Cryptococcal Meningitis Treatment Strategies in Resource-Limited Settings: A Cost-Effectiveness Analysis

David Boulware and colleagues assess the cost effectiveness of different treatment strategies in low- and middle-income countries for cryptococcal meningitis, one of the most common opportunistic infections of people with HIV. Background: Cryptococcal meningitis (CM) is the most common form of meningitis in Africa. World Health Organization guidelines recommend 14-d amphotericin-based induction therapy; however, this is impractical for many resource-limited settings due to cost and intensive monitoring needs. A cost-effectiveness analysis was performed to guide stakeholders with respect to optimal CM treatment within resource limitations. Methods and Findings:: We conducted a decision analysis to estimate the incremental cost-effectiveness ratio (ICER) of six CM induction regimens: fluconazole (800–1,200 mg/d) monotherapy, fluconazole + flucytosine (5FC), short-course amphotericin (7-d) + fluconazole, 14-d of amphotericin alone, amphotericin + fluconazole, and amphotericin + 5FC. We computed actual 2012 healthcare costs in Uganda for medications, supplies, and personnel, and average laboratory costs for three African countries. A systematic review of cryptococcal treatment trials in resource-limited areas summarized 10-wk survival outcomes. We modeled one-year survival based on South African, Ugandan, and Thai CM outcome data, and survival beyond one-year on Ugandan and Thai data. Quality-adjusted life years (QALYs) were determined and used to calculate the cost-effectiveness ratio and ICER. The cost of hospital care ranged from $154 for fluconazole monotherapy to $467 for 14 d of amphotericin + 5FC. Based on 18 studies investigating outcomes for HIV-infected individuals with CM in resource-limited settings, the estimated mean one-year survival was lowest for fluconazole monotherapy, at 40%. The cost-effectiveness ratio ranged from $20 to $44 per QALY. Overall, amphotericin-based regimens had higher costs but better survival. Short-course amphotericin (1 mg/kg/d for 7 d) with fluconazole (1,200 mg/d for14 d) had the best one-year survival (66%) and the most favorable cost-effectiveness ratio, at $20.24/QALY, with an ICER of $15.11 per additional QALY over fluconazole monotherapy. The main limitation of this study is the pooled nature of a systematic review, with a paucity of outcome data with direct comparisons between regimens. Conclusions: Short-course (7-d) amphotericin induction therapy coupled with high-dose (1,200 mg/d) fluconazole is “very cost effective” per World Health Organization criteria and may be a worthy investment for policy-makers seeking cost-effective clinical outcomes. More head-to-head clinical trials are needed on treatments for this neglected tropical disease. : Please see later in the article for the Editors' Summary. Background: Cryptococcal meningitis, a fungal infection of the membranes around the brain and spinal cord, affects about a million people every year (most of them living in sub-Saharan Africa and Southeast Asia) and kills about 640,000 people annually. People become infected with Cryptococcus neoformans, the fungus that causes cryptococcal meningitis and which is found in soil and dirt, by breathing it in. In healthy individuals, infection rarely causes disease. But in people living with AIDS, whose immune system has been damaged by HIV infection, and in people whose immune system is compromised for other reasons, the fungus can invade and damage many organs, including the brain. Cryptococcal meningitis, the symptoms of which include fever, stiff neck, headache, and vomiting, is diagnosed by looking for the fungus in fluid taken from the spinal cord in a procedure called a lumbar puncture. Cryptococcal meningitis is treated with antifungal drugs such as amphotericin, fluconazole, and flucytosine (induction therapy); recurrence of the infection is prevented by taking fluconazole daily for life or until the immune system recovers. Why Was This Study Done?: The World Health Organization (WHO) recommends a 14-day regimen of intravenous (injected) amphotericin and oral flucytosine or fluconazole for induction therapy of cryptococcal meningitis. Unfortunately, this regimen is impractical in many resource-limited settings because of the cost of the drugs and hospital care and the need for intensive monitoring—amphotericin is extremely toxic. Consequently, high-dose fluconazole monotherapy is the usual treatment for cryptococcal meningitis in resource-limited countries, although this regimen is much less effective. Another regimen that has improved survival in trials is flucytosine with fluconazole for two weeks. However, flucytosine is very expensive and is not licensed in most sub-Saharan African countries. Stakeholders in developing countries badly need guidance, therefore, on which induction treatment for cryptococcal meningitis they should recommend to optimize outcomes in their particular countries. In this cost-effectiveness analysis (a study that compares the costs and health effects of different interventions), the researchers use costs in Uganda to estimate the survival, cost, and cost per benefit associated with various induction treatments for cryptococcal meningitis in HIV-infected patients. What Did the Researchers Do and Find?: The researchers calculated the overall cost of six induction treatments using 2012 healthcare costs in Uganda for medications, supplies, and hospital care, and average laboratory costs for monitoring treatment from three African countries. They used data from published trials of cryptococcal meningitis treatment in resource-limited areas to estimate ten-week and one-year survival, life expectancy, and quality-adjusted life years (QALYs, the number of years of life added by an intervention, adjusted for the quality of life) for each intervention. Finally, they calculated the cost-effectiveness ratio (cost per QALY gained) and the incremental cost effectiveness ratio (ICER, the additional cost of a treatment strategy compared to fluconazole monotherapy divided by the incremental improvement in QALYs) for each intervention. The estimated costs per person for each induction treatment strategy ranged from US$154 for 14 days of fluconazole monotherapy to US$467 for 14 days of amphotericin plus flucytosine. Estimated average one-year survival was lowest for fluconazole (40%) and highest for short-course (seven days) amphotericin plus 14 days of fluconazole (66%), similar to other amphotericin-based treatments. Cost-effectiveness ratios ranged from US$20 per QALY for short-course amphotericin plus fluconazole to US$44 per QALY for 14 days of amphotericin plus flucytosine. Short-course amphotericin plus fluconazole had the lowest ICER (US$15.11 per additional QALY over fluconazole monotherapy). What Do These Findings Mean?: These findings suggest that, among the treatments investigated, a seven-day course of amphotericin with high-dose fluconazole for at least two weeks is the most cost-effective induction treatment for cryptococcal meningitis in Uganda. Although this result should be generalizable to other African countries, it needs to be treated with caution because very few trials have actually looked at the clinical effectiveness of this particular regimen. While short short-course amphotericin appears to be substantially more effective than fluconazole monotherapy, large-scale trials comparing short-course amphotericin regimens with more traditional 14-day regimens in resource-limited countries must be undertaken before short-course amphotericin-based treatments are adopted. Notably, however, if these trials confirm that survival with short-course amphotericin with fluconazole is about 30% better than with fluconazole alone, the researchers calculate that moving to short-course amphotericin could save about 150,000 lives every year in sub-Saharan Africa at a cost of US$220 per life saved. Additional Information: Please access these websites via the online version of this summary at http://dx.doi.org/10.1371/journal.pmed.1001316.