Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2021 Jun 24.
Published in final edited form as: Trop Med Int Health. 2014 Mar 3;19(5):490–498. doi: 10.1111/tmi.12285

Impact of choice of NRTI in first-line antiretroviral therapy: a cohort analysis of stavudine vs. tenofovir

Alana T Brennan 1,2, Kate Shearer 2, Mhairi Maskew 2, Lawrence Long 2, Ian Sanne 1,2,3,4, Matthew P Fox 1,2,5
PMCID: PMC8224497  NIHMSID: NIHMS1712989  PMID: 24589363

Abstract

OBJECTIVE

In April 2010, South Africa replaced stavudine with tenofovir in first-line antiretroviral therapy (ART) despite tenofovir’s higher cost. We examined treatment outcomes over 24 months amongst patients initiated on tenofovir-based vs. stavudine-based first-line regimens.

METHODS

Prospective cohort analysis of 3940 patients newly initiating either stavudine-based (April 2009 to March 2010) or tenofovir-based (April 2010 to March 2011) ART in Johannesburg, South Africa. Cox proportional hazards models and Fine and Gray’s competing risk regression accounting for death were used to model mortality and loss to follow-up, respectively. Linear and log-binomial regression were used to evaluate associations with immunologic response and unsuppressed virus (≥400 copies/ml), respectively.

RESULTS

About 1878 patients prescribed tenofovir and 2062 patients prescribed stavudine were included. One hundred and sixty-six (8.8%) tenofovir and 244 (11.8%) stavudine patients died. Three hundred and fifty (18.6%) tenofovir and 379 (18.4%) stavudine patients were lost to follow-up over 24 months on ART. Adjusted regression models showed tenofovir and stavudine were comparable regarding death, loss to follow-up, immunologic response and virologic status.

CONCLUSIONS

We found no difference in mortality, loss to follow-up, immunological and virologic outcomes over the first 24-months on ART associated with tenofovir compared with stavudine.

Keywords: antiretroviral therapy, tenofovir, stavudine, resource-limited setting, drug toxicities, treatment outcomes

Introduction

Over the past decade, HIV treatment in resource-limited settings has been scaled up through use of standardised regimens (World Health Organization 2007, 2010) with nucleoside reverse transcriptase inhibitor (NRTI) backbones. In South Africa, prior to April 2010, the most frequently used NRTI, besides lamivudine, was stavudine due to its limited requirements for laboratory monitoring and its relatively low cost (Gilks et al. 2006; World Health Organization 2007, 2010). Stavudine is an effective drug (Gallant et al. 2004; Ferradini et al. 2007), but is also associated with severe side effects, such as dyslipidemias, lipoatrophy and mitochondrial toxicities, notably peripheral neuropathy and lactic acidosis (McComsey & Lonergan 2004; Murphy et al. 2007; Subbaraman et al. 2007; Domingos et al. 2009; van Griensven et al. 2010). As a result of its poor side effect profile, in 2007, the World Health Organization (WHO) recommended reducing stavudine dosage from 40 mg to 30 mg for all adults on antiretroviral therapy (ART) (World Health Organization 2007). In 2009, WHO recommended discontinuing stavudine for initial HIV treatment (World Health Organization 2010). In 2010, South Africa followed this advice and substituted stavudine with tenofovir in first-line therapy for all new ART initiates, for treatment-experienced patients that were not tolerating stavudine well and for those at high risk of toxicity (i.e. older females, patients with low iron levels and those with high body mass index) (National Department of Health 2010).

In 2001, tenofovir was approved by the U.S. Food and Drug Administration (2001) for treatment of HIV infection, and in 2002, it went on to replace stavudine in resource-rich settings (Nelson et al. 2007). However, in resource-limited settings, the transition from stavudine to tenofovir in standard first-line ART has been slow due to cost (Clinton Foundation 2009; Medecins Sans Frontieres 2008) and difficulty in management of toxicities, predominantly renal insufficiency (Scherzer et al. 2012). Renal function for patients on tenofovir has to be measured (via creatinine clearance levels) at treatment initiation and routinely over the course of treatment in accordance with WHO (2010) guidelines. By 2011, all but three reporting low- and middle-income countries had adopted the WHO guidelines and started the transition away from stavudine use in first-line ART for either tenofovir or zidovudine (Global HIV/AIDS response 2011). Whilst it is a substantial accomplishment, the phase-out of stavudine remains slow in the majority of these countries, largely due to insufficient financial resources needed for procurement of antiretrovirals (ARV) (Global HIV/AIDS response 2011). Although the price of tenofovir fell in 2009 due to the help of the Clinton Foundation HIV ceiling price negotiations with manufacturers, the cost was still four times that ($99 USD vs. $25 USD) of stavudine (Clinton Foundation 2009). For example, in 2009, Malawi and Zimbabwe reported 95% of their patients on a stavudine-based regimen and 93% in 2010, only a 2% reduction in stavudine use since adoption of the WHO guidelines (Global HIV/AIDS response 2011). The lack of progress in these low-income countries was due to financial constraints (Global HIV/AIDS response 2011). In South Africa, the number of patients on stavudine has dropped substantially, from 67% in 2006 to 30% in 2013; however, an estimated 715 300 adults are still on a stavudine-based regimen (either newly initiated on since the change in the guidelines or treatment experienced) (Meyer-Rath et al. 2013). It is also important to note that 31 low- to middle-income countries do not report to WHO and are likely to still be using stavudine in first-line therapy (Global HIV/AIDS response 2011).

