Patterns of liver injury in HIV-positive patients in the medical admissions ward at Chris Hani Baragwanath Academic Hospital (CHBAH) Dr. Rudzani Wendy Ndwambi Student Number: 2302007 Supervisors: Prof Reidwaan Ally and Dr. Nazeer Chopdat Dissertation for Master of Medicine in Internal Medicine The study is submitted in partial fulfilment for the Degree of Masters of Medicine in Internal Medicine. Declaration I, Dr. Rudzani Wendy Ndwambi, declare that this dissertation is my work and is being submitted in partial fulfilment for the degree of Master of Medicine in Internal Medicine at the University of Witwatersrand, Johannesburg. This work has yet to be submitted for any degree or examination at this institution or other universities. Signed: RW Ndwambi …………………………………………………………………… Date: 19/04/2024 …………………………………………………………………… !1 Acknowledgement I thank my supervisors, Prof Reidwaan Ally and Dr. Nazeer Chopdat, for their motivation, guidance, and inspiration on this research project. I thank the Department of Internal Med- icine head for permitting the research in admissions ward 20. Mr. Thabo for assistance in retrieving the patients' files for the study. Special thanks to my loving, supportive husband, Dr. Michael N. Ndwambi, my children (Bella and Max), family, and friends for continuous support and encouragement. I am grateful to the Almighty God for his strength and wisdom. !2 Abstract Background Human immunodeficiency virus (HIV) infection is a major global public health concern, with approximately 37.9 million people living with HIV and acquired immunodeficiency syn- drome (AIDS) as of 2018. There has been an increase in HIV prevalence globally, with the African sub-Saharan region carrying a disproportionate burden, accounting for more than 70% of this burden. In 2018 South Africa had 7.52 million people living with HIV/AIDS (PLWHA), with approximately 115167 AIDS-related illnesses. In PLWHA, liver disease and failure contribute to more significant morbidity, mortality and higher cost of care. Aim And objectives To recognise and categorise the patterns of liver injury in people living with HIV and AIDS (PLWHA). To ascribe an etiology to the pattern of liver injury in PLWHA. Method This was a retrospective cohort, conducted at CHBAH medical admissions ward in Soweto Gauteng province, of patients living with HIV and AIDS with liver injury. All patients admitted to the medical admissions ward were selected and their hospital numbers retrieved from the admission register. Each hospital number was entered into the NHLS labtrack system to retrieve the LFT results. Any patient with abnormal LFT was checked for their HIV status. The two variables (abnormal LFT and HIV test) were matched and confirmed. After ethics approval was obtained, files were recorded and an- alysed. A data collection sheet was populated with all serological, histological and radio- logical investigations documented. Results This study included 208 patients (PLWHA) admitted to the medical admissions ward at CHBAH, with abnormal liver enzymes between January 2019 and March 2020, aged above 18 years. One hundred and five, 50.5% were males with a mean of 43.7 years, and 49.5% (n=103) were females with a mean of 39.9 years. One hundred and forty-three, 81.3% were taking antiretroviral therapy (ART), with the majority, 88.4% (n=107), on the first-line regimen. The most common pattern of liver injury was infiltrative, accounting for 67.8% (n=141), followed by mixed at 22.1% (n=46), hepatocellular at 7.7% (n=16), and lastly was cholestatic with 2.4% (n=5). Irrespective of the pattern of the liver injury, 36.1% !3 (n=75) of the patients had Mycobacterium tuberculosis (MTB) infection, and 22.5% (n=40) of them were on antituberculosis treatment (ATT), 6.3% (n=13) had hepatitis B viral (HBV) infection, 0.5% (n=1 ) had hepatitis C ( HCV) infection, while lymphomas (Hodgkin and non-Hodgkin lymphomas) contributed 1,9%(n=4) and 6,3% (n=13) respectively. Twelve, 5.7% (n=12) had drug induced liver Injury (DILI), and retroviral disease (RVD) cholan- giopathy contributed 1.4% (n=3). A significant number of patients, 34.1% (n=71), con- tributed to at least one-third of patients in the study, wherein the diagnosis was either un- known or not directly related to the liver injury. Those were, Other opportunistic infections