Prevalence and trends of advanced HIV disease among antiretroviral therapy-naïve and antiretroviral therapy-experienced patients in South Africa between 2010-2021: a systematic review and meta-analysis

Efforts to quickly expand access to human immunodeficiency virus (HIV) care and treatment have led to a significant number of individuals starting antiretroviral therapy (ART). The most notable increase has occurred in South Africa, which is home to 7.5 million of the estimated 37.7 million people living with HIV (PLHIV) [1]. Access to ART in South Africa has grown from 103,300 to 2005 to 5,500,000 in 2021. Despite significant progress in expanding access to ART, the burden of Advanced HIV Disease (AHD) in South Africa remains high [1‐4]. AHD is defined as a CD4 count of fewer than 200 cells/mm³ or a WHO stage 3 or 4 event in adults and adolescents, and all children with HIV younger than five years old are considered to have AHD [5, 6]. The proportion of people presenting with AHD has remained relatively unchanged in the past five years, despite the number of PLHIV receiving ART in low- and middle-income countries (LMIC) more than doubling over this period [4, 7, 8]. These patients are at high risk of death mainly due to Tuberculosis and cryptococcal meningitis, even after starting ART (which can increase the inflammatory response) and the risk increases with decreasing CD4 cell count [9].

AHD is prevalent in Sub-Saharan Africa (SSA), with studies reporting between 32% and 71% of patients initiating care with AHD, and up to 60% of patients presenting for care with AHD after disengagement [10, 11]. A large multi-country cohort study using data from HIV treatment programs in 6 SSA countries during 2005–2018 showed that South Africa had the highest percentage of adults with AHD (59.7%) starting ART. However, the study also noted that the percentage of adults starting ART with AHD steadily declined over time [4].

Contrary to the expectation that PLHIV will progress from testing and early ART initiation to consistent drug adherence and viral suppression, many PLHIV instead need assistance re-engaging in care with AHD following treatment failure or interruption [12]. This is a concerning trend observed in countries with established HIV programs, where an increasing number of patients present for care with AHD after a period of treatment interruption or ART-experience [13]. In a study by Osler et al. [2] in South Africa, the proportion of ART-experienced patients with a CD4 count less than 50 cells/mm³ increased from 14.3% to 2008 to 56.7% in 2017 and this group seeks hospital care following ART interruption or virological failure.

Previous meta-analysis studies have primarily focused on analysing temporal trends of CD4 count at presentation to care and treatment initiation, globally [14], in developed countries [15] and in low and middle-income countries [16]. It is crucial to estimate the prevalence of AHD among ART-naive and ART-experienced patients to understand the outcomes of ART programs, guide HIV prevention efforts, and predict the need for adjunctive therapies [17‐19]. The true estimates of AHD among ART-naïve and ART-experienced patients in South Africa are not well known. This systematic review and meta-analysis aimed to estimate the prevalence of AHD among ART-naive and ART-experienced patients in South Africa, using data from studies published between January 1, 2010 and May 31, 2022. Furthermore, we sought to examine trends in the prevalence of AHD throughout this period.

This study employed a systematic review and meta-analysis design and the reporting of this systematic review was guided by the standards of the Preferred Reporting Items for Systematic Reviews and Meta-analyses (PRISMA) Statement [20]. The protocol for this systematic review was registered and published on the International Prospective Register of Systematic Reviews (PROSPERO) with the registration number: CRD42021245429.

This systematic review analysed 53 studies from 5 provinces in South Africa, involving 11,545,460 patients who received ART services between 2010 and 2022. The majority of the studies were cross-sectional analyses (24 studies), followed by longitudinal analysis of HIV programmatic data (13 studies), cross-sectional surveys (2 studies), prospective cohort studies (10 studies), and retrospective studies (4 studies). The overall prevalence of AHD among patients presenting to HIV services between 2010 and 2021 was 43.4%, however, there was substantial heterogeneity with this ranging between 4.5 and 83.1%. Due to the high heterogeneity between included studies, the pooled prevalence estimate should be interpreted with caution. Estimation of prediction intervals showed narrow range of expected prevalence of AHD among ART-naïve patients between 38.0% and 50.0% in a future, new systematic review and meta-analysis.

Subgroup analyses related to AHD diagnostic methods (CD4 versus WHO) and age did not explain the observed heterogeneity. Among the studies (n = 2) reporting on the presentation of ART-experienced patients, AHD was 58.6%. A time trend analysis revealed a decline of approximately 9.0% in the prevalence of AHD over the 10 years, with a prevalence of 52.4% in 2010 and 43.4% in 2022.

Since the start of the HIV programme, CD4 counts have been used to guide treatment decisions and identify those at risk for morbidity, mortality, and increased medical and psychosocial needs [5]. During the review period ART initiation guidelines were revised several times (patients were deemed eligible for ART if their CD4 count was < 200 cells/mm³ prior to 1 May 2010 or < 350 cells/mm³ after 1 May 2010 and 500 cells/mm³ between January 2015 and 31 August 2015) to expand form ART eligibility based on CD4 count and other co-morbidity, to all PLHIV becoming eligible for ART irrespective of CD4 count by 2016. As such, prior to and in 2010 the majority of PLHIV were at risk of starting ART with AHD whereas after 2010 this decreases due to guideline changes [5, 6].

