Abstract
There is little information about the national burden of cryptococcal meningitis (CM) in African countries affected by the HIV/AIDS epidemic. From April 2005 onwards, we used national supervision visits of all health facilities that provided antiretroviral therapy to collect data on the number of new patients diagnosed and treated for CM in the previous quarters – using mainly fluconazole registers. For two 12-month reporting periods, there were 2125 and 2464 patients suffering from CM, giving an estimated annual incidence of 2.2% and 2.6%, respectively, of those infected with HIV in Malawi. Between 40–50% of all patients with CM were diagnosed and treated at central hospitals; no more than 1% were diagnosed and treated at smaller antiretroviral therapy sites. These data are useful for quantifying the need for better diagnostic reagents and antifungal drugs.
Introduction
Malawi is a poor country in central-southern Africa, with a population of 12.8 million. The first case of AIDS was reported in 1985. Since then, the country has experienced a rapid and massive epidemic. The total number infected with HIV is estimated to be 930,000. Every year another 90,000 are infected with HIV and 86,000 people die from AIDS. 1
Cryptococcal meningitis (CM) is one of the few AIDS-defining conditions that can be reliably diagnosed in the resource-constrained environments of sub-Saharan Africa. It requires a lumbar puncture and an India ink stain, or a more sensitive cryptococcal antigen test of cerebro-spinal fluid (CSF). CM is one of the major opportunistic infections in the high HIV-prevalence areas of sub-Saharan Africa. 2 It is also the leading cause of meningitis in central and southern Africa, accounting for 26% of cases in Malawi, 3 31% in the Central African Republic 4 and 45% in Zimbabwe. 5
Despite these reports, we have seen no published data on national estimates of the burden of CM. During our quarterly national supervision visits to health facility sites which provide antiretroviral therapy (ART), 6 we collected data on key opportunistic infections and we used this opportunity to obtain estimates of the burden of CM in Malawi.
Methods
The increase of the use of ART on a national level started in 2004. By the second quarter of 2005, all 60 health facility sites earmarked for Round-1 ART service provision had begun providing treatment. These sites included:
All central and district hospitals; Most major mission hospitals; Hospitals for the Malawi defence and police force; A small number of health centers or clinics.
Another 49 smaller health facility sites were earmarked for Round-2 ART service provision and were all providing treatment by early 2007.
All ART sites received central unit supervisory and monitoring visits on a quarterly basis from 2004. From April 2005, in addition to collecting specific ART data, we visited the pharmacy of each site and used the Fluconazole Register to count the number of new patients with CM (these registers are provided to health facilities by Pfizer Limited through the Government Central Medical Stores). Fluconazole is given to Malawi free of charge by Pfizer, and is only used for patients diagnosed with HIV-related CM and oesophageal candidiasis. The country does not use amphotericin B or flucytosine in the public health sector, and oral fluconazole is the only treatment available for CM. There are columns in the register to indicate provision of the drug for new patients and for follow-up patients, along with the date of each prescription. At each visit, the number of new patients with CM started on fluconazole was recorded for the previous quarter.
If an ART site did not have a Fluconazole register (which included some of the health centres and smaller clinics), or the register had not been completed for the previous quarter, we searched the ART master cards for the reasons for starting ART. 7 For patients classified in the World Health Organization as clinical Stage 4, we looked especially for CM. Additionally, in the small ART sites, we questioned clinical, nursing and laboratory staff to establish if any patient had been diagnosed and treated with CM in the previous quarter.
The number of patients with CM in an ART site for a quarter was entered to a structured proforma, and then transcribed to an Excel spread sheet. From the quarterly spread sheets, data were aggregated on the number of patients treated for CM in Malawi for six- and 12-month periods: these data were also analyzed in relation to the ART site. The Malawi National Health Science Research Committee provided approval for the collection and use of routine programmatic data for monitoring and evaluation.
Results
The number of patients with CM during six-monthly and 12-monthly periods, and the number and percentage diagnosed and treated at central hospitals and at Round-2 ART sites, are shown in the Table 1. For each of the 12-month reporting periods, there were 2125 and 2464 patients suffering from CM, giving an estimated annual incidence of 2.2% and 2.6%, respectively, of those infected with HIV. Between 40–50% of all patients with CM were diagnosed and treated at central hospitals - no more than 1% were diagnosed and treated at Round-2 ART sites.
The number of antiretroviral treatment (ART) sites and the number of patients diagnosed and treated for cryptococcal meningitis (CM) in Malawi between April 2005 and September 2007
Discussion
This study shows that 2000 to 2500 patients are diagnosed and treated for CM each year in the public sector in Malawi, giving a national estimated annual incidence of 2.0–2.5% of all those infected with HIV. Of course, patients newly diagnosed with HIV infection are not at risk of CM as this disease only becomes apparent with severe immunosuppression.
This study has several limitations. First, CM was only recorded from public sector health facilities providing ART. However, very few patients were diagnosed from Round-2 ART sites and this suggests that an equally small number of patients would be diagnosed in the other public sector health centres or health posts around the country. The private sector in Malawi is also small, and we believe, would add few cases of CM to the national total.
Second, the numbers of patients with CM were mainly obtained from the Fluconazole Registers. These may not have been properly completed, and, in particular, the columns for new and follow-up patients may have been wrongly ticked. However, the six-monthly national aggregates showed similar numbers of cases, suggesting that registration processes are largely correct.
Third, there is the issue about diagnosis. In Malawi, only the India ink stain is used for diagnosis, with cryptococcal antigen testing and/or culture for Cryptococcus neoformans generally being absent in the public sector. India ink stains alone may miss up to 25% of CM cases 8 and, thus, the true number of cases may be underestimated in this study.
Despite these limitations, the study gives one of the first estimates of the national burden of CM in an African country heavily affected by the HIV/AIDS epidemic. Also, the data are useful when it comes to estimating needs for better diagnostic reagents and drug forecasting.