The switch to tenofovir in South Africa was made with the hope of reducing toxicities and increasing regimen durability. Currently, we know little about how national programmes can switch safely and effectively from stavudine to tenofovir. The few studies done had mixed results regarding mortality, ranging from no effect to a large increase in mortality when comparing stavudine to tenofovir (Chi et al. 2010; Bygrave et al. 2011; Velen et al. 2013). While tenofovir has been shown to be highly effective in clinical trials (Gallant et al. 2004), because conditions of routine care differ strongly from trial conditions, the realised gains may be far smaller than anticipated. We set out to help build the evidence base on the move away from stavudine to tenofovir by evaluating the transition in a setting where routine viral load monitoring is standard of care. We compared 24-month treatment outcomes amongst patients initiated onto ART containing either tenofovir or stavudine (30 mg) in a large government HIV clinic in Johannesburg, South Africa.

Methods

Cohort

Themba Lethu Clinic is a large government clinic in Johannesburg, South Africa. The clinic started in 2004 as a public sector HIV treatment rollout site run by the South African Department of Health as part of its development of accredited Comprehensive Care, Management and Treatment sites. The clinic is located in an ambulatory care wing at Helen Joseph Hospital, a large urban secondary-level public-sector teaching hospital. Since the clinic’s inception in 2004, it has enrolled over 35 500 HIV-positive patients in care, 26 500 of whom have initiated treatment for HIV (Fox et al. 2013). Despite the large number of patients at the clinic, it functions with a small clinical staff; eight full time doctors, nine nurses, three pharmacists and a team of five administrative and eight data entry staff.

Care at Themba Lethu Clinic is provided according to national treatment guidelines (National Department of Health 2004, 2010, 2011). Details of the Themba Lethu Clinic have been described previously (Fox et al. 2013). Briefly, all data, including demographic, clinical conditions, laboratory test results and medications (ARV and non-ARV related) are entered into TherapyEdge- HIVtm in real-time by either a clinician or a data entry clerk at the clinic. Routine laboratory tests (CD4 count, full blood counts, liver function tests and renal function tests), with the exception of viral loads, are conducted at the time of ART initiation. Additional testing (tuberculosis microscopy and culture results, lactate levels, glucose and lipid profiles) are performed when clinically indicated (National Department of Health 2004, 2010). Prior to 2010, CD4 and viral load (concurrently with full blood counts, liver function tests and renal function tests) measurements were repeated every 6 months (National Department of Health 2004), but changes to the April 2010 guidelines called for these laboratories to be measured at 6 and 12 months post-treatment initiation and then yearly thereafter (National Department of Health 2010). Although the schedule for clinic visits varies depending on the regimen, typically, patients on treatment are seen for medical follow-up visits and at months 1, 3, 6, 7 or 8, 12 and 13 then 6-monthly thereafter. In regard to ARV pickups, patients have a pharmacy visit at months 1, 3, 6, 7 or 8, 10, 11 and 13 and then every 3 months thereafter (National Department of Health 2004, 2010).

Use of Themba Lethu Clinic data was approved by the Human Research Ethics Committee of the University of the Witwatersrand. Approval for analysis of de-identified data was granted by the Institutional Review Board of Boston University.

Study population

We performed a cohort analysis of data collected prospectively as part of routine care at the Themba Lethu clinic. All ART-naïve, non-pregnant, HIV-positive adult patients newly initiated on either a stavudine-based (between April 2009 and March 2010) or a tenofovir-based (between April 2010 and March 2011) regimen were included. These two 12-month time periods were chosen to minimise confounding factors due to time trends between patient populations (stavudine era and tenofovir era) before and after the change in the guidelines. Prior to April 2010, most patients initiated on a regimen containing stavudine also received lamivudine and efavirenz (National Department of Health 2004). Because in October 2007, a universal low dose stavudine policy was introduced, we limited our analysis to stavudine patients on 30 mg stavudine. After April 2010, new initiates were predominately given tenofovir plus lamivudine and efavirenz (National Department of Health 2010). Pregnant women were excluded as they are initiated on ART at higher CD4 counts, are on different ART regimens and have variable CD4 counts (due to the hemodilution effect of pregnancy) compared to the general population (Maini et al. 1996).

