Pruritic papular eruption in HIV-positive patients in Brazil

Abstract

Pruritic papular eruption (PPE) is a mucocutaneous manifestation associated with HIV infection which was first described in the early years of the HIV epidemic [1]. PPE is characterized by erythematous papules or macules with symmetrical distribution mainly on the trunk and extremities in the absence of other causes of pruritus; intense itching is the most common symptom, leading to scratching, excoriation and postinflammatory hyperpigmentation [2]. PPE tends to be a chronic condition, but individual lesions will appear and disappear. Various aetiological factors have been explored in relation to PPE including immunological responses to insect bites, while infectious agents other than the primary HIV infection do not appear to be implicated [2]. Histopathological and haematological findings have been inconclusive and sometimes inconsistent, with most studies being small [2]. The salient feature of PPE is that it has been strongly and consistently associated with low CD4+ cell count, as described recently in paediatric and adult HIV patients [3,4]. Here, we report the associations that we found between PPE, CD4+ cell count and HIV viral load in HIV patients in Brazil who attended the Hospital Universitário Cassiano Antonio Moraes (HUCAM), Vitória (Espírito Santo) between 2004 and 2010. Patients had a dermatological examination and 4-mm punch skin biopsy, with diagnosis of PPE based on clinical and histopathologic findings to exclude differential diagnosis as scabies, prurigo caused by insect bites, cutaneous drug reactions, lichen planus, acne, demodicosis, eosinophilic folliculitis and bacterial folliculitis [2]. CD4+ cell count and viral load measurements at the time of examination, if available, were obtained from medical records. Ethical permission for use of these data was obtained from HUCAM research ethics committee. Of 370 patients [190 (51.4%) male, 180 (48.7%) female], median age 39 (range 15–82), 17.6% (65/370) had PPE, with no difference in prevalence by age (Pt test = 0.13) or sex (PChi-squared = 0.23). PPE was generalized (more than one body area) in 60.0% (39/65) of cases and PPE manifested mostly [76.9% (50/65)] as papules. Median CD4+ cell count was lower in patients with PPE [266 (interquartile range (IQR) 114–524) cells/μl, n = 45] than without PPE [400 (IQR 238–568) cells/μl, n = 159] (PKruskal–Wallis = 0.02) (Fig. 1a) and viral load was higher [1331 (IQR 5–171797, n = 40) copies/ml compared with 5 (IQR 5–569) copies/ml, n = 133] (PKruskal–Wallis = 0.001) (Fig. 1b); 32.5% (13/40) of patients with PPE had undetectable viral load (<10 copies/ml) compared with 51.9% (69/133) of patients without PPE (PChi-squared = 0.03). Ten patients with PPE were not taking antiretroviral therapy (ART) (15.4%) compared with 56 patients (18.4%) without PPE (PChi-squared = 0.57).Fig. 1: (a) CD4+ cell count in HIV patients with (n = 45) and without (n = 159) pruritic papular eruption [each box indicates the median (horizontal line) and interquartile range, whiskers include data points within 1.5 interquartile range of nearer quartile]; (b) Log10 HIV viral load in HIV patients with (n = 40) and without (n = 133) pruritic papular eruption (each point represents a single measurement, horizontal lines indicate medians).The strong association that we found between low CD4+ cell count and PPE replicates findings in most, but not all [5], previous studies of HIV-related dermatoses. PPE in the early years of the HIV epidemic was considered pathognomonic of AIDS and is potentially prognostic of ART failure [6,7]. Given that the prevalence of PPE has remained relatively unchanged among HIV patients even in the era of ART and that countries such as Brazil continue to see large numbers of new HIV cases each year (44 000 in 2018 http://www.aids.gov.br/pt-br/pub/2019/boletim-epidemiologico-de-hivaids-2019), more consideration needs to be given to identifying effective treatments to relieve debilitating and stigmatizing symptoms of HIV-related dermatological conditions and to improve patient quality of life [2]. Acknowledgements We thank Professor Elton Lucas and Professor Tania Reuter from the Hospital Universitário Cassiano Antonio Moraes (HUCAM) for their advice. This study received no specific funding. Conflicts of interest The authors have no conflicts of interest to declare.