accounting for 7.7% (n=16), other diagnoses made up 13% (n=27), and no diagnosis at 13.4% (n=28). Of the infiltrative pattern, 33.13% (n=47) had MTB infection, 5.6% (n=8) had HBV, 0.3% (n=1) had HCV infection, 7.8% (n=11) had non-Hodgkin lymphoma, with Hodgkin lym- phoma, and DILI both accounting for 1,4%(n=2). Only 5.7% (n=12) of PLWHA had liver biopsies done. Conclusion Liver injury is common in PLWHA. The most common pattern of liver injury is an infiltrative pattern, and the most common etiology was MTB infection in this study. !4 Table of contents Topics Page numbers Title 1 Declaration 1 Acknowledgement 2 Abstract 3-4 List of tables 7 Chapter 1 8-18 Introduction 8 Epidemiology 8 Definitions 9 Patterns of liver injury 9-10 Guidelines to ascribe liver injury 10 Mechanism of liver injury 11 Aetiologies 11-12 Aims 13 Study objectives 13 Outcomes 13 Methods 13 Study design 13 Site of study 13 Study population 13 Sample size 13 Inclusion criteria 14 Exclusion criteria 14 Data collection 14 Data analysis 14 Ethics 15 Timing 15 Funding 15 References 16-18 !5 Chapter 2 19-41 Abstract 19-20 Background 21 Methods 21-22 Results 22-28 Discussion 29-30 Strength of the study 31 Limitations 31 Recommendations 31 Conclusion 31 References 32-34 Appendices 35-41 Annexure 1: Ethics approval letter 35 Annexure 2: medical advisory committee approval letter 36 Annexure 3: Head of Department of internal medicine 37 Annexure 4: Data collection sheet 38-40 Turnitin 41 !6 List of tables Table number Subject Page number Table 1 Demographics of patients in a cohort 23 Table 2 Baseline characteristics of patients abnormal in a cohort 24 Table 3 Patterns of liver injury 25 Table 4 Percentages of patterns found in different aetiologies 27 Table 5 Biopsies were undertaken in a cohort. 28 Table 6 Liver biopsy results 28 !7 Chapter 1 The research protocol and extended literature review TITLE Patterns of liver injury in HIV-positive patients in the medical admissions ward at Chris Hani Baragwanath Academic Hospital (CHBAH) Dr Rudzani Wendy Ndwambi Student Number: 2302007 MMed: Internal Medicine Supervisors: Prof Reidwaan Ally Head of Division of Gastroenterology, Department of Internal Medicine, Chris Hani Baragwanath Academic Hospital Dr. Nazeer Chopdat Gastroenterologist, Division of Gastroenterology, Department of Internal Medicine, Chris Hani Baragwanath Academic Hospital 1. Introduction Human Immunodeficiency Virus (HIV) infection is a major global public health concern, with approximately 37,9 million people living with HIV as of 2018 (1). There has been an increase in HIV prevalence globally, with the African sub-Saharan region carrying a dis- proportionate burden, accounting for more than 70% of this burden. In 2018 South Africa had 7,52 million people living with HIV/AIDS (PLWHA), with approximately 115167 AIDS- related illnesses (2). In PLWHA, liver disease and failure contribute to more significant morbidity, mortality, and higher cost of care (3). 2. Epidemiology A South African study by Hoffman (4) showed that 4% of PLWHA had abnormal LFT, which was five times the upper limit of normal before antiretroviral therapy (ART). Kasper and his !8 colleagues found that 13-18% of PLWHA died from liver-related causes (5). A Nigerian study found that 87% of their cohort had abnormal LFTs; of the hundred and twenty-nine (129) cases they reviewed, 14% showed a cholestatic pattern of liver injury, 83% were he- patocellular, and 3% mixed (6). 3. Definitions LFTs reflect the concentrations of various proteins and enzymes in the blood that are ei- ther produced by the liver cells or released when liver cells are damaged. They measure synthetic, biliary, and intracellular functions. Abnormalities of these proteins/enzymes re- flect abnormalities of these descriptive patterns. Abnormal LFTs are characterized by elevations of alanine transferase (ALT), aspartate transferase (AST), alkaline phosphatase (ALP), gamma-glutamyl transferase (GGT), and bilirubin (BIL) levels. In addition, a reduction in total protein/albumin may reflect diminished synthetic capacity. Though varied amongst laboratories, the upper limit of normal (ULN) is generally within similar ranges (7). ULN; ALT=40units per litre (u/l); AST =40u/l; ALP =130u/l; GGT =60u/l and total bilirubin (TBIL) =20 micromoles per liter (mmol/l). Any one value exceeding the ULN is considered abnormal (8). Liver disease can be described according to the abnormal LFT patterns, the molecular mechanism of the liver injury, or the etiology of liver injury (9). 