Declining number of patients with cryptococcosis in the Netherlands in the era of highly active antiretroviral therapy

Abstract

Cryptococcosis is caused by the encapsulated yeast-like fungus Cryptococcus neoformans. It is the most common life-threatening fungal infection in individuals with compromised cellular immunity, such as patients with malignancies, with AIDS, or receiving immunosuppressive therapy. Since the beginning of the AIDS epidemic, major changes in the epidemiology of cryptococcosis have occurred. In France, an increase in the number of cases was reported because of an increase in HIV-related cases [1]. The introduction of highly active antiretroviral therapy (HAART) in 1996 has changed the natural history of HIV infection by reducing the incidence of AIDS-related opportunistic disorders [2,3]. Before 1986, the annual incidence of cryptococcosis in the Netherlands varied from 0.2 to 0.5 per 1 million inhabitants [4]. To gain insight into the epidemiological spectrum of this disease during the following years we analysed cases of cryptococcosis retrospectively. Here we report the changes in the rate of new cases of cryptococcosis in the Netherlands between 1986 and 2000. The patient population described here was identified in two different ways. The majority was recovered from the registers of the Netherlands Reference Laboratory for Bacterial Meningitis in Amsterdam. Almost all clinical microbiology laboratories in the Netherlands collaborate by sending their results from patients with bacterial meningitis or cryptococcosis to the Netherlands Reference Laboratory for Bacterial Meningitis. Moreover, the medical microbiology departments of five major university hospitals were contacted for additional cases of cryptococcosis, because most patients with cryptococcosis are treated at university hospitals. Epidemiological data (age, sex, hospital) and relevant medical history (underlying disease, medication) for all patients were obtained through their treating physicians, or by reviewing their charts. The results of our study were compared with the AIDS Therapy Evaluation Netherlands (ATHENA) database. That nationwide multi-centre clinical cohort study started in 1998 and aims to monitor the use and effectiveness of HAART. A total 268 patients were identified with cryptococ cosis. HIV infection was present in 203 patients (76%). Other predisposing factors included: the use of immunosuppressive medication for various reasons (14 patients, 5%), including organ transplantation (seven patients, 2.5%) and connective tissue disorders (two patients, 1%); haematological malignancy (16 patients, 6%), and in two patients (1%) a rare hereditary case of impaired cellular immunity. In nine cases (3%) no predisposing factors could be detected despite an extensive evaluation. In 24 cases (9%), no information about an underlying condition was available. Most of the HIV-infected patients had advanced disease, with a median CD4 positive T cell count of 35 cells/μl (range 3–240). A sharp increase in the number of patients with cryptococcosis was seen in 1987, as a result of the increasing number of HIV-infected individuals. During the first 10 years, the mean annual rate of cryptococcosis was 21.3 cases per year. Interestingly, this declined by 55% to only nine and 10 cases per year in 1996 and 1997, respectively; and more or less stabilized at 13 patients over 1998 and 1999. This decrease was mainly caused by the reduced occurrence of cryptococcosis in HIV-infected individuals (Fig. 1).Fig. 1.: Number of patients with cryptococcosis during 1986–1999, accumulating estimated AIDS diagnoses and total number of patients on highly active antiretroviral therapy included in the AIDS Therapy Evaluation Netherlands (ATHENA) study cohort. ▪ HIV-infected patients; ▩ non-HIV-infected patients; □ unknown; —▪— total of HIV-infected patients included in ATHENA; —&●— accumulating AIDS diagnosis.The annual number of new AIDS cases in the Netherlands has decreased [5]. Although efforts have recently been made to estimate the incidence of HIV-infected individuals and monitor these patients on a regular basis [6], to our knowledge no published data exist that substantiate the direct relationship between the decline of cryptococcosis and the use of different treatment regimens for advanced HIV infection in the Netherlands. To strengthen further the suggestion that the decline of cryptococcosis is correlated with the introduction of HAART, we have looked at the ATHENA database. By the end of 1999, 1235 antiretroviral-naive and 1218 pretreated HIV-infected patients were enrolled. This representative cohort of 2500 HIV-infected patients shows that a large population of patients with AIDS before 1996 switched to HAART, and that newly diagnosed patients are offered HAART as the standard therapy for advanced HIV infection (Fig. 1). Other authors have reported a similar decrease in HIV-infected patients with cryptococcal meningitis up to 1997, and questioned whether this decrease would be sustained because of virological failure on the new antiretroviral drug regimens [7]. In our study, none of the HIV-infected individuals were on HAART in 1996 and 1997. However, four patients used HAART in 1998 and 1999, of which one was failing therapy. This retrospective analysis of cryptococcosis cases demonstrates changes in the occurrence of cryptococcosis that are mainly caused by changes in the epidemiology of AIDS in the Netherlands and are temporally associated with the introduction of HAART. To evaluate their course further in HIV-infected patients, continuous monitoring of opportunistic infections is essential. Leontine J. R. van Eldenab Annemiek M. E. Walenkampab Myriam M. Lipovskyab Peter Reissc Jacques F. G. M. Meisd Siem de Mariee Jacob Dankertf Andy I. M. Hoepelmanab