Abstract
Background:
Early mortality among those still initiating ART with advanced stages of HIV infection in resource-limited settings remains high despite recommendations for universal HIV treatment. We investigated risk factors associated with early mortality in PLHIV starting ART at low CD4 levels in the Asia-Pacific.
Methods:
PLHIV enrolled in TAHOD who initiated ART with CD4 <100 cells/μL between 2003 – 2018 were included. Early mortality was defined as death within one year of ART initiation. PLHIV in follow-up for more than one year were censored at 12 months. Competing risk regression was used to analyse risk factors with those lost-to follow-up as a competing risk.
Results:
1813 PLHIV were included, of which 74% were males. With 73 (4%) deaths, the overall first year mortality rate was 4.27 per 100 person-years (/100PYS). 38 (52%) of deaths were AIDS-related, 10 (14%) IRIS-related, 13 (18%) non-AIDS-related and 12 (16%) unknown. Risk factors included having BMI <18.5 (SHR 2.91, 95% CI 1.60-5.32) compared to BMI 18.5 – 24.9, and alanine aminotransferase (ALT) ≥5 times its upper limit of normal (ULN) (SHR 6.14, 95% CI 1.62-23.20) compared to ALT <5 times its ULN. Higher CD4 (51-100 cells/μL: SHR 0.28, 95% CI 0.14-0.55; and >100 cells/μL: SHR 0.12, 95% CI 0.05-0.26) were associated with reduced hazard for mortality compared to CD4 ≤25 cells/μL.
Conclusion:
52% of early deaths were AIDS-related. Efforts to initiate ART at CD4 above 50 cell/μl is associated with improved short-term survival rates, even in those with late stages of HIV disease.
Keywords: Asia-Pacific, HIV, advanced disease, mortality
Background
People living with HIV (PLHIV) in resource-limited settings continue to initiate antiretroviral therapy (ART) with advanced stages of HIV infection. (1, 2) Despite implementation of WHO “Treat All” guidelines,(3) a study of PLHIV with CD4 <100 cells/μL in Kenya, Malawi, Uganda, and Zimbabwe found that approximately 20%–25% presented for care at late stages of HIV disease and of them 12% experienced mortality 48 weeks post-ART initiation.(1)
A meta-analysis looking at death within 12 months of ART initiation in PLHIV with CD4 <100 cells/μL in low- and middle-income countries (LMIC) found an overall mortality of 14%, with sub-Saharan Africa (17%) leading, followed by Asia (11%) and Americas (7%). (2) CD4 cell count, body mass index (BMI), haemoglobin level, sex, and age were found to be associated with early mortality. Another common cause of death in the population of late ART initiators are opportunistic infections (OIs). While studies in sub-Saharan Africa have shown tuberculosis (TB), cryptococcal meningitis (CM), and pneumocystis pneumonia (PCP/PJP) as prevalent OIs, (4) countries in South America did not find CM as a leading AIDS defining illness. (5, 6) In addition, poor ART adherence, intolerability to antiretroviral drugs, and very low CD4 cell counts with slow response to ART have been found as other contributing factors. (7, 8)
Multiple studies have concluded that there is insufficient information available compared to other regions on early mortality of HIV patients who initiated late ART in the Asia-Pacific region. (2, 8) Therefore, we investigated risk factors associated with early mortality among late initiators in the TREAT Asia HIV Observational Database (TAHOD) of the International epidemiologic Databases to Evaluate AIDS (IeDEA) Asia-Pacific.
Methods
TAHOD is a collaborative observational cohort study of over 9600 adult PLHIV of ages 18 and older involving 21 participating sites in 12 countries in the Asia-Pacific region, whose detailed methods are published elsewhere. (9) Ethics approvals for the study were obtained from ethics committees of each participating site, the data management and biostatistical center at the Kirby Institute (The University of New South Wales (UNSW) Human Research Ethics Committee), and the coordinating center at TREAT Asia/amfAR.
Study population and analysis time
Late ART initiators were defined as PLHIV who started ART with CD4 <100 cells/μL. Early mortality was defined as mortality occurring in the first year of ART initiation.