4. Patterns of liver injury Liver enzyme abnormalities are common in PLWHA, and these abnormalities are classified as either hepatocellular, cholestatic, or mixed/infiltrative (10). It is essential to recognize and differentiate these patterns of liver injury because they can have different etiological and pathophysiological mechanisms. This arbitrary classification is not restrictive, since several diseases and drugs can present with more than one pattern of liver injury. Howev- er, early recognition may help identify the most suggestive cause and prevent unneces- sary investigations and delay in starting appropriate therapy in PLWHA (5). 4.1. Hepatocellular injury Acute cellular rupture releases more ALT (intra-cytoplasmic enzyme), whereas persistent and chronic disease will eventually elevate AST (intra-mitochondrial enzyme). When the !9 primary injury is to the hepatocytes, ALT and AST levels are elevated, with AST being less specific (found in other tissues as well) than ALT (liver-specific). In viral, ischaemic, and drug-induced hepatitis, ALT is elevated more than AST or ALP (7) since it reflects acute rupture of the cell. Slower onset and chronic diseases are associated with longstanding cell destruction and hence greater AST levels, an example being chronic alcohol abuse. 4.2. Cholestatic injury The ALP and GGT levels are elevated in cholestatic disease since the primary injury is to the bile ducts/collecting system. Bile duct inflammation and obstruction lead to cholestatic injury (7). According to the literature, cholestasis may be extrahepatic (choledocholithiasis, malignant obstruction) or intrahepatic (drug toxicity, infiltrative diseases, viral hepatitis). Nevertheless, some diseases, e.g., AIDS cholangiopathy, may affect the intrahepatic and extrahepatic biliary systems. 4.3. Infiltrative patterns An infiltrative pattern occurs when the disease affects both the hepatocytes and the bile ducts. It is often seen when the liver is infiltrated or replaced by a granulomatous or malig- nant infiltration (7). True extrahepatic obstructive causes also elevate the bilirubin (conju- gated), whereas these mixed patterns have significant elevations of the ALP/GGT without elevating the bilirubin. Often, the only way to differentiate infiltrative patterns from obstructive patterns is through liver imaging (ultrasound, computerized tomography scan, magnetic resonance imaging, or cholangiography) and biopsy. 4.4. Mixed patterns In a mixed pattern, there is both cholestatic and hepatocellular enzyme elevation, as well as elevation in bilirubin. 5. Guidelines to ascribe liver injury and severity The European association for the study of liver (EASL) uses a grading system for the severity of the liver injury. This system was recently modified while evaluating their cancer therapy program (11) and is accepted as robust and applicable. In this system, the follow- ing ranges are used to assess the liver injury and severity: !10 Grade 1 indicates mild liver injury, grade 2 moderate liver injury, grade 3 severe liver injury, and grade 4 life-threatening liver injury (8). 6. Mechanisms of liver injury The molecular events that disrupt cellular transport function are complex and interactive. They can either be biochemically or immune-mediated. The biochemical-mediated liver injuries are mainly via the cytochrome p-450 and conjugation reaction systems. The toxic metabolites may alter the plasma membrane and mitochondrial function, change intracellu- lar ion homeostasis, or interfere with enzyme activity. The immune-mediated liver injuries are through immunological interference by various cytokines, nitric oxide, and complement systems (5). Overlapping factors contribute to the development of liver injury and hepatic fibrosis. In HIV-infected individuals, albeit the molecular events mentioned above are the same as in non-HIV-infected individuals, there is the added burden of HIV itself and ART therapy (20). A single disease state often employs several mechanisms in its pathogenesis (5). There- fore, a broad understanding of these mechanisms and linking them to diseases that em- ploy them may be helpful to both clinicians and investigators caring for PLWHA and liver disease. 