Study variables

We compared all-cause mortality, loss to follow-up, mean increase in CD4 count from ART initiation and having a detectable viral load (≥400 copies/ml) at 24-months postinitiation for tenofovir vs. stavudine. For patients with a national ID, mortality was ascertained via linkage with the National Vital Registration system in November 2012 (Timaeus et al. 2002; Statistics South Africa 2005). Because there is a 6-month delay in updating the registry, all patients lost to follow-up in our analysis (n = 739) were checked against the death registry again in July 2013. Loss to follow-up was defined as at least 3 months late for last scheduled visit. For analyses of death and loss, person-time ended at the earliest of (i) substitution of NRTI within first-line ART; (ii) switch to second-line treatment (iii) loss; (iv) death; (v) transfer; or (vi) completion of 24-months of follow-up. It is important to note that we do not include single-drug substitutions due to side effects or toxicity in this analysis, because we published on those outcomes previously in this population (Brennan et al. 2013).

Statistical analysis

We used a Cox proportional hazards model to evaluate the relationship between tenofovir and stavudine on 24-month all-cause mortality. Fine and Gray’s competing risks regression method (Fine & Gray 1999) was used to evaluate the relationship between tenofovir and stavudine on loss over 24-months, accounting for death as a competing risk. We used linear regression to evaluate predictors associated with mean CD4 change from ART initiation and log-binomial regression to evaluate risk factors associated with a viral load ≥400 copies/ml at 24-months on ART.

We used multiple imputation to account for missing data at ART initiation including CD4 count (2.6%), haemoglobin (6%) and body mass index (11.3%) (http://support.sas.com/rnd/app/papers/miv802.pdf). All models were fit using 25 imputed datasets, estimated coefficients were combined (http://support.sas.com/rnd/app/papers/mianalyzev802.pdf) and appropriate standard errors calculated (Rubin 1987).

Results

The 3940 patients included in the analysis (1878 tenofovir, 2062 stavudine) had a median CD4 count of 114 cells/mm3 (IQR: 47–185) at ART initiation and were on ART for a median of 24 months (IQR: 10.6–24.0; Table 1). Compared with stavudine patients, those on tenofovir had similar demographic and clinical characteristics with the exception of a slightly higher median CD4 cell count (122 vs. 109 cells/mm3) at treatment initiation. However, we would not consider the estimates to be significantly different from each other as the confidence intervals have substantial overlap — stavudine 109 cells/mm3; 95% CI: 42–179 cells/mm3 vs. tenofovir 122 cells/mm3; 95% CI: 51–191 cells/mm3.

Table 1.

Clinical and demographic characteristics at ART initiation and outcomes at 24-months of follow-up stratified by stavudine and tenofovir status at Themba Lethu Clinic, Johannesburg, South Africa (n = 3940)

Nucleotide reverse transcriptase inhibitor
Characteristics Stavudine (n = 2062) n (%) Tenofovir (n = 1878) n (%) Total (n = 3940) n (%)
Gender
 Female 1170 (56.7) 1164 (62.0) 2334 (59.2)
 Male 892 (43.3) 714 (38.0) 1606 (40.8)
Age (years)
 18–24.9 73 (3.5) 86 (4.6) 159 (4.0)
 25–29.9 289 (14.0) 262 (14.0) 551 (14.0)
 30–39.9 887 (43.0) 779 (41.5) 1666 (42.3)
 40–49.9 588 (28.5) 540 (28.8) 1128 (28.6)
≥50 225 (11.0) 211 (11.1) 436 (11.1)
Age at ART initiation, median (IQR) 37.5 (32.2–44.5) 37.7 (31.5–43.9) 37.6 (31.8–44.2)
CD4 at ART initiation
 0–49 cells/mm3 563 (27.3) 465 (24.8) 1028 (26.1)
 50–99 cells/mm3 398 (19.3) 328 (17.5) 726 (18.4)
 100–199 cells/mm3 743 (36.0) 685 (36.5) 1428 (36.2)
≥200 cells/mm3 358 (17.4) 400 (21.2) 758 (19.2)
CD4 at ART initiation (cells/mm3), median (IQR) 109 (42–179) 122 (51–191) 114 (47–185)
WHO stage at ART initiation
 I/II 1351 (65.5) 1254 (66.8) 2605 (66.1)
 III 558 (27.1) 500 (26.6) 1058 (26.9)
 IV 153 (7.4) 124 (6.6) 277 (7.0)
Haemoglobin at ART initiation
 ≥10 g/dl 1503 (72.9) 1457 (77.6) 2960 (75.1)
 <10 g/dl 559 (27.1) 421 (22.4) 980 (24.9)
 Median (IQR) 11.5 (9.8–12.9) 11.6 (10.1–12.9) 11.5 (10.0–12.9)
Body mass index at ART initiation
 ≥18.5 kg/m2 1610 (78.1) 1552 (82.6) 3162 (80.3)
 <18.5 kg/m2 452 (21.9) 326 (17.4) 778 (19.8)
 Median (IQR) 21.2 (18.8–24.6) 22.3 (19.4–25.8) 21.7 (19.1–25.2)
Tuberculosis at ART initiation
 No 1770 (85.8) 1641 (87.4) 3411 (86.6)
 Yes 292 (14.2) 237 (12.6) 529 (13.4)
Non-nucleotide reverse transcriptase inhibitor
 Nevirapine 142 (6.9) 104 (5.5) 246 (6.2)
 Efavirenz 1920 (93.1) 1774 (94.5) 3694 (93.8)
Time on ART (months), median (IQR) 24.0 (10.5–24.0) 24.0 (10.7–24.0) 24.0 (10.6–24.0)
Vital status over 24-months of follow-up, n (%)
 Death 244 (11.8) 166 (8.8) 410 (10.4)
 Loss to follow-up 379 (18.4) 350 (18.6) 729 (18.5)
 Transfers 138 (6.7) 156 (8.3) 294 (7.5)
 Alive 1301 (63.1) 1206 (64.2) 2507 (63.6)
Virologic and immunological outcomes over 24-months of follow-up
 Detectable viral load (≥400 copies/ml), n (%) 486 (23.6) 366 (19.5) 852 (21.6)
 CD4 change from ART initiation, median (IQR) 218 (128.0–334.1) 223 (133.1–325.0) 220.9 (130.1–331.0)
Open in a new tab