Patients enrolled in TAHOD were included if they had initiated ART between 2003 and 2018 with CD4 <100 cells/μL and had a subsequent follow up visit. PLHIV in follow-up for more than one year were censored at 12 months. Patients were also censored at their date of death or lost-to follow-up (LTFU) if it occurred within the analysis time. LTFU was defined as not seen in clinic for >12 months without evidence of transfer.
Statistical analyses
The standardized Cause of Death (CoDe) form by the Data collection on Adverse events of Anti-HIV Drugs (D:A:D) group is used in the cohort to ascertain causes of death. These forms were used to review and categorize causes of death as AIDS-related, Immune Reconstituted Inflammatory Syndrome (IRIS)-related, non-AIDS or unknown. The AIDS-causes of death were further reviewed for their AIDS-defining illness collected in the cohort.
Factors associated with early all-cause mortality were analysed using Fine and Gray’s competing risk regression model. Risk time began from date of ART initiation or date of cohort entry for those who started treatment prior to cohort entry and ended on the date of death for those who died in the first year. PLHIV who became LTFU in the first year were censored on the date of LTFU and included as a competing risk. Other patients, including those who transferred out, were censored on the date of last visit or one year from ART initiation, whichever occurred first. Time-fixed covariates were sex, HIV mode of exposure, World Bank country income level, hepatitis B and hepatitis C co-infection, prior AIDS defining illnesses including TB and PCP, and receipt of prophylaxes for OIs. Time-dependent covariates, i.e variables that were allowed to change in values over time during the first year of follow up included age, HIV viral load (VL), CD4 cell counts, BMI, haemoglobin, alanine aminotransferase (ALT) elevations, creatinine, early ART interruptions, ART adherence, and calendar year of follow-up. BMI was categorized as <18.5 kg/m2, 18.5 – 24.9 kg/m2, and ≥ 25 kg/m2.
Haemoglobin levels were grouped as <13.0 g/dL (anaemic) and ≥13g/dL (non-anaemic) for males and <12.0 g/dL (anaemic) and ≥12 g/dL (non-anaemic) for females. (10) ALT was categorized as <5 times its upper limit of normal (ULN) (not severe) or ≥ 5 times ULN (severe). Similarly, creatinine levels were grouped as <1.8 times ULN (not severe) and ≥ 1.8 times ULN (severe). (11) Early ART interruptions were indicated by the absence of ART for ≥ 14 days in the first year of follow-up. Self-reported ART adherence was obtained from the visual analogue scale. (12)
Except for country income level which was included as a priori, all co-variates in the univariate analysis with p <0.10 were fit in the multivariate model using backward stepwise selection process. Covariates with p <0.05 in the multivariate model were considered significant.
All data management and statistical analyses were performed using SAS software version 9.4 (SAS Institute Inc., Cary, NC, USA) and Stata software version 14.2 (Stata Corp., College Station, TX, USA).
Results
A total of 1813 TAHOD patients from Cambodia, China and Hong Kong SAR, India, Indonesia, Japan, Malaysia, Philippines, Singapore, South Korea, Taiwan, Thailand, and Vietnam who initiated ART with CD4 <100 cells/μL were included. The median age at ART start was 35 years [interquartile range (IQR) 29–41], with majority being male (74%) and reporting heterosexual HIV exposure (64%). The median BMI was 19.6 kg/m2 (IQR 17.6-21.8) and the median CD4 cell counts was <50 cells/μL (34 cells/μL (IQR 14–60)). More patients initiated treatment between 2008–2012 (64%) compared to 2003–2007 (28%) or 2013–2018 (8%) to which extent reflects recruitment patterns into TAHOD (Table 1). No significant difference (p=0.930) was found in the median CD4 cell counts over the year groups (2003–2005:33 cells/μL; 2006-2009:33 cells/μL; 2010-2012: 35 cells/μL; 2013-2018: 33 cells/μL) among those initiating treatment late.
Table 1.