Our study found that the proportion of patients presenting with AHD remains high but has decreased over time due to increased HIV testing and access to ART. Time trend analysis showed a 2% yearly decline in AHD among ART-naive patients. Our findings align with previous studies from both high-income and low- and middle-income settings [11, 14‐16, 28, 29], which also reported either a high proportion or a modest decline in patients with AHD. Given the moderate decline in the proportion of AHD among ART-naïve patients, these results emphasize the importance of expanding HIV testing and timely linkage to care, to promote early ART initiation.

A meta-analysis data, which reported a mean CD4 count at ART initiation in 27 countries of 186 cells/mm³, suggest at least one in four ART patients in SSA initiated ART with advanced disease in 2015 and CD4 count at presentation to care and at ART initiation have not increased over the past decade [16]. When restricted analysis to studies conducted in South Africa, the meta-analysis found a statistically significant increase in CD4 counts at presentation but no change at ART initiation. To reduce the prevalence of AHD among ART-naïve patients in South Africa, continued attention to programmatic strategies facilitating earlier HIV testing and linkage to care are needed, including referral to primary health care and post-discharge support for adherence and retention, as well as implementing WHO-recommended universal ART eligibility (“treat-all policy”) guidelines for PLHIV [30, 31]. Furthermore, community awareness of AHD needs to be increased, as it is still limited [31].

Our study also found that 58.6% of patients with AHD were ART-experienced, this is a newly alarming trend observed in countries with long-standing HIV treatment programs and even in high ART coverage settings. Among ART-experienced, progress to AIDS either because they either interrupt treatment or because of treatment failure. The decline in the number of patients presenting for the first time with AHD could be offset by an increasing number of patients on long-term ART who have interrupted, stopped, or failed therapy [32‐34]. Only two studies contributed to this outcome, therefore we cannot draw a firm conclusion using this finding. This evidence is consistent with findings from a multi-country cohort study conducted in 2 SSA countries evaluating characteristics and hospital outcomes of HIV-infected patients which showed that 65.1% of patients were admitted with AHD. Among inpatients with AHD, 71.7% were ART-experienced at admission with a median of 55.9 months on treatment (2). This means ongoing transmission due to ongoing viraemia among those who are out of care [36]. This supports a need for strategies to reduce gaps in HIV care and encourage return (re-engagement interventions such as welcome back services) before marked deterioration [37, 38], Given that more and more people who are in HIV programs are treatment experienced. This work supports the current reconceptualized non-linear HIV care cascade, which accounts for the cycling of patients in and out of care and the mobile populations served by ART programs in many SSA countries [12].

Furthermore, our subgroup analyses revealed a statistically significant difference in the prevalence of AHD among ART-naive patients between studies that used the WHO clinical staging system and those that used CD4 count criteria to identify patients with AHD, with 58.78% and 40.52% respectively. The higher prevalence among studies using the WHO clinical staging system should be interpreted with caution due to the poor accuracy associated with the WHO clinical staging system [39]. A systematic review and meta-analysis assessing the accuracy of WHO clinical staging system compared with CD4 count found that the WHO clinical stage 3 or 4 had a sensitivity of 60% and specificity of 73% for a CD4 threshold of ≤ 200 cells/mm3, and that sensitivity and specificity decreased as the CD4 count threshold increased [40]. Therefore, relying on the WHO clinical staging system alone risks missing a substantial number of PLHIV with severe immune suppression. Additionally, a multi-country study from Kenya, Malawi, Uganda, and Zimbabwe found that almost half of the people with CD4 count < 100 cells/mm3 were classified as having WHO clinical stage 1 or 2 disease [41].

The role of the CD4 count in HIV management is changing, with a shift away from using CD4 count to decide when to start ART or monitor treatment efficacy. Nevertheless, CD4 count remains the best measurement of patients’ immune and clinical status, the risk of opportunistic infections, and diagnosis decision-making, particularly for patients with AHD [30, 42].

Hence, identifying people with AHD who are eligible for elements of a package of care requires CD4 cell count testing. In addition, determining the immune status of PLHIV whose treatment is failing by virological criteria can help in guiding clinical management decisions (providing a package of care to reduce mortality and morbidity among people with AHD, which package includes screening, treatment, and/or prophylaxis for major opportunistic infections, rapid ART initiation and intensified adherence support interventions) [43].

Furthermore, the shift from monitoring CD4 cell count to monitoring viral load among individuals on ART has contributed to the de-prioritisation of testing CD4 cell count. Monitoring of viral load and CD4 cell count complement one another and should not be placed in competition. Monitoring VL supports viral suppression and triage into differentiated models of care, whereas CD4 cell count identifies people for whom an additional intervention is most urgently needed [44]. For these reasons, WHO recommends that HIV programs should retain the capacity to perform CD4 cell count at baseline and in the case of virological failure.

Despite progress in making ART available to millions of PLHIV in South Africa, the burden AHD among ART-naive and ART-experienced patients remains consistently high. This is despite South Africa having a well-established ART program with excellent population-level coverage. Our findings highlight that CD4 cell count measurement remains an essential component of HIV care and is critically important in identifying individuals with low CD4 counts and enabling differentiated care focused on reducing the high morbidity and mortality in this vulnerable patient group. These results emphasize the importance of expanding HIV testing and timely linkage to care, to promote early ART initiation. There is a need to programmatically improve and decentralize screening for AHD at primary health clinics.