National adult antiretroviral therapy guidelines in South Africa: concordance with 2003 WHO guidelines?

Abstract

We read with interest the article by Beck and colleagues [1] who examined the adult antiretroviral therapy (ART) guidelines in 43 World Health Organization (WHO) ‘3 by 5’ focus countries. The authors found that the national guidelines of a majority of countries had a good degree of concordance with the WHO 2003 guidelines [2]. Although concordance was noted to be inversely related to health expenditure per capita, the authors did not further explore the reasons why some countries have adopted guidelines that differ from the current WHO recommendations. One such country is South Africa, which has among the highest per capita income of countries in subSaharan Africa and also has much better healthcare infrastructure than most. Despite these resources, the South African national ART programme currently bases its treatment guidelines on the former WHO 2002 guidelines that recommend ART only for patients with WHO stage 4 disease (AIDS) or a blood CD4 cell count of less than 200 cells/μl [3]. We believe these guidelines advocate treatment at too late a stage of disease and that they represent a compromise that may substantially undermine the effectiveness of the programme in the long term. The key goals of an ART programme are to prevent morbidity and mortality. We have recently demonstrated in a historical natural history cohort in Cape Town, South Africa, that 52% of HIV-associated deaths actually occur before the development of stage 4 disease [4]. Moreover, the rate of progression to AIDS among patients with CD4 cell counts of 200–350 cells/μl was almost twice that reported in natural history cohorts in high-income countries [4,5]. Therefore, under the current South African national guidelines, much HIV-associated morbidity and mortality in this setting occurs among patients before the eligibility for ART. Data from ART programmes also suggest that treatment is being initiated too late. Within a community-based programme in Cape Town, the mortality rate among referred individuals eligible to start ART is extremely high, exceeding 30 deaths per 100 person-years [6]. Moreover, during 3 years follow-up of this cohort, 87% of mortality occurred in the interval just before treatment initiation or during the first 16 weeks of ART [7]. Collectively, these data indicate that patients are arriving with disease that is too far advanced. More worryingly, these data are likely to be indicative of a huge burden of mortality occurring upstream of the ART programme. South Africa is at the epicentre of the continent's HIV-associated tuberculosis epidemic, with tuberculosis notification rates in many townships reaching almost unprecedented levels [8]. The extent to which ART may assist in tuberculosis control at the community level is highly dependent upon how early ART is initiated and upon the extent to which ART reduces the incidence of tuberculosis during long-term treatment [9]. Under the South African ART guidelines, pulmonary tuberculosis (WHO stage 3 disease) per se is not a criterion for the consideration of ART. This may well represent a critical missed opportunity; by the time our patients initiate ART, 52% of them have already completed treatment for one or more episodes of tuberculosis and 25% are receiving treatment for current tuberculosis [10]. The later the initiation of ART, the greater the burden of tuberculosis that accrues. Moreover, the long-term incidence of tuberculosis during ART in this cohort remains five to 10-fold higher than that among HIV-non-infected individuals despite excellent virological outcomes [10]. This is likely to reflect the suboptimal restoration of tuberculosis-specific immunity, which is a further factor strongly associated with the late initiation of ART [11]. In view of these outcomes, the late initiation of ART could paradoxically undermine tuberculosis control in South Africa [12]. Clearly, in addition to medical factors, economic and logistic constraints also affect the ART eligibility criteria set by different national ART programmes. However, economic data also indicate the benefits of earlier treatment. Cost-effectiveness analyses using local data from Cape Town indicate that it is more cost effective to treat patients with less advanced disease, although the overall investment is greater [13,14]. Collectively, these data point in the same direction, favouring earlier initiation of ART than is at present recommended in South Africa. Good medical practice mandates that treatment guidelines should not be ‘cast in stone’ but should be continually reviewed and revised in the light of emerging data. We conclude that the six million people living with HIV in South Africa might be much better served by the national ART guidelines being brought up to date with current revised WHO guidelines. As Beck and colleagues [1] point out, the WHO guidelines have been developed so that ‘an effective and appropriate standard of care can be maintained’. Sponsorship: S.D.L. is funded by the Wellcome Trust, London, UK, and R.W. is funded partly by the National Institutes of Health, USA, RO1 grant A1058736-01A1.