7. Etiologies According to Sterling (10), the alterations and abnormal LFTs that occur in PLWHA are of- ten caused by different etiologies coexisting together. FEATURE GRADE 1 GRADE 2 GRADE 3 GRADE 4 ALT (u/l) <3 x ULN 3-5 x ULN 5-20 x ULN >20 x ULN AST (u/l) <3 x ULN 3-5 x ULN 5-20 x ULN >20 x ULN ALP (u/l) <2.5 x ULN 2.5-5 x ULN 5-20 x ULN >20 x ULN Bilirubin (mmol/l) <1.5 x ULN 1.5-3 x ULN 3-10 x ULN >10 xULN !11 7.1. Infective causes Several infective agents (viral, bacterial, fungal, or parasitic) cause liver injury that can be hepatic, cholestatic, or mixed. Viral infections, e.g., viral hepatitis B, C, E, cytomegalovirus (CMV), and Epstein-Barr virus, cause hepatocellular injury with AST and ALT levels >25 times the ULN being the predominant pattern. Bacterial infections (e.g., syphilis), dissemi- nated fungal infections (histoplasmosis and cryptococcosis), and parasitic infections (toxo- plasmosis, isosporidiosis, and cryptosporidiosis) can cause either hepatocellular or mixed/ infiltrative patterns (13). 7.2. Granulomatous diseases Granulomatous diseases (sarcoid, amyloid, and tuberculosis) have also been associated with liver involvement in either HIV-infected or non-HIV-infected individuals. They usually present with a mixed pattern that may appear similar to cholestatic injury. 7.2. Drugs and toxins Alcohol, antituberculosis therapy (ATT), and antiretroviral therapy ( ART) are associated with the highest risk of drug-induced liver injury (DILI)(12). Drugs and toxins can cause hepatocellular, cholestatic, or mixed patterns. Studies have found that drug-associated hepatotoxicity and viral hepatitides are the most common causes of liver disease in PLWHA (13). 7.5. Malignancies Kaposi's sarcoma and lymphoma are commonly associated with HIV and have been shown or associated with liver injury in PLWHA (13). The other malignancy implicated in liver injury in PLWHA is hepatocellular carcinoma (HCC). It has recently been found that hepatitis, immunosuppression due to HIV, the direct impact of HIV on liver parenchyma, and the use of hepatotoxic ARVs, all can contribute to HCC pathogenesis (14). 7.6. Autoimmune Autoimmune liver injury, though rare, can occur in patients with well-controlled HIV infec- tion (15). 7.7. Vascular HIV-infected patients can present with a vascular occlusive disease that can impair blood flow to other organs, including the liver (ischaemic liver injury) (16). !12 It is anticipated that the findings of this study will help guide the management plans for PLWHA presenting with liver injury at Chris Hani Baragwanath Academic Hospital (CHBAH). Recognition of liver injury patterns will help clinicians develop a more focused approach to ascribing etiology and instituting appropriate management plans more readily. 8. Aims 8.1. To recognize the patterns of liver injury in PLWHA presenting to the medical admis- sions ward at CHBAH 8.2. To ascribe an etiology to the pattern 9. Study objectives 9.1. To be able to recommend a more straightforward management plan in diagnosing and treating PLWHA who have abnormal LFTs 10. Outcomes 10.1. To provide an understanding of patterns of liver disease in PLWHA 10.2. To provide a straightforward approach to the management of liver injury in PLWHA 11. Methods 11.1. Study design: A retrospective, cohort study. 11.2. Site of study: Chris Hani Baragwanath Academic Hospital medical admissions ward. 11.3. Study population: Patients over the age of 18 years who are living with HIV and have abnormal liver enzymes. 11.4. Sampling 11.4.1. Sample size: All HIV-positive patients admitted to the medical admissions ward with abnormal LFTs from January 2019 to March 2020. Test files were reviewed for further analysis using a frequency rate of approximately 14% (average from previous studies) of the HIV-positive patients with abnormal liver function. 