By 2011, an estimated 780,000 people were living with human immunodeficiency virus (HIV) in China [1]. To respond to the HIV epidemic, the Chinese government started the National Free Antiretroviral Treatment Program (NFATP) in 2003 to provide free antiretroviral therapy (ART) to all eligible patients [2–5]. Overall HIV-associated mortality rate decreased from 39.3% in 2002 to 14.2% in 2009 due to the increasing coverage of NFATP [6]. Yet a substantial number of deaths still occur each year among people living with HIV/acquired immunodeficiency syndrome (AIDS), especially those not on ART and those with only a short treatment history [7]. HIV infection remains the leading cause of death among all notifiable infectious diseases in China [8]. Therefore, it is of great importance to explore additional interventions to further reduce HIV-related deaths.

Co-trimoxazole (CTX) prophylaxis has been associated with a reduction in mortality and morbidity among HIV-infected people both not on ART [9–15] and on ART [16–18]. A meta-analysis of nine articles found CTX prophylaxis reduced mortality by 58% among HIV-infected individuals older than 13 years who were receiving ART [16]. Furthermore, two studies in Africa showed CTX prophylaxis had the most significant effect during the first 3–6 months after ART initiation [17, 18]. In addition, from an economic perspective, CTX prophylaxis is cost-effective in averting mortality among patients on ART [19–22].

In 2005, the Chinese Ministry of Health issued national guidelines on CTX prophylaxis for preventing HIV-related opportunistic infections (OIs) [23]. There has been insufficient data in China to illustrate the actual coverage of CTX prophylaxis among people living with HIV/AIDS. However, a five-year national observational cohort study showed that less than a third (26.2%) of 69,943 HIV-infected individuals who were on ART in China received CTX either for prophylaxis or treatment of opportunistic infections during 2002 to 2009 [6]; these data suggest that implementation of the guidelines has been suboptimal. The aim of this study was to investigate the effect of CTX prophylaxis on survival among HIV-infected adults on ART by utilizing the data from the Henan Provincial Free Antiretroviral Treatment Program database.

A total of 2,529 eligible HIV-infected individuals were identified with 2,429 person-years (PY) of follow-up. Of these, 2,018 received CTX prophylaxis treatment at ART initiation and 511 did not. By 12 months after ART, 154 patients had died, 50% of whom died within 3 months (21 in the non-CTX group vs 56 in the CTX group). Over half the patients were male (54.2%; 58% were infected through plasma donation. Nearly half of the patients (44.2%) were aged between 35 and 54 years (Table 1). Overall, 5.6% had a history of active TB one year prior to ART initiation. A majority (78.9%) had a CD4 count <200 cells/µl at ART initiation. The CTX group had a higher percentage of patients with baseline CD4 count <200 cells/µl (80.8%) than the non-CTX group (71%). The median baseline CD4 count was 91 cells/µl and 132 cells/µl for the CTX group and the non-CTX group, respectively. There were no significant differences between the two groups in gender, age at ART initiation, proportion with TB history, and transmission route.

Overall mortality at 3-months after ART initiation was nearly doubled that at 12-months (12.4 per 100 PY vs 6.3 per 100 PY, p<0.01). Univariate analyses demonstrated that patients who were male, aged ≥55 at ART initiation, or had a history of TB prior to ART treatment had a higher probability of death. Patients with a baseline CD4 count < 50 cells/µl had a higher risk of death within one year of treatment. Transmission route was not related to patients’ survival. The Kaplan-Meier curve (Fig. 1) showed that patients receiving CTX had higher survival than those without prophylaxis, although this was not statistically significant (log rank =1.14, p = 0.285).

The 3-month mortality was 11.3 per 100 PY for patients on CTX at ART initiation and 16.8 per 100 PY for those not on CTX. The 12-month mortality was lower than the 3-month mortality, with 6.1 per 100 PY for the CTX group compared to 7.4 per 100 PY for the non-CTX-group. While these differences were not statistically significant before adjustment, after controlling for other variables, patients who took CTX at ART initiation were less likely to die at one year after ART treatment (AHR: 0.65, 95% CI: 0.44 – 0.95; p = 0.027) compared to those who did not take CTX (Table 2). This protective effect was stronger at 3 months after ART initiation (AHR: 0.53, 95% CI: 0.32 – 0.89; p = 0.017). Other factors significantly associated with increased risk of death at both at 3-months and 12-months after ART initiation were male sex (3-month mortality AHR 1.99, 95% CI 1.20 – 3.31, p=0.008; 12-month mortality AHR 1.56, 95% CI 1.11 – 2.20, p=0.011) and TB history in the past year (3-month mortality AHR 3.62, 95%CI 2.00 – 6.54, p<0.001; 12-month mortality AHR 2.59, 95% CI 1.63 – 4.14, p<0.001). Age ≥55 years at ART initiation was associated with an increased mortality only at 12-months after treatment (AHR 2.21, 95%CI 1.24 – 3.95, p=0.007). Higher CD4 cell counts at ART initiation was an independent beneficial factor to patients’ survival. Patients with baseline CD4 count of 200 cells/µL or more had a nearly 80% reduction in death both at 3-months and 12-months after ART initiation, compared with those with CD4 count < 50 cells/µL at baseline (3-month AHR 0.22, 95% CI 0.10 – 0.53, p<0.001; 12-month AHR 0.21, 95% CI 0.12 – 0.39, p<0.001). The protective effect of baseline CD4 count of 50–199 cells/µL was weaker but was still associated with a more than 50% decline both in 3-month and 12-month mortality compared with persons with a baseline CD4 count <50 cells/uL (3-month AHR 0.48 95%CI 0.29 – 0.78, p<0.001; 12-month AHR 0.36 95%CI 0.25 – 0.52, p<0.001).