Patient characteristics at ART start in patients who have commenced ART with CD4<100 cell/uL and those who died within 12 months in the TREAT Asia HIV Observational Database (TAHOD)
| All patients | Those who died within 12 months | |
|---|---|---|
| n(%) | n(%) | |
| Total | 1813 (100) | 73 (4) |
| Age (years) | ||
| Median (IQR) | 35 (29-41) | 34 (28-42) |
| Sex | ||
| Male | 1335 (74) | 58 (79) |
| Female | 478 (26) | 15 (21) |
| HIV Exposure | ||
| Heterosexual | 1208 (67) | 45 (62) |
| MSM | 268 (15) | 8 (11) |
| Injecting drug use | 186 (10) | 14 (19) |
| Other/Unknown | 151 (8) | 6 (8) |
| CD4 cell count (cells/ul) at ART initiation | ||
| Median (IQR) | 34 (14-60) | 22 (12-41) |
| Viral Load (copies/mL) at ART initiation | ||
| Median (IQR) | 150000 (70185-440000) | 170000 (83984-430000) |
| HCV co-infection antibody | ||
| Negative | 1110 (61) | 36 (49) |
| Positive | 253 (14) | 19 (26) |
| Not reported | 450 (25) | 18 (25) |
| BMI at ART initiation | ||
| Median IQR | 19.6(17.6-21.8) | 18.1(16-20.5) |
| Haemoglobin in g/dL at ART initiation | ||
| Median (IQR) | 12(10-13) | 11 (10-12) |
| ALT at ART initiation | ||
| <5.0 | 1405 (78) | 56 (77) |
| >=5.0 | 11 (1) | 1 (1) |
| Not reported | 397 (22) | 16 (22) |
| Adherence | ||
| Always adherent (>=95%) | 1141 (63) | 32 (44) |
| Ever non-adherent (<95%) | 57 (3) | 4 (5) |
| Not reported | 615 (34) | 37 (51) |
| World Bank country income | ||
| Lower Middle | 779 (43) | 32 (44) |
| Upper Middle | 800 (44) | 30 (41) |
| High | 234 (13) | 11 (15) |
| Year of ART initiation | ||
| 2003-2005 | 385 (21) | 23 (32) |
| 2006-2009 | 724 (40) | 27 (37) |
| 2010-2012 | 559 (31) | 22 (30) |
| 2013-2018 | 145 (8) | 1 (1) |
No., number; MSM, Men who have sex with men; IQR, interquartile range; HCV, hepatitis C virus; BMI, body mass index; ALT, alanine aminotransferase; ART, Antiretroviral Therapy.
Baseline was defined as the date patients started their first triple regimen.
One year after initiating ART, 73 patients (4%) experienced early mortality, of which 38 (52%) were AIDS-related, 10 (14%) related to IRIS, 13 (18%) non-AIDS-related and 12 (16%) due to unknown causes. TB from Mycobaterium tuberculosis (18%) and Pneumocystis pneumonia (PCP/PJP) (16%) were found prevalent among the AIDS deaths.
The overall early mortality rate was 4.27 per 100 person-years (/100PYS). This rate decreased from 7.33/100 PYS during follow-up years 2003-2005 to 0.64/100 PYS in more recent years 2013-2018 (p=0.008). It was found that BMI <18.5 (SHR 2.91, 95% CI 1.60-5.32, p=0.001) compared to BMI 18.5-24.9, HCV co-infection (SHR 2.67; 95% CI 1.40-5.11, p=0.003), and severe ALT elevations (ALT ≥ 5 times its ULN) (SHR 6.14, 95% CI 1.62-23.20, p=0.007) compared to not severe ALT elevations (ALT <5 time its ULN), being anaemic (haemoglobin <13 g/dL for males and <12 g/dL for females) (SHR 2.33, 95% CI 1.15-4.74, p=0.019) compared to non-anaemic levels (≥ 13.0 g/dL for males and ≥ 12.0 g/dL for females), and ever non-adherent (<95%) to ART (SHR 3.38, 95% CI 1.11-10.36, p=0.033) compared to always adherent (≥ 95%) had increased risk for early mortality. Higher current CD4 cell counts (51-100 cell/μL: SHR 0.28, 95% CI 0.15-0.55, p<0.001); and >100 cell/μL: SHR 0.11, 95% CI 0.05-0.23, p<0.001) compared to CD4 ≤ 25 cell/μL was associated with improved survival (Figure 1).