11.4.2. Selection of subjects: Subjects for the study were selected from the CHBAH Ward 20 admission register. Patients with HIV and abnormal LFT were selected, and their hospi- tal numbers were retrieved from the admission register. Each hospital number that was re- trieved was entered into the NHLS lab track system to retrieve the LFT results, and any patient with abnormal LFTs was checked if they had an HIV test done. The two variables (abnormal LFT and HIV test) were matched and confirmed, and the patients' names and !13 hospital numbers with the confirmed abnormal LFTs and HIV tests were recorded. After ethics approval was obtained, patients files were retrieved from the CHBAH filing room. Data from these patients’ files were recorded and analyzed. A data collection sheet (see appendix) was populated with all serological, histological, and radiological investigations available and documented. 11.4.3. Inclusion criteria: All patients over the age of 18 who are HIV positive and were admitted via the medical admissions ward with abnormal LFTs. 11.4.4. Exclusion criteria: Those with normal LFTs and negative HIV tests 12. Data collection Once patients have been identified from the ward 20 admission registry, their HIV test and LFTs correlated, their hospital files were retrieved, and all necessary information extracted. This information was entered into a data collection sheet (see appendix) without identifying the patients. All screened cases were reviewed for clinical presentation, clinical examina- tion findings, investigations done, and provisional diagnosis ascertained by the attending physicians. Further information, including CD4 cell counts, viral loads, and other investiga- tions, was accessed via the NHLS laboratory results portal and analyzed. The data collec- tion sheets and records were electronically recorded through Redcap. Patterns of liver in- jury were recorded as described earlier. All other information that led to the diagnosis was also recorded. For this study, all other relevant available investigations, e.g., bilirubin and ammonia levels, international normalized ratio (INR), prothrombin time (PT), albumin, glucose, viral hepati- tis serology, cytomegalovirus (CMV) staining, alpha-fetoprotein (AFP), Imaging and en- doscopy and histology were also used for analysis. 13. Data analysis After the data had been collected, a statistician was consulted. Statistical analysis was un- dertaken to compute measures of central tendency (mean, median, and mode) and dis- persion measurements (standard deviation, range, variance) where appropriate. The dis- tribution of continuous variables, such as age, was expressed as medians. Frequency ta- bles were generated for gender, ART regimen, CD4 cell count, and viral loads. Testing for the significance of association for categorical data was performed using the Pearson chi- square test. For non-normal distributed data, non-parametric tests, I.e., Mann-Whitney test, to test associations were used. The analysis was performed using STATA version 14 !14 statistical software. Results were displayed in tables. All tests were performed at a 95% confidence interval for statistical significance p-value <0.05. The data sheet is attached. 14. Ethics Ethical approval for the study submitted to the Wits human research ethics committee was obtained (attached). Permission from the Head of Internal Medicine to conduct the study was requested and granted(attached). 15. Timing The expected duration of the study: 16. Funding The expenses of this study, including the printing of documents, were personally funded. Septembe r 2019 October 2019 January 2020 June 2020 July 2020 December 2020 Literature review Protocol preparatio n Protocol assessme nt Ethics approval Data collection Data analysis Writing up thesis !15 References 1. United Nations Programme on HIV/AIDS. Global HIV and AIDS statistics. Geneva: UNAIDS; 2018. Available from: https://www.unaids.org/sites/default/files/media_as- set/unaids-data-2018_en.pdf 2. Statistics South Africa. Statistical Release P0302: Mid-year population estimates 2018. Pretoria: Stats SA; 2018. Available from: http://www.statssa.gov.za/publica- tions/P0302/P03022018.pdf 3. Sherman KE, Rockstroh J, Thomas D. Human Immunodeficiency Virus and Liver Disease: An Update. Hepatology. 2015 Dec;62(6):1871-82. 4. Hoffmann CJ, Charalambous S, Thio CL, Martin DJ, Pemba L, Fielding KL, et al. Hepatotoxicity in an African Antiretroviral Therapy Cohort: The Effect of Tuberculo- sis and Hepatitis B. AIDS. 2007 Jun 19;21(10):1301-8. 5. Kaspar MB, Sterling RK. Mechanisms of liver disease in patients infected with HIV. BMJ Open Gastroenterol. 2017 Oct 26;4(1):e000166. 