Determinants of Survival Among HIV/AIDS Patients With The Second-Line Antiretroviral Therapy In Henan Province of China

Background: To estimate the survival and effectors of mortality among HIV/AIDS patients switch to second-line highly active antiretroviral therapy (ART) in rural China. Methods: A three years’ retrospective cohort study was conducted and HIV/AIDS patients switched to the second-line ART between January 2009 to December 2014 enrollment. The data collected from medical records and analysis using Kaplan-Meier statistics and COX regression models. Findings: A total of 2883 HIV/AIDS participants followed up for 8445 person-years, 183 (6.5%) died, 14(0.5%) lost follow-up and the mortality rate 2.17/100 person-years. After adjusting other confounding factors by multivariable COX regression, age older than 50 years (HR,3.37; 95%CI, 1.92-5.92), Traditional Chinese medicine therapy (HR,0.48; 95%CI, 0.33-0.71), CD4 cell count littler than 200 cells/μl (HR,2.97; 95%CI, 1.90-4.64), AST or ALT higher than 50 u/L (HR,1.55; 95%CI, 1.15-2.11) were each independently associated with mortality among HIV/AIDS patients switch to second-line ART. Conclusions: Our retrospective cohort study indicates that mortality among HIV/AIDS patients switch to second-line ART lower than most other studies. However, the limitations of a retrospective cohort could have biased the study, so prospective studies should be carried out to conrm our primary results. The result of our study suggest that Chinese therapy was potential treatment for HIV/AIDS patients.

Henan province lies in the middle of china, being one of HIV High Incidence areas. From 2003 year, HAART had been used among those patients according to the handbooks of national ART. Second-line ART was introduced for Henan province in 2009, and from then on, most of studies on second-line ART was focusing on drug resistances and treatment failure [9,10], little on the mortality. Hence, we design a retrospective cohort study to analysis the mortality of HIV/AIDS after second-line ART to explore the determinants of survival. Methods

study design, period and setting
A three years' retrospective cohort study was conducted in Henan province, which is in center of China, based on standard medical record registers. Most HIV-infected people in Henan are living in resource limited regions and were infected through paid blood supply and illegal blood plasma collecting in the 1990s [11]. China initiated a large-scale program for free HAART in 2003, with zidovudine (AZT)/ stavudine (D4T) + didanosine (DDI) + nevirapine (NVP) as the main rst-line therapy regimen. But in 2009, adult patients with failure of the rst-line treatment protocol have to switch to second-line ART, which is comprise of lamivudine (3TC) + tenofovir (TDF) + lopinavir/ritonavir (Lpv/r) [12]. Henan was one of the earliest areas to begin free rst-line and second-line HAART as described in the National Free Antiretroviral Treatment Program (NFATP). In the last decade, Henan province has made signi cant strides in treating HIV/AIDS with Chinese medicine (CM) and this program reported elsewhere [13][14][15]. The information of HIV/AIDS patients researched in this study were collected in the medical record registers of NFATP and CM therapy.

Study population
All individuals in this study were in the area of taking CM therapy before year of 2009, switched to second-line ART between January 2009 to December 2014, older than 18 years and younger than 65 years when begin second-line ART. Individual who were incomplete data on basic variables were excluded from the study. The primary endpoint was all-cause death. Individuals without recorded deaths after three years following, or recorded lost follow-up or withdrawal ART were taken as censored individuals.

Data collection and variables
Information of individuals including age, gender, marital status, race, education, occupation, route of infection, year of HIV con rmed time, year of beginning ART and beginning second-line ART, whether taken Chinese medicine, CD4 cell count, hemoglobin (Hb), alanine aminotransferase (ALT), aspartate aminotransferase (AST) year of death, and year of censored were collected. All the information from standard medical record registers adopted by o ce of NTCMTP or Center for Disease Control and Prevention (CDC), Henan. The laboratory tests such as CD4 cell count, Hb, AST, ALT were the recorded value latest time from the beginning time of second-line cART in half year, if no value in half year, the variable was missing.

Data analysis
Categorical variables were reported as whole numbers with proportions, and continuous variable were reported as medians with interquartile range (IQR) unless indicated otherwise. Life time table method was used to compute the cumulative survival rates and mortality density. Cox proportional hazard regression models was used to identify the effectors of mortality and calculate hazard ratios (HR). Univariable COX proportional hazard regression models was test rst and those variables which become tted on the univariable regression at 0.10 level of signi cance were included in the multivariable analysis. The results of these models were expressed as HR with 95% con dence interval (CI). Statistical analyses were performed by using SPSS 19.0 (SPSS Inc., Chicago, IL, USA).
The variable at 0.05 level of signi cance computed by multi-variable analysis were the explanatory variable on time to death after initiation of second-line ART. The Kaplan-Meier model was used to estimate the survival and log-rank test were used to compare survival curve of different groups classi ed by the explanatory variable. The gures of survival curve were performed by using R (version 3.6.1) software. All statistical analyses were performed Two-sided P-values less than 0.05 were considered to indicate statistical signi cance.

Results
Summary of population 2,883 records of individuals initiating second-line ART in center China during 2009-2014 were eligible for inclusion in this analysis, 1513(52.5%) were female and 1370 (47.5%) were male, the mean age was 48.0±8.2 years. 2867 (99.4%) were Han and 2791 (96.8%) were farmer. The comparison of Scio-Demographic characteristics of the patients at initiation of study is shown in Table 1.