Figure 1:
Forest plot of factors associated with early mortality in patients starting ART with CD4<100 cells/μL
Not reported/not tested categories were included in the multivariate analysis, however, were not graphically displayed. Factors not significant in the multivariate analysis were not displayed.
Discussion
In our cohort, 18.8% initiated treatment with CD<100 cells/μL between 2003 – 2018 with lower proportions in later years. Majority of early mortality was found to be AIDS-related (52%), with HCV co-infection, high ALT elevations, low haemoglobin, low BMI and poor adherence to ART associated with early mortality. Furthermore, the study highlighted that even among those who started treatment with severe immunosuppression levels, those with CD4 >50 cells/μL had better survival in the first year.
The overall early mortality rate of 4.27/100PYS found in our study is comparable to 1-year mortality rates of < 5.0/100PYS found among patients 18-73 years of age initiating ART with CD4 groups 0-49 and 50-99 cells/μL in clinical trials done in Uganda and Zimbabwe. (13) While the median CD4 cell count remained consistent across years groups, studies on the wider clinical populations in the Asia-Pacific region has found CD4 cell counts at ART initiation to have increased over time. (14) The decrease in the rate of early mortality from 7.33/100 PYS to 0.64/100 PYS over follow-up years 2003-2005 to 2013-2018, is believed to be due to availability of new ART regimen and patients initiating ART at higher CD4 levels in the later years. Furthermore, the high proportion of deaths related to AIDS (52%) was in the range of 51-54% reported among late initiators who died 1-year post ART in other cohorts (2, 13) indicating that AIDS-defining illnesses are an important cause of death in late presenting PLHIV. (2) Common AIDS-defining illnesses among those with low CD4 counts in sub-Saharan Africa and South America are TB and PCP. (4-6) In alignment with these literatures, our study found similar results with TB (18%) and PCP (16%) to be the most common types of AIDS-defining illnesses.
Consistent with other studies from the African and Asian regions on late ART initiators, low haemoglobin, low BMI, HCV co-infection were associated with early mortality. (1, 2, 8, 13) These factors impact an already affected immune system contributing to early mortality. Furthermore, low haemoglobin (which could result from HIV disease, co-infections, poor nutritional status, or a combination of all), along with BMI are risk markers known to be found useful in the prognosis of HIV disease progression. (1, 2, 15) Unlike most other studies which found age as a factor known to contribute to early mortality, (1, 2, 7, 8, 13) our study did not find significant association with age. Adherence to ART is vital at all stages of the HIV disease. With 66% of those included in our study reporting on adherence, non-adherence to antiretroviral treatment was associated with early mortality. Studies in Asia with similar findings demonstrated that adherence to ART must be maintained and low adherence increases the risk of mortality. (2, 12) Interestingly, these findings which are similar to other cohorts, are also common to those with CD4 ≥ 100 cells/μL (8, 15) indicating that these factors affect the immune system increasing mortality rates at all CD4 levels.
It is widely known for lower CD4 counts to be associated with mortality (1, 2, 8, 13); in addition, a significant finding of our study was that among those with advanced stages of the HIV disease, those with CD4 cell counts >50 cells/μL have a reduced hazard for mortality compared to those with CD4 cell counts <25 cells/μL improving survival outcomes in the first year. More studies on early mortality and survival rates among late initiators, could further highlight the better prognosis of survival even among with advanced stages of disease enhancing the care provided.
The limitations in our study included not being able to link to death registries; therefore, it is possible that mortality rates may be underestimated due to unascertained mortality among those who were LTFU. Furthermore, patient recruitment for cohort enrolment in TAHOD is not random. While TAHOD contains a rich dataset of information on PLHIV from different income levels, its patterns of treatment and care may be less representative of the broader clinical population.
Conclusion
Early mortality in PLHIV who started late ART in the Asia-Pacific region, is primarily due to AIDS-related causes. With PLHIV still presenting to care at late stages of the HIV disease, and with those initiating ART at CD4 >50 cells/μL having a better survival outcome in the first year, greater effort should be applied to initiating treatment at even modestly higher CD4 counts, improving ART adherence, and careful monitoring of BMI, HCV co-infection, liver function, and anaemic status, even in those with late stages of HIV disease.