ART, antiretroviral therapy; WHO, World Health Organization.

Even though Themba Lethu clinic made the shift from stavudine-to tenofovir-based first-line ART in 2010, patients were still initiated onto stavudine after the 2010 change in the guidelines (Figure 1). We chose to exclude patients initiated onto stavudine after April 2010 because they could potentially represent a sicker patient population. According to the 2010 guidelines, patients that have contraindications to tenofovir (renal disease or the use of other nephrotoxic drugs, such as aminoglycosides) should be initiated on a zidovudine-based regimen, while those that have contraindications to tenofovir and zidovudine (renal disease and anaemia or the use of other nephrotoxic drugs) should be initiated on a stavudine-based regimen (National Department of Health 2010). It is also important to note that there have been documented ARV stock-outs and shortages throughout South Africa since April 2010 (http://stockouts.org/uploads/3/3/1/1/3311088/stock_outs_a_national_crisis.pdf); however, there was no documented stock out of tenofovir at Themba Lethu between April 2010 and March 2011. Additionally, although not depicted in the graph, 2 (0.09%) patients were initiated on abacavir and 52 (2.3%) patients initiated zidovudine between April 2009 to March 2010, while 10 (0.41%) initiated abacavir and 42 (1.7%) initiated zidovudine between April 2010 to March 2011.

Figure 1.

Figure 1

Open in a new tab

Proportion of patients initiated onto tenofovir and stavudine quarterly at Themba Lethu Clinic since the roll-out of ART in South Africa in 2004.

Death and loss to follow-up at 24-months

A total of 166 (8.8%) tenofovir and 244 (11.8%) stavudine patients died. Among those who died median time to death was 2.4 months (IQR: 0.8–7.6). The crude mortality rate was lower among tenofovir (5.8/100-person years (pys); 95%CI: 4.9–6.7) than stavudine patients (7.8/100-pys; 95%CI: 6.9–8.8; Table 2). However, Cox regression adjusted for age and gender and clinical characteristics at ART initiation (CD4 count, hemoglobin, body mass index, tuberculosis, WHO stage) showed no difference in the hazard of death for patients on tenofovir compared to stavudine (aHR 0.9; 95%CI: 0.7–1.0; Table 2). Loss to follow-up was also similar with 350 (18.6%) tenofovir and 379 (18.4%) stavudine patients being lost in a median of 8.3 months (IQR: 4.8–14.5). Crude loss rates were comparable between the two drugs (tenofovir-12.1/100-pys; 95%CI: 10.9–13.5 vs. stavudine-12.1/100-pys; 95%CI: 10.9–13.4). Competing risks regression for loss to follow-up (Table 2) accounting for death as a competing risk and adjusted for demographic and clinical characteristics at ART initiation showed no difference between patients on tenofovir vs. stavudine (aHR 1.1; 95%CI: 0.9–1.3; Table 2).

Table 2.

Crude and adjusted predictors of death, loss to follow-up, mean CD4 change from ART initiation and viral load status in the first 24-months on ART in Johannesburg, South Africa (n = 3940)