Analysis the risk group
After the multivariable analysis, age, CD4 cell count, Chinese medicine therapy and unmoral AST/ALT were the explanatory variables on time to death after initiation of second-line cART. The group of age<50 years followed for 3 years including 5395 person-years, 81 (5.5%) died, and the mortality rate 1.50/100 person-years. The group of age >50 years followed for 3 years including 3050 person-years, 102(10.0%) died, and the mortality rate 3.34/100 person-years ( g 1 A).The group of cd4>200 followed for 3 years including 6099 person-years, 95 (4.7%) died, and the mortality rate 1.56/100 person-years. The group of CD4 <200 years followed for 3 years including 2346 person-years, 88 (11.0%) died, and the mortality rate 3.75/100 person-years ( g 1 B).No Chinese therapy group followed for 3 years including 6024 personyears, 149 (7.4%) died, and the mortality rate 2.47/100 person-years. Chinese therapy group followed for 3 years including 2421 person-years, 34 (4.2%) died, and the mortality rate 1.40/100 person-years ( g 1 C). The group of AST or ALT <50 followed for 3 years including 4753.5 person-years, 88 (5.7%) died, and the mortality rate 1.85/100 person-years. The group of AST or ALT >50 followed for 3 years including 3691.5 person-years, 95 (7.6%) died, and the mortality rate 2.57/100 person-years ( g 1 D).

Discussion
The main goal of this study was to estimate the mortality rate of HIV/AIDS patients after second-line ART and to identify its effectors in rural China. The overall mortality rate of was 2.17/100 person-years with a lowest level compared with the study on the survival of HIV/AIDS patients after ART don' care rst or second-line in Guizhou in China, where the mortality rate was with 8.53/100 person-years [16], or a study in and Henan in China, where the mortality rate was with 3.9/100 person-years [17], which it maybe the second-line ART had good e cacy. Another study conducted in African and Asian in 2010 reported the mortality rate was 4.42/100 person-years after patients switch to second-line ART [18], higher than this study. According to the global HIV statistics fact sheet reported by WHO, the mortality rate of HIV/AIDS patients was decline sharply [1]. The different results among the studies might be due to difference in characteristics sample, in length of study or in the study period.
The effect of gender on survival has often been the focused, but the result had varied in the literature. Most studies report women had lower risk of death than men [19,20], but in our study, there was no difference between women and men, even though adjust other effectors. Older age was a higher risk of death, we report that the HR of death among HIV/AIDS patients older than 50 years was 3.37 times that of those under 40, patients with 40-50 years old was not show higher HR of death compared with those under 40. Some studies have reported that older HIV/AIDS patients had more comorbidities and be signi cantly associated with second-line ART failure [21]. Married status and education status were not associated with mortality in our study.
lower CD4 cell count at switch to second-line was independent risk factor on time to death, which accordance with many studies. The HR of death among HIV/AIDS patients with CD4 cell count lower than 200 cells/μl was 2.97 times that of higher than 500 cells/μl. Patients with lower CD4 cell count were found to be signi cantly associated with second-line ART failure and have higher probability of developing different opportunistic infections [22], all those more were apt to cause death. Anemia has been documented as a risk factor for morbidity and mortality in these patients, even if the CD4 cell count and viral load are controlled [23][24][25]. However, in our study the anemia had not affect the mortality of HIV/AIDS patients with second-line ART. AST/ALT≥50 u/L, which was de ned as Liver injury [26], was the risk factor of death of HIV/AIDS patients. Liver injury would induce cirrhosis and hepatocellular carcinoma and increase the HR of death in HIV/AIDS patients [27,28].
In our study, the mortality rate of HIV/AIDS patients in CM group was 1.40/100 person-years and the HR was 0.48 (95%CI, 0.33-0.71), compares with mortality of non-CM group, which mortality rate was 2.47/100 person-years. The result suggested CM combined with second-line ART could better increased the survival and lengthened the lifetime of HIV/AIDS patient. In Henan, CM treatment for AIDS for tens year, many effects of TCM on HIV/AIDS have been shown, such as reducing plasma HIV viral load, increasing CD4+ T cell counts, promoting immune reconstitution, ameliorating signs and symptoms, improving the health related quality of life, and counteracting the side effects of anti-retroviral drugs [29][30][31][32].
The data of this study extracted from the normal medical record which can veritable re ect the outcome of the current second-line ART management. As a retrospective study, selection bias was possibly introduced because of the fact that patients with incomplete records would be excluded. Some important variables associated with mortality like the viral load and body mass index when switch to second-line ART which were not recorded for most of the patients. Adherence of ART was con rmed variable don't record in the normal medical record.

Conclusions
In brief, our retrospective cohort study indicates that mortality among HIV/AIDS patients switch to second-line ART lower than most other studies. The result of our study suggest that CM based on ART was a potential treatment for HIV/AIDS patients. However, the limitations of a retrospective cohort could have biased the study, so prospective studies should be carried out to con rm our primary results.

Declarations Ethical Approval and Consent to participate
This study was approved by the institutional review board of the rst hospital a liated to Henan University of Traditional Chinese Medicine(2019HL-068). Individual informed consent was not achieved because this analysis used currently existing data collected during the course of routine treatment, and the data were reported in aggregate without the use of individual identifying information.