Page 224 - SAHCS HIVMed Journal Vol 20 No 1 2019

Page 5 of 6 Original Research

standard test. In contrast, AST:ALT ratio performance was not In conclusion, this is the first study in Zimbabwe to
comparable to any of the other three algorithms, showing poor demonstrate that algorithms such as FIB4-index, APRI test
agreement with any of the three algorithms. Therefore, we and FibroTest together with individual serum biomarkers
speculate that this discordance could be owing to overexpression like A-2M can be used as alternative methods for assessing
of AST from non-hepatic sources and also delayed clearance of liver fibrosis in asymptomatic, HIV-infected individuals on
AST relative to ALT. 35,36 Thus, making AST-based algorithms ART. The moderately high prevalence of asymptomatic liver
problematic in assessing liver fibrosis by overestimating the fibrosis obtained in this study warrants adequate monitoring
degree of liver fibrosis. AST:ALT ratio consequently becomes a among ART-experienced individuals. The discordance of
very unreliable test for estimating significant fibrosis in ART- fibrosis results among the algorithms, and individual
experienced individuals. Among the comparable algorithms, biomarkers call for further work in identifying optimal
FIB-4 index is the best in terms of reliability and ease of biomarkers for detection of asymptomatic fibrosis. However,
performing because it includes age in its calculations and the AST:ALT ratio does not require further work, as it has been
tests involved in calculation are routinely available. shown to be an unreliable test for assessing liver fibrosis.
16
Whilst a number of studies, including the Ugandan study,
We compared individual biomarkers between fibrotic and have demonstrated high levels of fibrosis in HIV-
non-fibrotic individuals as defined by the FibroTest and APRI monoinfected patients, the natural history and long-term
test to determine the possible utility of using a single biomarker liver outcomes in this group of patients have not been well
in prediction of fibrosis. Only AST and A2M remained described. There is, therefore, an urgent need to have well-
significantly higher in patients with liver fibrosis compared to designed cohort studies looking at long-term outcomes in
those without, adjusting for multiple comparisons. We this group of patients and an easy to apply non-invasive test
therefore speculate that the observed significance could be in this setting. Introducing better non-invasive markers of
because of the fact that (1) most of the AST in hepatocytes is liver fibrosis in Zimbabwe and sub-Saharan Africa has a
located in the mitochondria and damage to hepatocytes causes potential of simplifying and improving the way ART-
the release of both the cytoplasmic and mitochondrial AST, experienced patients are monitored for liver fibrosis.
leading to a raised AST in liver fibrosis, and (2) alpha-2
35
macroglobulin, a PI, is produced by hepatocytes, granuloma Acknowledgements
cells and hepatic stellate cells during inflammation. Hence, its
synthesis is increased as the body tries to inhibit breakdown of The authors would like to thank the Letten foundation for
ECM proteins, which favours fibrosis. In our study, A-2M funding this project and all the participants for participating
37
was able to strongly distinguish fibrosis from non-fibrosis and in this study. The authors extend their gratitude to the
this has also been demonstrated in a study conducted personnel of the Department of Chemical Pathology, College
37
in Romania. Aspartate aminotransferase significantly of Health Sciences, University of Zimbabwe for all the
distinguished fibrosis from non-fibrosis, but the enzyme is not support throughout the whole project.
a good individual biomarker of liver fibrosis as it is not solely
produced by the liver and this results in false-positives. On the Competing interests
contrary, A-2M is an excellent individual biomarker of liver
fibrosis, but analysis of this protein is costly, thus making it not The authors declare that they have no financial or personal
ideal for routine assessment of liver fibrosis in poor resource relationship(s) which may have inappropriately influenced
settings. them in writing this article.

This study was limited by the unavailability of liver LB, Authors’ contributions
which is the gold standard method for assessing liver
fibrosis. We could not assay LB on our patients as it is B.N., K.M., Prof. B.S.-P. and Prof. H.T.M. conceived the study
ethically unacceptable for routine monitoring of liver and its design. B.N., A.M. and T.K.N. participated in the
fibrosis, particularly in the HIV-infected individuals because collection of clinical material and study participant
it increases the risks of coagulopathies. 8,35,38 However, recruitment. B.N., A.M. and C.M. conducted laboratory tests.
including LB in our study could have potentially helped us B.N., P.T.M. and K.M. performed data analysis. B.N. wrote
in drawing a stronger conclusion on the diagnostic the first draft with inputs from K.M., T.K., A.M., C.M., P.T.M.,
performance of each algorithm. Transient elastography was Prof. H.T.M. and Prof. B.S.-P. All authors reviewed and
going to be a better comparator among the non-invasive approved the final version of the manuscript.
tools of assessing liver fibrosis. However, the test was not
performed owing to a lack of availability of the test and also References
budget constraints. Cut-off values used in our study were
based on studies performed mainly in Europe, with most 1. Spearman CW, Sonderup MW. Health disparities in liver disease in sub-Saharan
Africa. Liver Int. 2015;35(9):2063–2071. https://doi.org/10.1111/liv.12884
participants having HCV. 18,39 Although we have no reason to 2. Price JC, Thio CL. Liver disease in the HIV-coinfected individual. Clin Gastroenterol
believe that the cut-off values would differ in our setting, Hepatol. 2010;8(12):1002–1012. https://doi.org/10.1016/j.cgh.2010.08.024
studies to validate the cut-off values in an African population 3. Weber R, Sabin CA, Friis-Møller N, et al. Liver-related deaths in persons infected
with the Human Immunodeficiency Virus: The D: A: D study. Arch Intern Med.
are recommended. 2006;166(15):1632–1641. https://doi.org/10.1001/archinte.166.15.1632

http://www.sajhivmed.org.za 217 Open Access