A superiority of viral load over CD4 cell count when predicting mortality in HIV patients on therapy

Abstract

Background CD4 cell count has been identified to be an essential component in monitoring HIV treatment outcome. However, CD4 cell count monitoring sometimes fails to predict virological failure resulting in unnecessary switch of treatment lines which causes drug resistance and limitations of treatment options. This study assesses the use of both viral load (HIV RNA) and CD4 cell count in the monitoring of HIV/AIDS progression. Methods Time-homogeneous Markov models were fitted, one on CD4 cell count monitoring and the other on HIV RNA monitoring. Effects of covariates; gender, age, CD4 baseline, HIV RNA baseline and adherence to treatment were assessed for each of the fitted models. Assessment of the fitted models was done using prevalence plots and the likelihood ratio tests. The analysis was done using the “msm” package in R. Results Results from the analysis show that viral load monitoring predicts deaths of HIV/AIDS patients better than CD4 cell count monitoring. Assessment of the fitted models shows that viral load monitoring is a better predictor of HIV/AIDS progression than CD4 cell count. Conclusion From this study one can conclude that although patients take more time to achieve a normal CD4 cell count and less time to achieve an undetectable viral load, once the CD4 cell count is normal, mortality risks are reduced. Therefore, both viral load monitoring and CD4 count monitoring can be used to provide useful information which can be used to improve life expectance of patients living with HIV. However, viral load monitoring is a better predictor of HIV/AIDS progression than CD4 cell count and hence viral load is deemed superior. Electronic supplementary material The online version of this article (10.1186/s12879-019-3781-1) contains supplementary material, which is available to authorized users.

BoLA-DRB3 Polymorphism is Associated with Differential Susceptibility to Bovine Leukemia Virus-Induced Lymphoma and Proviral Load

“…Viral load in chronic infections with viruses, such as hepatitis B virus (HBV), hepatitis C virus (HCV), human T cell leukemia virus type 1 (HTLV-1), and human immunodeficiency virus type 1 (HIV-1), has been reported to determine the likelihood of pathogenesis and disease progression [1][2][3][4]. For retroviruses, whose genome integrates with the host genome, proviral load (PVL) is an important risk factor of viruss-associated disease prediction [5,6].…”

Validation of the BD FACSPresto system for the measurement of CD4 T-lymphocytes and hemoglobin concentration in HIV-negative and HIV-positive subjects

“…The use of the BD FACSPresto system for POC testing of AbsCD4 and CD4% potentially could improve the viral load monitoring of people living with HIV in China, particularly in rural areas and resource-limited settings. CD4 T cell testing is an important diagnostic tool that offers valuable insights of the immune system status for monitoring and long-term care management, disease stage and progression, risk of opportunistic infections and mortality risk 37 , 38 ; therefore, enumerating absolute CD4 counts is key to prioritizing decisions related to the initiation of ART in settings where standard treatment is unavailable 39 . Furthermore, although viral load is considered superior to CD4 cell count for monitoring the response to ART 37 , CD4 cell count remains the single most important parameter in places where viral load testing is not available 8 – 10 .…”

A Critical Review of Multistate Bayesian Models Modelling Approaches in Monitoring Disease Progression: Use of Kolmogorov-Chapman Forward Equations with in WinBUGS

“…In scenarios whereby the absorbing state is not considered, the K-progressive models which follow a sequential process (15) for instance health, mild, moderate and severe sequence or the fertility model which is used to describe the reproductive life history of a woman, Figure 1D, where each state is defined by the number of children born are commonly used. Application of multistate models is not limited to biomedical studies like the evaluation of disease progression patterns (16)(17)(18) but cuts across various life history data, including health economics. In health economics studies inclined to the monitoring of disease progression, issues on the costeffectiveness of prevention strategies (19), treatment (20), and diagnosis intervention (21) to inform policy decision-making process (22) can be addressed using multistate models.…”