to Mr. Harrold Carter and Mr. Randal Stegmeyer for the development of the photographic techniques used in this study.
Source Information
From the Dermatopharmacology Unit, Department of Dermatology, University of Michigan Medical Center, Ann Arbor. During part of this study Dr. Ellis and Dr. Voorhees served as consultants to the Johnson & Johnson Corporation (of which the Ortho Pharmaceutical Corporation is a subsidiary).
Address reprint requests to Dr. Voorhees at the Department of Dermatology, University of Michigan Medical Center, 1910 Taubman Ctr., 1500 E. Medical Ctr. Dr., Ann Arbor, MI 48109-0314.
References
| |||||||||||||||||||||||||||||||||||||||
Related Letters:
Visceral Leishmaniasis in Desert Storm Veterans
Boyer M. H., Magill A. J., Grogl M., Gasser R. A., Oster C. N.
Extract |
Full Text
N Engl J Med 1993;
329:1503-1504, Nov 11, 1993.
Correspondence
This article has been cited by other articles:
HOME | SUBSCRIBE | SEARCH | CURRENT ISSUE | PAST ISSUES | COLLECTIONS | PRIVACY | HELP | beta.nejm.org Comments and questions? Please contact us. The New England Journal of Medicine is owned, published, and copyrighted © 2008 Massachusetts Medical Society. All rights reserved. |
Previous
