Abstract
The scale-up of antiretroviral treatment (ART) of HIV/AIDS in resource-limited settings has been one of the largest public health operations of our time, and by the end of 2009 more than 5 million people were receiving ART in low- and middle-income countries. Although several studies on ART efficacy in Africa have been published, the majority have been carried out in larger cities and usually with short follow-up time. To date, there is a paucity of research from rural settings, which often face additional challenges, such as shortages of health workers, transport difficulties and other logistical constraints. In the first part of the present study we reported long-term clinical and virological outcomes of ART in a rural hospital in Tanzania.
In the second part of this study we addressed the lack of field-friendly tools for virological monitoring of patients on ART. Viral load and resistance testing, as recommended in high-income countries, are rarely available in resource-limited settings due to high costs and stringent requirements for storage and transport of plasma. Consequently, treatment failure in such settings usually passes unnoticed until patients develop severe immunodeficiency, at which stage widespread resistance is likely. Dried blood spots (DBS) are easy to collect and store, and can be a convenient alternative to plasma. Unlike plasma, DBS can be stored and shipped at ambient temperature, thus avoiding the need for cold chain and speedy transport to the laboratory. In our study, under field conditions in rural Tanzania, we assessed the performance of DBS in virological monitoring of patients on ART.
Haydom Lutheran Hospital has provided ART to HIV-infected patients since 2003. A combination of stavudine or zidovudine with lamivudine and either nevirapine or efavirenz is the standard first-line regimen, in accordance with WHO and Tanzanian guidelines. All patients who started ART from October 2003 through December 2007 were included in a longitudinal cohort study. Standard techniques of survival analysis were used to estimate mortality and identify predictors of mortality. Virological efficacy and emergence of drug resistance was assessed in patients who had completed at least 6 months of first-line ART. Viral load and resistance results obtained with a standard plasma-based method were compared with results obtained with the use of DBS.
Paper I and II were epidemiological studies. We found a high mortality in this cohort, particularly the first months after initiating ART: estimated mortality was 19.2%, 29.0% and 41.7% after 3, 12 and 36 months, respectively. Anemia, malnutrition and thrombocytopenia were strong and independent predictors of mortality. A prognostic model based on hemoglobin level appeared to be a useful tool for initial risk assessment: estimated one year mortality was 3.7% in patients without anemia compared to 55.2% in those with severe anemia (<8 g/dL). Among patients who survived the first 6 months and remained in care, however, we found good long-term virological efficacy: viral suppression (<400 copies/mL) was observed in 187 of 212 (88.2%) patients after a median follow-up time of 22.3 months. In total, 18 patients harbored at least one clinically significant drug-resistance mutation, of whom 5 had thymidine analogue mutations associated with broad cross-resistance to nucleoside reverse transcriptase inhibitors. Although the overall prevalence of drug resistance was relatively low, it increased with time and reached approximately 15% after 3-4 years on ART.
Paper III and IV were laboratory studies. First, we compared the viral load levels in 98 plasma-DBS pairs from patients on ART. In a linear regression model there was a strong correlation, with an R2 value of 0.75, between the two specimen types. Viral loads were on average slightly higher in plasma than DBS, but the mean difference was only 0.04 log10 copies/mL. However, DBS had reduced sensitivity to detect HIV-1 RNA in samples with low-level viraemia (<3000 copies/mL). Subsequently, we compared genotypic resistance results from DBS with those of plasma in 36 ART-experienced individuals with treatment failure (viral load >1000 copies/mL). Overall, 34 of 36 (94%) DBS specimens were successfully genotyped, and there was high concordance between mutations found in plasma and DBS. Thirty of 34 (88%) patients had identical resistance profiles to antiretroviral drugs in plasma and DBS.
In conclusion, we found a high early mortality in this cohort. Simple laboratory markers, especially hemoglobin level, appeared to be useful for initial risk assessment, and can be of particular use in settings without access to CD4 cell counts. Long-term virological efficacy rates were favorable, and drug resistance appeared to develop at the same rate as in high-income countries. Finally, we found that the use of DBS was a feasible and reliable option for viral load and resistance testing, and we believe that DBS could simplify virological monitoring in resource-limited settings.