Mortality
 Loss to follow-up
 Viral load detectable
 Mean CD4 change from ART initiation
Variable No.
events
(%)		 Crude(95% CI) Adjusted HR*
(95% CI)		 No.
events
(%)		 Crude(95% CI) Adjusted HR*
(95% CI)		 No.
events
(%)		 Crude(95% CI) Adjusted RR*
(95% CI)		 Crude(95% CI) Adjusted*
(95% CI)
NRTI
 Stavudine 244 (11.8) Reference Reference 379 (18.4) Reference Reference 486 (23.6) Reference Reference Reference Reference
 Tenofovir 166 (8.8) 0.7 (0.6–0.9) 0.9 (0.7–1.0) 350 (18.6) 1.0 (0.9–1.2) 1.1 (0.9–1.3) 366 (19.5) 0.9 (0.8–1.0) 0.9 (0.7–1.0) −10.3 (−20.6, −0.01) −5.6 (−17.7, 6.4)
NNRTI
 Efavirenz 390 (10.6) Reference Reference 682 (18.5) Reference Reference   52 (21.1) Reference Reference Reference Reference
 Nevirapine 20 (8.1) 0.8 (0.5–1.2) 1.1 (0.7–1.7)   47 (19.1) 1.0 (0.7–1.3) 1.0 (0.7–1.4) 800 (21.7) 1.2 (1.0–1.5) 1.1 (0.8–1.5) −3.8 (−25.0, 17.3) −7.0 (−32.6, 18.6)
Gender
 Female 204 (8.7) Reference Reference 368 (15.8) Reference Reference 466 (20.0) Reference Reference Reference Reference
 Male 206 (12.8) 1.5 (1.2–1.9) 1.3 (1.1–1.6) 361 (22.5) 1.5 (1.3–1.7) 1.6 (1.4–1.9) 386 (24.0) 1.1 (1.0–1.3) 1.2 (1.0–1.4) −56.9 (−67.0, −46.8) −49.0 (−63.0, −35.1)
Age (years)
 18–24.9 13 (8.2) 0.9 (0.5–1.5) 0.9 (0.5–1.6)   43 (27.0) 1.6 (1.2–2.2) 1.9 (0.3–2.6)   36 (22.6) 1.2 (0.9–1.6) 1.2 (0.8–1.6) 48.3 (15.9, 80.8) 41.6 (7.7, 75.5)
 25–29.9 41 (7.4) 0.8 (0.5–1.1) 0.8 (0.6–1.1) 131 (23.8) 1.3 (1.1–1.7) 1.5 (1.2–1.8) 128 (23.2) 1.1 (0.9–1.4) 1.1 (0.9–1.4) 26.1 (9.1, 43.1) 16.2 (−2.3, 34.7)
 30–39.9 167 (10.0) Reference Reference 314 (18.9) Reference Reference 360 (21.6) Reference Reference Reference Reference
 40–49.9 118 (10.5) 1.1 (0.8–1.3) 1.1 (0.9–1.4) 188 (16.7) 0.9 (0.7–1.1) 0.9 (0.8–1.1) 234 (20.7) 1.0 (0.9–1.2) 1.0 (0.8–1.2) −21.0 (−32.9, −9.1) −20.8 (−36.0, −5.6)
≥50 71 (16.3) 1.7 (1.3–2.2) 2.0 (1.5–2.7)   53 (12.2) 0.6 (0.5–0.9) 0.7 (0.5–0.9)   94 (21.5) 1.1 (0.9–1.3) 1.0 (0.7–1.3) −36.1 (−52.5, −19.7) −31.5 (−51.1, −11.9)
CD4 count (cells/mm3)
≥200 39 (5.2) Reference Reference 120 (15.8) Reference Reference 134 (17.7) Reference Reference Reference Reference
 100–199 90 (6.3) 1.2 (0.8–1.7) 1.1 (0.8–1.7) 259 (18.1) 1.1 (0.9–1.4) 1.1 (0.9–1.4) 303 (21.2) 1.1 (0.9–1.3) 1.1 (0.9–1.4) 28.1 (12.7, 43.6) 24.7 (5.6, 43.3)
 50–99 89 (12.3) 2.4 (1.7–3.5) 2.0 (1.4–2.9) 140 (19.3) 1.2 (0.9–1.5) 1.1 (0.9–1.4) 173 (23.8) 1.2 (0.9–1.4) 1.2 (0.9–1.5) 35.1 (18.1, 52.1) 25.5 (4.1, 46.8)
<50 192 (18.7 3.9 (2.8–5.5) 2.8 (2.0–4.1) 210 (20.4) 1.4 (1.1–1.7) 1.1 (0.9–1.5) 242 (28.4) 1.3 (1.1–1.6) 1.2 (1.0–1.6) 41.1 (25.6, 56.7) 29.7 (7.3, 52.1)
Tuberculosis
 No 313 (9.2) Reference Reference 623 (18.3) Reference Reference 713 (20.9) Reference Reference Reference Reference
 Yes 97 (18.3) 2.1 (1.7–2.7) 1.3 (1.0–1.7) 106 (14.5) 1.1 (0.9–1.4) 0.8 (0.6–1.0) 139 (26.3) 1.1 (1.0–1.3) 0.9 (0.7–1.2) 39.7 (24.0, 55.5) 32.0 (8.3, 55.7)
WHO stage
 I/II 210 (8.1) Reference Reference 440 (16.9) Reference Reference 520 (20.0) Reference Reference Reference Reference
 III/IV 200 (15.0) 2.0 (1.6–2.4) 1.1 (0.8–1.4) 289 (21.7) 1.3 (1.2–1.6) 1.2 (1.0–1.5) 332 (24.9) 1.1 (1.0–1.3) 1.1 (0.9–1.3) 24.3 (13.3, 35.4) 8.2 (−7.3, 23.7)
Hb (μg/dl)
 ≥10.0 222 (7.5) Reference Reference 507 (17.1) Reference Reference 609 (20.6) Reference Reference Reference Reference
 <10.0 188 (19.2) 2.8 (2.3–3.4) 2.1 (1.7–2.6) 222 (22.6) 1.4 (1.2–1.7) 1.3 (1.1–1.6) 243 (24.8) 1.2 (1.0–1.3) 1.2 (1.0–1.4) 34.1 (21.5, 46.6) 17.0 (0.5, 33.6)
BMI (kg/m2)
 ≥18.5 257 (8.1) Reference Reference 541 (17.1) Reference Reference 665 (21.0) Reference Reference Reference Reference
 <18.5 153 (19.7) 2.6 (2.1–3.1) 1.7 (1.3–2.1) 188 (24.2) 1.4 (1.1–1.7) 1.2 (1.0–1.5) 187 (24.0) 1.2 (1.0–1.4) 1.0 (0.8–1.3) 3.8 (−10.0, 17.5) −1.0 (−17.6, 15.6)
Open in a new tab