6. Ejilemele AA, Nwauche CA, Ejele OA. Pattern of abnormal liver enzymes in HIV pa- tients presenting at a Nigerian Tertiary Hospital. Niger Postgrad Med J. 2007 Dec; 14(4):306-9. 7. Ganesan M, Poluektova LY, Kharbanda KK, Osna NA. Liver as a target of human immunodeficiency virus infection. World J Gastroenterol. 2018 Nov 14;24(42):4728- 37. 8. Qin F, Jiang J, Qin C, Huang Y, Liang B, Xu Y, et al. Liver damage in patients living with HIV on antiretroviral treatment with normal baseline liver function and without HBV/HCV infection: an 11-year retrospective cohort study in Guangxi, China. BMJ Open. 2019 Apr 2;9(4):e023140. 9. Cullen JM. Mechanistic classification of liver injury.Toxicol Pathol. 2005;33(1):6–8. 10. Sterling RK, Chiu S, Snider K, Nixon D. The prevalence and risk factors for abnor- mal liver enzymes in HIV-positive patients without hepatitis B or C coinfections. Dig Dis Sci. 2008 May;53(5):1375–82. 11. European Association for the Study of the Liver. EASL Clinical Practice Guidelines: Drug-induced liver injury. J Hepatol. 2019 Jun;70(6):1222-1261. !16 https://www.unaids.org/sites/default/files/media_asset/unaids-data-2018_en.pdf https://www.unaids.org/sites/default/files/media_asset/unaids-data-2018_en.pdf http://www.statssa.gov.za/publications/P0302/P03022018.pdf http://www.statssa.gov.za/publications/P0302/P03022018.pdf 12. Jong E, Conradie F, Berhanu R, Black A, John MA, Meintjes G, et al. Consensus statement: management of drug-induced liver injury in HIV-positive patients treated for TB. S Afr J HIV Med. 2013;14(3):113-9. 13. Rockstroh J, editor. Liver disease in the HIV patient [Internet]. New York: Infectious Disease Advisor - Haymarket Media, Inc.; 2017 [cited 2020 Apr 15]. Available from: https://www.infectiousdiseaseadvisor.com/home/decision-support-in-medicine/ infectious-diseases/liver-disease-in-the-hiv-patient/ 14. Dika IE, Harding JJ, Abou-Alfa GK. Hepatocellular carcinoma in patients with HIV. Curr Opin HIV AIDS. 2017 Jan;12(1):20-5. 15. Kia L, Beattie A, Green RM. Autoimmune hepatitis in patients with human immun- odeficiency virus (HIV): Case reports of a rare, but important diagnosis with thera- peutic implications. Medicine (Baltimore). 2017 Feb;96(7):e6011. 16. Pillay B, Ramdial PK, Naidoo DP. HIV-associated Large-Vessel Vasculopathy: A Review of the Current and Emerging Clinicopathological Spectrum in Vascular Sur- gical Practice. Cardiovasc J Afr. 2015 Mar-Apr;26(2):70-81. 17. Cohen J, Torres C. HIV-associated cellular senescence: A contributor to accelerated aging. Aging Res Rev. 2017 Jul;36:117-124. 18. Murali AR, Carey WD, editors. Liver Test Interpretation - Approach to the Patient with Liver Disease: A Guide to Commonly Used Liver Tests [Internet]. Ohio: Cleve- land Clinic – Center for Continuing Education; 2017 [cited 2020 Apr 15]. Available from: https://www.clevelandclinicmeded.com/medicalpubs/diseasemanagement/he- patology/guide-to-common-liver-tests/ 19. Debes JD, Bohjanen PR, Boonstra A. Mechanism of accelerated liver fibrosis pro- gression during HIV infection. J Clin Transl Hepatol. 2016 Dec 28;4(4):328-35. 20. Naseer M, Dailey FE, Juboori AA, Samiullah S, Tahan V. Epidemiology, determi- nants, and management of AIDS cholangiopathy: A review. World J Gastroenterol. 2018 Feb 21;24(7):767–74. 21. Puoti M, Nasta P, Gatti F, Matti A, Prestini K, Biasi L, et al. HIV-related Liver Dis- ease: ARV Drugs, Coinfection, and Other Risk Factors. J Int Assoc Physicians AIDS Care (Chic). 2009;8(1):30-42. 22. Cai J, Osikowicz M, Sebastian G. Clinical significance of elevated liver transami- nases in HIV-infected patients. AIDS. 2019 Jul 1;33(8):1267-82. 23. Lodenyo H, Schoub B, Ally R, Kairu S, Segal I. Hepatitis B and C virus infections and liver function in AIDS patients at Chris Hani Baragwanath Hospital, Johannes- burg. East Afr Med J. 2000 Jan;77(1):13–5. !17 https://www.infectiousdiseaseadvisor.com/home/decision-support-in-medicine/infectious-diseases/liver-disease-in-the-hiv-patient/ https://www.infectiousdiseaseadvisor.com/home/decision-support-in-medicine/infectious-diseases/liver-disease-in-the-hiv-patient/ https://www.infectiousdiseaseadvisor.com/home/decision-support-in-medicine/infectious-diseases/liver-disease-in-the-hiv-patient/ https://www.clevelandclinicmeded.com/medicalpubs/diseasemanagement/hepatology/guide-to-common-liver-tests/ https://www.clevelandclinicmeded.com/medicalpubs/diseasemanagement/hepatology/guide-to-common-liver-tests/ https://www.clevelandclinicmeded.com/medicalpubs/diseasemanagement/hepatology/guide-to-common-liver-tests/ https://www.ncbi.nlm.nih.gov/pubmed/?term=Lodenyo%252525252525252525252525252525252520H%25252525252525252525252525252525255BAuthor%25252525252525252525252525252525255D&cauthor=true&cauthor_uid=10944831 https://www.ncbi.nlm.nih.gov/pubmed/?term=Schoub%252525252525252525252525252525252520B%25252525252525252525252525252525255BAuthor%25252525252525252525252525252525255D&cauthor=true&cauthor_uid=10944831 https://www.ncbi.nlm.nih.gov/pubmed/?term=Ally%252525252525252525252525252525252520R%25252525252525252525252525252525255BAuthor%25252525252525252525252525252525255D&cauthor=true&cauthor_uid=10944831 https://www.ncbi.nlm.nih.gov/pubmed/?term=Kairu%252525252525252525252525252525252520S%25252525252525252525252525252525255BAuthor%25252525252525252525252525252525255D&cauthor=true&cauthor_uid=10944831 https://www.ncbi.nlm.nih.gov/pubmed/?term=Segal%252525252525252525252525252525252520I%25252525252525252525252525252525255BAuthor%25252525252525252525252525252525255D&cauthor=true&cauthor_uid=10944831 24. Ocama P, Katwere M, Piloya T, Feld J, Ohio KC, Kambugu A, et al. The spectrum of liver diseases in HIV-infected individuals at an HIV treatment clinic in Kampala, Uganda. Afr Health Sci. 2008 Mar;8(1):8-12. 25. Pacheco A, Perazzo H, Cardoso S; Fonseca MJ, Griep R, Lotufo P, et al. HIV infec- tion is an independent risk factor for liver steatosis: a study in HIV mono-infected patients compared to uninfected paired controls and associated risk factors [ab- stract]. Amsterdam: 22nd International AIDS Conference; 23-27 Jul 2018. Available from: https://onlinelibrary.wiley.com/doi/full/10.1002/jia2.25148 26. Jadeja RN, Devkar RV, Nammi S. Oxidative Stress in Liver Diseases: Pathogene- sis, Prevention, and Therapeutics. Oxid Med Cell Longev. 2017;2017:8341286. 27. Hulgan T, Gerschenson M. HIV, and Mitochondria: more than just drug toxicity. J infect Dis .2012 Jun 15; 205(12):1769-71. 28. Rodrigues EM Filho, Fernandes R, Susin R, Fior B. Immune reconstitution inflam- matory syndrome as a cause of autoimmune hepatitis and acute liver failure. Rev Bras Ter Intensiva. 2017 Jul-Sep;29(3):382-5. 29. Robinson MW, Harmon C, O'Farrelly C. Liver immunology and its role in inflamma- tion and homeostasis. Cell Mol Immunol. 2016 May;13(3):267-76. !18 https://onlinelibrary.wiley.com/doi/full/10.1002/jia2.25148 Chapter 2 (Submissible format) Patterns of liver injury in HIV-positive patients in the medical admissions ward at Chris Hani Baragwanath Academic Hospital Authors: Dr. RW Ndwambi, Prof R Ally, Dr. N Chopdat Abstract Background Human immunodeficiency virus (HIV) infection is a major global public health concern, with approximately 37.9 million people living with HIV and acquired immunodeficiency syn- drome (AIDS) as of 2018. There has been an increase in HIV prevalence globally, with the African sub-Saharan region carrying a disproportionate burden, accounting for more than 70% of this burden. In 2018 South Africa had 7.52 million people living with HIV/AIDS (PLWHA), with approximately 115167 AIDS-related illnesses. In PLWHA, liver disease and failure contribute to more significant morbidity, mortality and higher cost of care. Aim and objectives To recognize and categorize the patterns of liver injury in people living with HIV and AIDS (PLWHA). To ascribe an etiology to the pattern of liver injury in PLWHA. Method This was a retrospective cohort, conducted at CHBAH medical admissions ward in Soweto Gauteng province, of patients living with HIV and AIDS with liver injury. All patients admitted to the medical admissions ward were selected and their hospital numbers retrieved from the admission register. Each hospital number was entered into the NHLS labtrack system to retrieve the LFT results. Any patient with abnormal LFT was checked for their HIV status. The two variables (abnormal LFT and HIV test) were matched and confirmed. After ethics approval was obtained, files were recorded and an- !19 alysed. A data collection sheet was populated with all serological, histological and radio- logical investigations documented. Results This study included 208 patients (PLWHA) admitted