Acknowledgments
The study team would like to acknowledge all TAHOD study members, steering committee, and patients for their support:
TREAT Asia HIV Observational Database (TAHOD) contributors include:
PS Ly* and V Khol, National Center for HIV/AIDS, Dermatology & STDs, Phnom Penh, Cambodia;
FJ Zhang* †, HX Zhao and N Han, Beijing Ditan Hospital, Capital Medical University, Beijing, China;
MP Lee*, PCK Li, W Lam and YT Chan, Queen Elizabeth Hospital, Hong Kong SAR;
N Kumarasamy*, S Saghayam and C Ezhilarasi, Chennai Antiviral Research and Treatment Clinical Research Site (CART CRS), YRGCARE Medical Centre, VHS, Chennai, India;
S Pujari*, K Joshi, S Gaikwad and A Chitalikar, Institute of Infectious Diseases, Pune, India;
S Sangle*, V Mave and I Marbaniang, BJ Government Medical College and Sassoon General Hospital, Pune, India;
TP Merati*, DN Wirawan and F Yuliana, Faculty of Medicine Udayana University & Sanglah Hospital, Bali, Indonesia;
E Yunihastuti*, D Imran and A Widhani, Faculty of Medicine Universitas Indonesia - Dr. Cipto Mangunkusumo General Hospital, Jakarta, Indonesia;
J Tanuma*, S Oka and T Nishijima, National Center for Global Health and Medicine, Tokyo, Japan;
JY Choi*, Na S and JM Kim, Division of Infectious Diseases, Department of Internal Medicine, Yonsei University College of Medicine, Seoul, South Korea;
BLH Sim*, YM Gani, and NB Rudi, Hospital Sungai Buloh, Sungai Buloh, Malaysia;
A Kamarulzaman*, SF Syed Omar, S Ponnampalavanar and I Azwa, University Malaya Medical Centre, Kuala Lumpur, Malaysia;
R Ditangco*, MK Pasayan and ML Mationg, Research Institute for Tropical Medicine, Muntinlupa City, Philippines;
WW Wong*, WW Ku and PC Wu, Taipei Veterans General Hospital, Taipei, Taiwan;
OT Ng* ‡, PL Lim, LS Lee and Z Ferdous, Tan Tock Seng Hospital, Singapore;
A Avihingsanon*, S Gatechompol, P Phanuphak and C Phadungphon, HIV-NAT/Thai Red Cross AIDS Research Centre, Bangkok, Thailand;
S Kiertiburanakul*, A Phuphuakrat, L Chumla and N Sanmeema, Faculty of Medicine Ramathibodi Hospital, Mahidol University, Bangkok, Thailand;
R Chaiwarith*, T Sirisanthana, W Kotarathititum and J Praparattanapan, Research Institute for Health Sciences, Chiang Mai, Thailand;
S Khusuwan*, P Kantipong and P Kambua, Chiangrai Prachanukroh Hospital, Chiang Rai, Thailand;
KV Nguyen*, HV Bui, DTH Nguyen and DT Nguyen, National Hospital for Tropical Diseases, Hanoi, Vietnam;
CD Do*, AV Ngo and LT Nguyen, Bach Mai Hospital, Hanoi, Vietnam;
AH Sohn*, JL Ross* and B Petersen, TREAT Asia, amfAR - The Foundation for AIDS Research, Bangkok, Thailand;
MG Law*, A Jiamsakul* and D Rupasinghe, The Kirby Institute, UNSW Sydney, NSW, Australia.
* TAHOD Steering Committee member; † Steering Committee Chair; ‡ co-Chair
Funding and conflicts of interest section
This study was supported by the TREAT Asia HIV Observational Database which is funded by International Epidemiology Databases to Evaluate AIDS (IeDEA; U01AI069907). The Kirby Institute is funded by the Australian Government Department of Health, and is affiliated with the Faculty of Medicine, UNSW Sydney. The content of this publication is solely the responsibility of the authors and does not necessarily represent the official views of any of the governments or institutions mentioned above. The authors have no conflict of interest to disclose.