Hb, haemoglobin; BMI, body mass index; WHO, World Health Organization; NRTI, Nucleotide/Nucleoside Reverse Transcriptase Inhibitor; NNRTI, Non-Nucleoside Reverse Transcriptase Inhibitor.

*

All prediction equations included log age at initiation of treatment, gender, square root of CD4 count at ART initiation and at 24-months, log of 24-month viral load measurement, haemoglobin at ART initiation (continuous), body mass index at ART initiation (continuous), WHO stage (I/II, III and IV) and tuberculosis at ART initiation. Indicator variables for death, loss to follow-up were also added to the equations but were not imputed.

Immunologic and virologic response at 24-months

The median increase in CD4 count over 24 months for those on tenofovir was 223 cells/mm3 (IQR: 133.1–325.0) and 218 cells/mm3 (IQR: 128.0–334.1) for patients on stavudine (Table 1). Analyses adjusted for demographic and clinical characteristics at ART initiation comparing tenofovir to stavudine showed no difference in mean CD4 change from ART initiation (−5.6 cells/mm3; 95%CI: −17.7–6.4) (Table 2). A total of 366 (19.5%) patients of 1878 on tenofovir and 486 (23.6%) of 2062 on stavudine had a detectable viral load (≥400 copies/ml) at 24 months, but adjusted log-binomial regression showed no difference between the two drugs (adjusted Risk Ratio (aRR): 0.9; 95%CI: 0.7–1.0) by the end of follow-up (Table 2).

Discussion

Our results support the notion that changing from stavudine-based to tenofovir-based ART regimens as part of the national treatment programme in South Africa was not associated with any increase in negative treatment outcomes. Our results are in agreement with previous studies comparing the two drugs. An observational study conducted in Zambia found the rates of mortality comparable between patients on stavudine vs. tenofovir (Chi et al. 2010). However, in two more recent observational studies in Lesotho (Bygrave et al. 2011) and South Africa (Velen et al. 2013), stavudine was associated with a 30–170% increase, respectively, in mortality compared with tenofovir. Our results are consistent with previous studies showing no difference in mortality associated with the substitution of tenofovir with stavudine in first-line ART. We also found no difference in risk of loss to follow-up, viral suppression or increase in CD4 count, consistent with previous studies (Gallant et al. 2004; Bygrave et al. 2011; Velen et al. 2013).

A main strength of our analysis is our large sample size coupled with robust mortality data generated through linkage with a highly sensitive national death registry (Timaeus et al. 2002; Statistics South Africa 2005). However, because only 56% of patients at Themba Lethu clinic have an identification number allowing linkage to the registry, we are likely underestimating mortality. Second, although loss from care was comparable between groups, overall loss was high (18%). Patients lost to follow-up were more likely to be male, of younger age and have poorer health status than those included. As a result, there is likely some selection bias as patients who leave care are more likely to stop treatment, increasing their risk of death (Fox et al. 2010; Druyts et al. 2013). Third, we recognise that there may be confounding between the stavudine and tenofovir groups due to the change in the patient characteristics (due to the aging epidemic in addition to the changes in the National ART guidelines) over time; however, restricting the population to 1 year prior and 1 year after the 2010 guidelines minimised confounding factors between groups (visible in Table 1). However, to confirm this, we re-ran the analysis restricting the population to 3 months prior and 3 months after the change to further minimise the potential effect of a time trend and results remained unchanged. Fourth, we recognise that due to our sample size, we may not be able to tease out smaller differences in outcomes when comparing tenofovir to stavudine. However, increasing the population outside of the period 12 months prior and 12 months after the change in the 2010 guidelines would increase the probability of confounding due to time trends between the groups potentially affecting the validity of the results. Fifth, while multiple imputation helps improve the validity of research, the validity depends on correct modelling and the assumption that our data is missing at random (Rubin 1987). Deviations from this could have led to unpredictable biases in our parameter estimates.

