Abnormal insect bite reactions: A manifestation of immunosuppression of HIV infection?

Introduction

Since 1986, our department of sexually transmitted disease has been monitoring the trend of the HIV epidemic and the prevalence of HIV rose from 0.2% in 1986[1] to 31% in 1998 (unpub-lished data).

Pruritic eruptions are a common initial mani-festation in persons with acquired immunodeficiency syndrome (AIDS).[2],[3] These extremely pruritic lesions are of uncertain significance. According to earlier reports, more than 50% of these patients have associated opportunistic infections.[4] These pruritic eruptions can be an initial manifestation of HIV infection. Our experience is not very different and we have a number of patients who were diagnosed to have HIV infection since they presented to us with itchy papules.

Pruritic eruptions in HIV infected patients refer to a number of conditions which include eosinophilic folliculitis, scabies or other hypersensi-tivity phenomena. Many of our patients presenting with pruritic eruptions and HIV infection have lesions predominently on the exposed parts of the body. Considering the history of insect bites, high prevalence of insect bite in this part of the world and development of lesions morphologically similar to insect bite reactions at the site of insect bite, it is likely that these are due to delayed hypersensi-tivity reactions to insect bites (to which they have already been desensitised in early life) which reappear as a result of failing immunity. This prompted us to do a pilot study to determine whether the appearance of pruritic papular erup-tions on the exposed parts of these patients is a predictor of impending immunosuppression due to HIV infection. We used the CD4 count as an indicator of immunosuppression in these patients.

Patients and Methods

Ten consecutive patients who presented with persistent pruritic eruptions of recent onset (range 1 week to 2 months) with no opportunistic infection other than tuberculosis of recent onset (not more that 2 months) were selected for the study. There were 6 males and 14 females of age ranging from 18 to 55 years. All these patients had erythematous excoriated papules, predomi-nantly on the exposed parts of the body. Of these, 8 patients were diagnosed earlier by the Western blot method to be infected with HIV. The remain-ing two patients presented to us with papular erup-tions. At the inclusion to the study 8 of them were not taking any drugs other than symptomatic measures like antihistamines. Two patients (no: 4 and 8) were taking antituberculosis therapy with rifampicin, INH, ethambutol and pyrazinamide, but they gave a history of pruritic lesions even before initiation of treatment. All patients were investigated for evidence of hepatic and renal disease and HIV related morbidity. Haemogram, chest x-ray, creatinine and liver function test were done for all patients. Biopsies were taken from representative skin lesions in 2 patients who had tuberculosis; in order to rule papulonecrotic tuber-culid presenting as pruritic lesion, biopsy was taken from a representative lesion of patient no: 3 also. Biopsy showed perivascular lymphocytic infiltration and nonspecific dermatitis in all the 3 cases.

CD4 counts were done in all these patients. For CD4 counts, peripheral blood samples were collected between 8 and 9 AM, in order to mini-mize any influence due to diurnal variation. Labora-tory reference ranges for CD4 counts in South Indian adults were previously generated by collect-ing blood samples as above and processing them according to the protocol of Becton Dickinson, USA. Fresh whole blood was analysed for cells bearing both CD3 and CD4 surface markers by flow cytometry (FACScan), using Simultest reagents and Simulset software (Becton Dickinson USA)

Results

The CD4 counts in these 10 patients are given in [Table - 1]. CD4 counts less than 400/cmm were obtained in 8 (mean +- SD was 215 +- 90, with a range of 170-430/cmm). Of the two pa-tients with tuberculosis, one had a count of 320/ cmm and the other 170/cmm. The normal range of CD4 counts in healthy adults in southern India (N=110) is 500-1200/cmm (mean +- SD = 850 +-376/cmm).[3],[5] The difference between the mean +- SD of the CD4 counts of the patients with papular eruptions and the normal healthy controls was statistically significant (p< 0 001). The biopsied skin of the three patients showed perivascular lymphocytic infiltration.

Discussion

In our study all the 10 cases with pruritic eruptions and HIV infections showed CD4 count less than 500 and 8 of them had CD4 count less than 400. Two patients who had already devel-oped tuberculosis had still lower CD4 counts (pa-tients no: 4 and 8). Two of our patients were de-tected to be infected with HIV when they came to us for the treatment of pruritic eruptions (pa-tients No: 9 and 10). These findings support ourhypothesis that the appearance of pruritic erup-tions of the exposed parts of the body in HIV infected patients who are exposed to repeated in-sect bites can be a marker of fall in CD4 count and impending immuno deficiency. An exaggerated re-sponse to mosquito bites has been reported ear-lier in HIV infection,[6] natural killer cell Iymphocyto-sis,[7] malignant histiocytosis[8] and leukemia.[9] Immunosuppression with cyclophosphamide reverses immunological tolerance to dinitrochlorobenzene in guinea pigs.[10] Similar lesions are reported in individuals with idiopathic CD4+ lymphocytopenia.[11] It is possible that as a result of immunosuppression, these patients react to antigens to which they were earlier desensitized (from repeated insect bites). The normal response to insect bite is determined by previous exposure. An individual not previously exposed to insect bites does not develop any response to it. Subsequent bites produce a delayed reaction with pruritic wheals which develop 24 hours after the bite and persist for about 24 hours. After repeated bites over several weeks, the response changes with the appearance of an immediate wheal followed by a delayed reaction.[12]

Recall reaction to arthropod bites in which mosquito salivary gland antigens in the skin from previous bites react with elevated levels of anti mosquito salivary gland antigen antibodies from nonspecific B cell activation of HIV infection is suggested as a possible mechanism of these le-sions by Penneys et al.[13] But this does not explain the occurrence of similar lesion, following chemo-therapy, malignancy and idiopathic CD4 lymphocy-topenia.[9],[10],[11] It is possible that a reversal of CD4+: CD8+ and Th1: Th2 lymphocyte reaction occurs as part of HIV infection, malignancy or chemotherapy resulting in faulty recognition of antigen. This has been suggested as a mechanism of development of pruritic eruptions in HIV infections.[14] Qualitatively and quantitatively less effective Langerhans cells in the skin of HIV infected individuals also can be a contributory factor for the altered immunological response to the insect bite.

If proved to be a marker of immunosuppres-sion, appearance of pruritic papular eruption will be a useful marker to predict immunosuppression in HIV infected individuals. This assumes even more importance when effective anti-retroviral therapy and prophylaxis regimens are becoming popular.