References
- 1.Siika A, McCabe L, Bwakura-Dangarembizi M, Kityo C, Mallewa J, Berkley J, et al. Late Presentation With HIV in Africa: Phenotypes, Risk, and Risk Stratification in the REALITY Trial. Clin Infect Dis. 2018;66(suppl_2):S140–S6. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Gupta A, Nadkarni G, Yang WT, Chandrasekhar A, Gupte N, Bisson GP, et al. Early mortality in adults initiating antiretroviral therapy (ART) in low- and middle-income countries (LMIC): a systematic review and meta-analysis. PLoS One. 2011;6(12):e28691. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Organization WH. CONSOLIDATED GUIDELINES ON THE USE OF ANTIRETROVIRAL DRUGS FOR TREATING AND PREVENTING HIV INFECTION. World Health Organization; 2016. [PubMed] [Google Scholar]
- 4.Salami AK, Olatunji PO, Oluboyo PO. Spectrum and prognostic significance of opportunistic diseases in HIV/AIDS patients in Ilorin, Nigeria. West Afr J Med. 2006;25(1):52–6. [DOI] [PubMed] [Google Scholar]
- 5.Reinhardt SW, Spec A, Melendez J, Cordon AA, Ross I, Powderly WG, et al. AIDS-Defining Illnesses at Initial Diagnosis of HIV in a Large Guatemalan Cohort. Open Forum Infect Di. 2017;4(4). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Saraceni V, King BS, Cavalcante SC, Golub JE, Lauria LM, Moulton LH, et al. Tuberculosis as primary cause of death among AIDS cases in Rio de Janeiro, Brazil. Int J Tuberc Lung Dis. 2008;12(7):769–72. [PMC free article] [PubMed] [Google Scholar]
- 7.Lee CY, Tseng YT, Lin WR, Chen YH, Tsai JJ, Wang WH, et al. AIDS-related opportunistic illnesses and early initiation of HIV care remain critical in the contemporary HAART era: a retrospective cohort study in Taiwan. BMC Infect Dis. 2018;18(1):352. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Kiertiburanakul S, Boettiger D, Lee MP, Omar SF, Tanuma J, Ng OT, et al. Trends of CD4 cell count levels at the initiation of antiretroviral therapy over time and factors associated with late initiation of antiretroviral therapy among Asian HIV-positive patients. J Int AIDS Soc. 2014;17:18804. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Zhou J, Kumarasamy N, Ditangco R, Kamarulzaman A, Lee CK, Li PC, et al. The TREAT Asia HIV Observational Database: baseline and retrospective data. J Acquir Immune Defic Syndr. 2005;38(2):174–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Organisation WH. Haemoglobin concentrations for the diagnosis of anaemia and assessment of severity. Vitamin and Mineral Nutrition Information System. 2011.
- 11.U.S. Department of Health and Human Services NIoH, National Institute of Allergy and Infectious Diseases, Division of AIDS. Division of AIDS (DAIDS) Table for Grading the Severity of Adult and Pediatric Adverse Events, Corrected Version 2.1.; July 2017.
- 12.Jiamsakul A, Kumarasamy N, Ditangco R, Li PC, Phanuphak P, Sirisanthana T, et al. Factors associated with suboptimal adherence to antiretroviral therapy in Asia. J Int AIDS Soc. 2014;17:18911. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Walker AS, Prendergast AJ, Mugyenyi P, Munderi P, Hakim J, Kekitiinwa A, et al. Mortality in the year following antiretroviral therapy initiation in HIV-infected adults and children in Uganda and Zimbabwe. Clin Infect Dis. 2012;55(12):1707–18. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.De La Mata NL, Ly PS, Ng OT, Nguyen KV, Merati TP, Pham TT, et al. Trends in CD4 cell count response to first-line antiretroviral treatment in HIV-positive patients from Asia, 2003–2013: TREAT Asia HIV Observational Database Low Intensity Transfer. Int J STD AIDS. 2017;28(13):1282–91. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Kwantwi LB, Tunu BK, Boateng D, Quansah DY. Body Mass Index, Haemoglobin, and Total Lymphocyte Count as a Surrogate for CD4 Count in Resource Limited Settings. J Biomark. 2017;2017:7907352. [DOI] [PMC free article] [PubMed] [Google Scholar]