The global economic crisis has threatened the substantial gains made over the last decade in the scale-up of ART in resource-limited settings. Failure to maintain current financial commitments could lead to an increase in HIV-related morbidity and mortality and HIV transmission (UNAIDS 2009). With constrained budgets and the higher cost of tenofovir (Rosen et al. 2008), treatment programmes must make difficult decisions around how to maximise benefits. Further research is needed to continue to monitor the impact of the transition to tenofovir in other resource-limited settings.

Acknowledgements and disclaimer

We express our gratitude to the directors and staff of Themba Lethu Clinic and to Right to Care, the Non-Governmental Organization supporting the study site through a partnership with United States Agency for International Development. We also thank the Gauteng and National Department of Health for providing for the care of the patients at the Themba Lethu Clinic as part of the Comprehensive Care Management and Treatment plan. Most of all we thank the patients attending the clinic for their continued trust in the treatment provided at the clinic. Funding was provided by USAID under the terms of Cooperative Agreement 674-A-00-09-00018-00 to Boston University and Cooperative Agreement 674-A-12-00020 to Right to Care; INROADS USAID-674-A-12-00029 and the National Institute of Allergy and Infectious Diseases (NIAID) Award Number K01AI08309. This study is made possible by the generous support of the American people through the United States Agency for International Development (USAID) and the National Institutes of Health. The contents are the responsibility of the authors and do not necessarily reflect the views of USAID, NIAID, the United States Government, the Themba Lethu Clinic or Right to Care.