We found CTX prophylaxis provided an approximately 35% reduction in mortality in the first year of ART among HIV-infected individuals in central China. This result is slightly lower than but consistent with what has been previously reported. Currently there is limited information evaluating the effect of CTX prophylaxis on mortality among HIV-infected patients receiving ART in China. Most publications on CTX prophylaxis in the Asia-Pacific region are from Cambodia and Thailand [28, 29]. Our study provides additional evidence in the region and highlights the benefit of promoting CTX prophylaxis in conjunction with ART to reduce mortality among HIV patients on ART.

Another important finding was that the protective effect of CTX was particularly strong in the first 3 months after ART initiation with a 47% reduction in mortality compared to those not on CTX. Many studies demonstrate that the first three months after ART initiation is a critical period. Lowrance et al. reported that a third to half of deaths occurred within three months of ART initiation [17]. Zhang et al. showed that mortality (22.6 per 100 PY) was greatest during the first three months of treatment [7]. In Zambia 71% of deaths after initiating ART occurred within 90 days of ART initiation, with an early mortality of 26.0 per 100 PY [30]. Similarly, our study found exactly half of the deaths occurred in the first three months. This high mortality shortly after ART initiation may be mainly attributed to untreated AIDS rather than the immune reconstitution inflammatory syndromes [7].

In our study, we demonstrated that CTX prophylaxis decreased 3-month mortality after ART by nearly half. This is consistent with previously reported data. Walker et al. showed mortality among patients on ART with daily CTX prophylaxis was more than halved at 3 months of treatment compared with those not on CTX [18]. The survival difference in the Lowrance paper was apparent as early as 40 to 45 days after ART initiation [17]. A substantial proportion of patients in our cohort had already progressed to advanced HIV with a median CD4 count <200 cells/µl at ART initiation. Even though ART was initiated, these patients were at a high risk for death. Their immune system was being under restoration during the initial ART period and was similar to those not on ART. The beneficial effect of CTX prophylaxis in reducing the risk of death among those patients who have only recently initiated ART was similar to those patients who were not on ART. Although the baseline median CD4 count was lower in the CTX group than that in the non-CTX group (100 cells/µl vs130 cell/µl), the risk of death associated with lower baseline immune function was compensated for by CTX prophylaxis.

However, our study has three limitations that need to be considered. First, the majority of participants (58.5%) contracted HIV through commercial plasma donation. This is a manifestation of the unusual nature of the local HIV epidemic in Henan Province. According to the recent Chinese HIV epidemic report unprotected sexual contact is becoming the major transmission route of HIV infections with 46.5% by heterosexual and 17.4% by homosexual contacts, respectively [1]. Therefore, the subjects in our study are not representative of the HIV-infected population in other parts of the country. It is possible that the survival of HIV infected people in Henan Province might be different from those who acquired the infection through sexual transmission since patients in this analysis who were infected by plasma donation in the early 1990s [24] survived without initiating ART for several years until 2003 when ART began to be available in China. However, in the multivariable analysis, we did not identify statistically significant differences in impact of CTX prophylaxis on 3-month or 12-month mortality among different transmission groups. The conclusion of our study provided strong evidence in China to support to scale up CTX prophylaxis together ART to reduce mortality among patients who are eligible for treatment.

Second, CTX adherence was not documented in our study. We used ART dosage as a proxy for overall medication adherence (Table 2), but this may not accurately reflect the adherence to CTX prophylaxis. Last, only information on CTX prophylaxis at ART initiation was collected. Initiating CTX prophylaxis at any time during ART treatment, given that the criteria for CTX prophylaxis are met, may bring benefits to patients. Some patients may have initiated CTX prophylaxis after ART initiation and been miscategorized as not on CTX in our study. This could lead to an underestimation of the effect of CTX prophylaxis.

Our study demonstrated that CTX significantly reduced the mortality of HIV patients within the first 12 months after ART initiation in Henan Province, China with the strongest impact during the first 3 months. Overall uptake of OI prophylaxis and treatment among eligible HIV-infected individuals in China is low. In 2006, the World Health Organization (WHO) recommended providing CTX prophylaxis to all HIV patients with CD4 count < 350 cells/µl or with WHO clinical stages 2, 3 or 4 disease [31]. However, China is still using the earlier criteria of CD4 count ≤200 cells/µl or WHO clinical 3 or 4 disease [23]. Although our study did not differentiate the effect of CTX prophylaxis in person with CD4 count <350 cells/µl vs <200 cells/µl, we would like to recommend updating the Chinese recommendations to be consistent with the current WHO recommended guidelines which suggest initiating CTX prophylaxis at CD4 counts <350 cells/µl. This recommendation not only targets prevention of OIs like Pneumocystis jiroveci pneumonia (PCP) and toxoplasmosis but also enteric bacterial infections and isosporiasis and malaria even though there is limited information on these infections among HIV/AIDS patients in China.

Given the consistent results in the literature, CTX prophylaxis, in addition to ART, should be promoted nationwide to all eligible HIV-infected individuals-- both adults and children. This may help to achieve the national goal of reducing HIV-associated mortality by 30% by 2015. More studies are needed to investigate factors that may hinder provision and uptake of CTX prophylaxis, as well as the optimal duration for CTX. To improve CTX prophylaxis monitoring, it is important to add a CTX prophylaxis component in the current web-based HIV case reporting system in China rather than only in the NFATP database.