Correspondence

Cutaneous Disease and Drug Reactions in HIV Infection

N Engl J Med 1993; 329:1582-1583November 18, 1993

Article

To the Editor:

Coopman et al. (June 10 issue)1 found that cutaneous diseases, including drug reactions, are extremely common in patients with human immunodeficiency virus (HIV) infection, and that their incidence increases as immune function deteriorates. They discussed ascertainment bias as a possible contributing factor to these associations, with cutaneous abnormalities possibly precipitating the discovery of HIV infection in some people. They did not mention the possibility of ascertainment bias in the reverse direction -- i.e., the presence of HIV infection may have increased the likelihood that cutaneous disorders would come to medical attention. People with HIV infection see physicians more frequently as their disease progresses. This increased contact would predictably lead to more medical-record entries related to skin problems per patient per unit of time among HIV-infected patients than among patients without HIV infection, even in the absence of true differences between these groups in the prevalence of skin disorders. Furthermore, the greater importance attached to skin disorders by patients and physicians alike in the context of HIV infection would be likely to bring to medical attention skin conditions that might otherwise be disregarded or that might be addressed without producing a medical-record entry. Thus, differences in the frequency and intensity of patient evaluation, related to HIV status, rather than (or in addition to) true biologic differences between groups of patients could have produced the associations noted.

James R. Johnson, M.D.
University of Minnesota, Minneapolis, MN 55455

1 References

1. 1

   Coopman SA, Johnson RA, Platt R, Stern RS. Cutaneous disease and drug reactions in HIV infection. N Engl J Med 1993;328:1670-1674
   Full Text | Web of Science | Medline

To the Editor:

Coopman et al. do not mention cutaneous pigmentary changes that accompany clofazimine therapy. This phenazine dye has been used in the treatment of leprosy, discoid lupus erythematosus, and pyoderma gangrenosum and in combination with other agents, such as clarithromycin, in the treatment of patients with AIDS who are infected with Mycobacterium avium complex1.

Clofazimine reportedly produces pink-to-brownish-black discoloration in up to 75 to 100 percent of patients receiving it,2 and it has been associated with ocular disorders, including keratopathy and retinal degeneration3. We have noted intense, mahogany-brown darkening of the skin in relation to clofazimine therapy in patients with AIDS who were being treated for M. avium complex infection. The changes become obvious after a six-month course of therapy with the usual dose (100 mg daily). There is no association with exposure to light, cortisol deficiency, or hyperbilirubinemia. Light-microscopical and electron-microscopical studies of the skin in one of these patients did not demonstrate specific histologic abnormalities like those described in two patients with leprosy4.

Since clofazimine is likely to be used more frequently as adjunctive therapy when drug-resistant strains of M. avium complex are encountered, clinicians should be aware of this dramatic side effect.

Lynne C. Krop, Pharm.D.
David P. Johnson, M.D.
Veterans Affairs Medical Center, Bay Pines, FL 33504

4 References

1. 1

   Holdiness MR. Clinical pharmacokinetics of clofazimine: a review. Clin Pharmacokinet 1989;16:74-85
   CrossRef | Web of Science | Medline

2. 2

   Clofazimine. In: McEvoy GK, ed. AHFS drug information 93. Bethesda, Md.: American Society of Hospital Pharmacists, 1993:493-7.
   

3. 3

   Font RL, Sobol W, Matoba A. Polychromatic corneal and conjunctival crystals secondary to clofazimine therapy in a leper. Ophthalmology 1989;96:311-315
   Web of Science | Medline

4. 4

   Job CK, Yoder L, Jacobson RR, Hastings RC. Skin pigmentation from clofazimine therapy in leprosy patients: a reappraisal. J Am Acad Dermatol 1990;23:236-241
   CrossRef | Web of Science | Medline

Author/Editor Response

The authors reply:

To the Editor: We shared Dr. Johnson's concern that the high frequency of dermatologic diagnoses that we observed in our cohort with HIV infection and the increases in these rates as HIV infection progressed might in part reflect the biases that he suggests. Our data, partly presented in Table 3 of our article, suggest that these biases did not substantially affect our results. To evaluate the extent to which members of our cohort were more likely to have skin diagnoses recorded independently of an increase in their rate of occurrence, we examined the frequency of common disorders, including epidermal cysts, acne, and nevi, whose occurrence and diagnosis are unlikely to be related to either HIV infection or progression of this disease. As we noted, the rates of visits by our cohort for these diagnoses were not substantially different from the rates for the subjects without infection. Furthermore, these rates did not increase significantly as HIV infection progressed.

Krop and Johnson point out in their letter that pigmentary changes, which may be asymptomatic and hence not prompt the patient to seek medical advice, are commonly observed in association with some drugs used to treat HIV infection. Diagnoses of hyperpigmentation were relatively infrequent in our cohort (nine patients). Therefore, we did not separately report the frequency of this diagnosis (and many other infrequent diagnoses). Of the nine patients with a diagnosis of hyperpigmentation, seven had filled prescriptions for zidovudine (2 percent of the patients taking zidovudine), another well-known cause of hyperpigmentation. None of the four persons who filled prescriptions for clofazimine had a diagnosis of hyperpigmentation recorded. The frequency of the diagnosis of hyperpigmentation among patients in our cohort who took zidovudine was comparable to that reported in studies of adverse effects of this drug, but it was below that noted in some studies that have specifically attempted to determine the true prevalence of this finding irrespective of its clinical importance1,2. Our study principally documented skin diagnoses that were clinically important to clinicians and patients. We agree that our study is likely to have underestimated the frequency of incidental skin changes, such as drug-associated hyperpigmentation, that are well recognized by clinicians who prescribe these drugs, are not amenable to therapy, and are usually not a reason to discontinue drug therapy.

Robert S. Stern, M.D.
Richard Platt, M.D.
Serge Coopman, M.D.
Harvard Medical School, Boston, MA 02215

2 References

1. 1

   Poizot-Martin I, Gamby T, Dhiver C, et al. Hyperpigmentation of skin in HIV-infected patients. In: Abstracts of the Sixth International Conference on AIDS, San Francisco, June 20-24, 1990. San Francisco: University of California, 1990.
   

2. 2

   Groark SP, Hood AF, Nelson K. Nail pigmentation associated with zidovudine. J Am Acad Dermatol 1989;21:1032-1033
   CrossRef | Web of Science | Medline