References

  1. The MI Procedure. http://support.sas.com/rnd/app/papers/miv802.pdf. [Google Scholar]
  2. The MIANALYZE Procedure. http://support.sas.com/rnd/app/papers/mianalyzev802.pdf. [Google Scholar]
  3. Brennan AT, Maskew M, Ive P et al. (2013) Increases in regimen durability associated with the introduction of tenofovir at a large public-sector clinic in Johannesburg, South Africa. Journal of the International AIDS Society 16, 18794. [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Bygrave H, Ford N, Van CG et al. (2011) Implementing a tenofovir-based first-line regimen in rural Lesotho: clinical outcomes and toxicities after two years. Journal of Acquired Immune Deficiency Syndromes 56, e75–e78. [DOI] [PubMed] [Google Scholar]
  5. Chi BH, Mwango A, Giganti M et al. (2010) Early clinical and programmatic outcomes with tenofovir-based antiretroviral therapy in Zambia. Journal of Acquired Immune Deficiency Syndromes 54, 63–70. [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Clinton Foundation (2009) Antiretroviral (ARV) Price List. http://www.givewell.org/files/DWDA%202009/Interventions/WJCF-ARVPrices-2009.pdf. [Google Scholar]
  7. Domingos H, Cunha RVD, Paniago AMM, Martins DM, Elkhoury EB & Souza ASD (2009) Metabolic effects associated to the highly active antiretroviral therapy (HAART) in AIDS patients. Brazilian Journal of Infectious Diseases 13, 130–136. [DOI] [PubMed] [Google Scholar]
  8. Druyts E, Dybul M, Kanters S et al. (2013) Male sex and the risk of mortality among individuals enrolled in antiretroviral therapy programs in Africa: a systematic review and metaanalysis. AIDS 27, 417–425. [DOI] [PubMed] [Google Scholar]
  9. Ferradini L, Laureillard D, Prak N et al. (2007) Positive outcomes of HAART at 24 months in HIV-infected patients in Cambodia. AIDS 15, 2293–2301. [DOI] [PubMed] [Google Scholar]
  10. Fine JP & Gray RJ (1999) A proportional hazards model for the sub distribution of a competing risk. JSTOR 94, 496–509. [Google Scholar]
  11. Fox MP, Brennan A, Maskew M, MacPhail P & Sanne I (2010) Using vital registration data to update mortality among patients lost to follow-up from ART programmes: evidence from the Themba Lethu Clinic, South Africa. Tropical Medicine and International Health 15, 405–413. [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Fox MP, Maskew M, Macphail AP et al. (2013) Cohort profile: The Themba Lethu Clinical Cohort, Johannesburg, South Africa. International Journal of Epidemiology 42, 430–439. [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Gallant JE, Staszewski S, Pozniak AL et al. (2004) Efficacy and safety of Tenofovir DF vs stavudine in combination therapy in antiretroviral-naive patients: a 3-year randomized trial. JAMA 292, 191–201. [DOI] [PubMed] [Google Scholar]
  14. Gilks CF, Crowley S, Ekpini R et al. (2006) The WHO public-health approach to antiretroviral treatment against HIV in resource-limited settings. Lancet 5, 505–510. [DOI] [PubMed] [Google Scholar]
  15. Global HIV/AIDS response (2011) Epidemic update and health sector progress towards Universal Access. Progress Report 2011. [Google Scholar]
  16. van Griensven J, Zachariah R, Rasschaert F, Mugabo J, Atté EF & Reid T (2010) Stavudine- and nevirapine-related drug toxicity while on generic fixed-dose antiretroviral treatment: incidence, timing and risk factors in a three-year cohort in Kigali, Rwanda. Transactions of the Royal Society of Tropical Medicine and Hygiene 104, 148–153. [DOI] [PubMed] [Google Scholar]
  17. Maini MK, Cilson RJC, Chavda N et al. (1996) Reference ranges and sources of variability of CD4 counts in HIV seronegative women and men. Genitourinary Medicine 72, 27–31. [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. McComsey G & Lonergan JT (2004) Mitochondrial dysfunction: patient monitoring and toxicity management. Journal of Acquired Immune Deficiency Syndromes 37 (Suppl. 1), S30–S35. [DOI] [PubMed] [Google Scholar]
  19. Medecins Sans Frontieres (2008) Untangling the web of antiretroviral price reductions, 11th edn. [March 2, 2009]. http://www.msfaccess.org. [Google Scholar]
  20. Meyer-Rath G, Brennan A, Long L, Rosen S & Fox MP (2013) National ART Cost Model South Africa. Health Economics and Epidemiology Research Office, Johannesburg, South Africa. [Google Scholar]
  21. Murphy R, Sunpath H, Kuritzkes DR, Venter F & Gandhi RT (2007) Antiretroviral therapy-associated toxicities in the resource-poor world: the challenge of a limited formulary. Journal of Infectious Diseases 196 (Suppl. 3), S449–S456. [DOI] [PubMed] [Google Scholar]
  22. National Department of Health (2004) National Antiretroviral Treatment Guidelines. Jacana Publishers, Johannesburg, Republic of South Africa. [Google Scholar]
  23. National Department of Health (2010) The South African Antiretroviral Treatment Guidelines, Republic of South Africa [http://www.doh.gov.za/docs/factsheets/guidelines/art.pdf]. [Google Scholar]
  24. National Department of Health (2011) Circular on new criteria for initiating adults on ART at CD4 count of 350 cells/μl and below, National Department of Health, Johannesburg, Republic of South Africa. [Google Scholar]
  25. Nelson MR, Katlama C, Montaner JS et al. (2007) The safety of tenofovir disoproxil fumarate for the treatment of HIV infection in adults: the first 4 years. AIDS 21, 1273–1281. [DOI] [PubMed] [Google Scholar]
  26. Rosen S, Long L, Fox M & Sanne I (2008) Cost and cost-effectiveness of switching from stavudine to tenofovir in first-line antiretroviral regimens in South Africa. Journal of Acquired Immune Deficiency Syndromes 48, 334–344. [DOI] [PubMed] [Google Scholar]
  27. Rubin DB (1987) Multiple imputation for Nonresponse in Surveys. John Wiley & Sons, New York. [Google Scholar]
  28. Scherzer R, Estrella M, Li Y et al. (2012) Association of tenofovir exposure with kidney disease risk in HIV infection. AIDS 26, 867–875. [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Southern African HIV Clinicians Society (SAHIVSOC). Stock Outs in South Africa — A National Crisis. http://stockouts.org/uploads/3/3/1/1/3311088/stock_outs_a_national_crisis.pdf [Google Scholar]
  30. Statistics South Africa (2005) Mortality and causes of death in South Africa, 1997–2003. Findings from death notification. P0309.3. [Google Scholar]
  31. Subbaraman R, Chaguturu SK, Mayer KH, Flanigan TP, Kumarasamy N (2007) Adverse effects of highly active antiretroviral therapy in developing countries. Clinical Infectious Diseases 45, 1093–1101. [DOI] [PubMed] [Google Scholar]
  32. Timaeus I, Dorrington R, Bradshaw D & Nannan N (2002) Mortality trends in South Africa 1985–2000: From apartheid to AIDS. South African Medical Research Council, Cape Town. [Google Scholar]
  33. UNAIDS (2009) The Global Economic Crisis and HIV Prevention and Treatment Programmes: Vulnerabilities and Impact. UNAIDS, Geneva, Switzerland. [Google Scholar]
  34. U.S. Food and Drug Administration (2001) Approval letter VI-READ™. http://www.accessdata.fda.gov/drugsatfda_docs/app-letter/2001/21356ltr.pdf [Google Scholar]
  35. Velen K, Lewis JJ, Charalambous S, Grant AD, Churchyard GJ & Hoffmann CJ (2013) Comparison of tenofovir, Zidovudine, or Stavudine as part of first-line antiretroviral therapy in a resource-limited-setting: a cohort study. PLoS One 8, e64459. [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. World Health Organization; (2007) Addendum to 2006 WHO Guidelines on Antiretroviral Therapy for HIV Infection in Adults and Adolescents. World Health Organization, Geneva. [Google Scholar]
  37. World Health Organization; (2010) Antiretroviral Therapy for HIV Infection in Adults and Adolescents: Recommendations for a Public Health Approach — 2010 Revision. World Health Organization, Geneva, Switzerland. [PubMed] [Google Scholar]

RESOURCES