Abstract

Cerebral toxoplasmosis or toxoplasmic meningoencephalitis (hereafter referred to as TE) was one of the first opportunistic infections to be described in human immunodeficiency virus (HIV) -infected patients. Treatment of TE has been relatively successful in comparison to other opportunistic infections. Prior to the introduction of highly active antiretroviral therapy (HAART), a median survival of over a year was reported for patients who could tolerate the toxicity of TE treatment. HAART is becoming increasingly widely available in sub-Saharan Africa, where the majority of HIV-infected patients live. Many patients in Africa are diagnosed with HIV only after developing opportunistic infections such as TE. Hence, the optimal management of opportunistic infections such as TE is important if the benefits of subsequently initiating HAART are to be seen.

The purpose of this review is to determine the most effective therapy for TE in HIV-infected adults. Different treatment regimens have been compared with regard to clinical and radiological response, mortality, morbidity, and serious adverse events.

A comprehensive search of relevant databases and other sources was conducted to identify relevant studies.

Randomised double-blinded trials were included.

Data were extracted using standardised forms and analysed using Rev Man 4.2.7 software.

Three trials were found to meet the inclusion criteria. Dannemann et al 1992 and Katlama et al 1996 compared pyrimethamine plus sulfadiazine (P+S) with pyrimethamine plus clindamycin (P+C). Torre 1998 compared P+S with trimethoprim-sulfamethoxazole (TMP-SMX). For the purposes of this review, clinical outcomes were analysed as complete or partial resolution vs. failure. Patients who crossed over or were lost to follow-up were analysed as failures.

Dannemann et al 1992 assessed 59 patients. Five of 26 (19%) patients randomised to P+C died in the first 6 weeks compared with 2 of the 33 (6%) patients randomised to P+S (relative risk (RR) 3.17; 95% CI 0.67-15.06). Complete or partial clinical response was obtained in 12 (46.2%) patients receiving P+C vs. 16 (48.5 %) patients receiving P+S (RR 0.95; 95% CI 0.55-1.64).

Katlama et al 1996 assessed 299 patients. Twenty-nine (19%) of the 152 patients randomised to P+C died compared with 22 (15%) of the 147 patients randomised to P+S (RR 1.27; 95% CI 0.77-2.11). We were unable to obtain data on the outcomes of patients who crossed over and therefore excluded these data from the analysis.

Dannemann et al 1992 and Katlama et al 1996 were analysed together for the outcome of death. The two treatment arms did not differ for death (RR 1.41; 95% CI 0.88-2.28).

Torre et al 1998 assessed 77 patients. There were no deaths during the study period. Twenty-eight (70%) of 40 patients randomised to TMP-SMX had a complete or partial clinical response compared with 26 (70%) of 37 patients randomised to P+S (RR 1.0; 95% CI 0.74-1.33).

The available evidence fails to identify any one superior regimen for the treatment of TE. The choice of therapy will often be directed by available therapy. Given the current evidence, TMP-SMX appears to be an effective alternative therapy for TE in resource-poor settings where P+S are not available.