to the medical admissions ward at CHBAH, with abnormal liver enzymes between January 2019 and March 2020, aged above 18 years. One hundred and five, 50.5% were males with a mean of 43.7 years, and 49.5% (n=103) were females with a mean of 39.9 years. One hundred and forty-three, 81.3% were taking antiretroviral therapy (ART), with the majority, 88.4% (n=107), on the first-line regimen. The most common pattern of liver injury was infiltrative, accounting for 67.8% (n=141), followed by mixed at 22.1% (n=46), hepatocellular, at 7.7% (n=16), and lastly was cholestatic with 2.4% (n=5). Irrespective of the pattern of the liver injury, 36.1% (n=75) of the patients had Mycobacterium tuberculosis (MTB) infection, and 22.5% (n=40) of them were on antituberculosis treatment (ATT), 6.3% (n=13) had hepatitis B viral (HBV) infection, 0.5%(n=1 ) had hepatitis C ( HCV) infection, while lymphomas (Hodgkin and non-Hodgkin lymphomas) contributed 1,9%(n=4) and 6,3% (n=13) respectively. Twelve, 5.7% (n=12) had drug induced liver Injury (DILI), and retroviral disease (RVD) cholan- giopathy contributed 1.4% (n=3). A significant number of patients, 34.1% (n=71), con- tributed to at least one-third of patients in the study, wherein the diagnosis was either un- known or not directly related to the liver injury. Those were, Other opportunistic infections accounting for 7.7% (n=16), other diagnoses made up 13% (n=27), and no diagnosis at 13.4% (n=28). Of the infiltrative pattern, 33.13% (n=47) had MTB infection, 5.6% (n=8) had HBV, 0.3% (n=1) had HCV infection, 7.8% (n=11) had non-Hodgkin lymphoma, with Hodgkin lym- phoma, and DILI both accounting for 1,4%(n=2). Only 5.7% (n=12) of PLWHA had liver biopsies done. Conclusion Liver injury is common in PLWHA. In this study, the commonest pattern of liver injury was infiltrative and the most frequent aetiology of MTB infection. Background Human immunodeficiency Virus (HIV) infection is a major global public health concern, with approximately 37,9 million people living with HIV as of 2018 (1). There has been an !20 increase in HIV prevalence globally, with the African sub-Saharan region carrying a dis- proportionate burden, accounting for more than 70% of this burden. In 2018 South Africa had 7,52 million people living with HIV/AIDS (PLWHA), with approximately 115167 AIDS- related illnesses (2). In PLWHA, liver disease and failure contribute to more significant morbidity, mortality, and higher cost of care (3). A South African study by Hoffman (4) showed that 4% of PLWHA had abnormal LFT, which was five times the upper limit of normal before antiretroviral therapy (ART). Kasper and his colleagues found that 13-18% of PLWHA died from liver-related causes (5). A Nigerian study found that 87% of their cohort had abnormal LFTs; of the hundred and twenty-nine (129) cases they reviewed, 14% showed a cholestatic pattern of liver injury, 83% were he- patocellular, and 3% mixed (6). Liver enzyme abnormalities are common in PLWHA, and these abnormalities are classified as either hepatocellular, cholestatic, or mixed/infiltrative (10). It is essential to recognize and differentiate these patterns of liver injury because they can have different etiological and pathophysiological mechanisms. This arbitrary classification is not restrictive, since several diseases and drugs can present with more than one pattern of liver injury. Howev- er, early recognition may help identify the most suggestive cause and prevent unneces- sary investigations and delay in starting appropriate therapy in PLWHA (5). Methods A retrospective cohort study at CHBAH medical admissions ward in Soweto, Gauteng Province, South Africa. The study included patients who are over the age of 18 years living with HIV and have abnormal liver enzymes from January 2019 to March 2020. Medical records of patients admitted to the medical admissions ward were reviewed for further analysis using frequency rate of approximately 14% (average from previous studies) of HIV-positive patients with abnormal liver function. Subjects for the study were selected from the CHBAH ward 20 admission register. Patients hospital numbers from admission register were retrieved. Each hospital number that was retrieved was entered into the Na- tional health laboratory services (NHLS) labtrack system to retrieve liver function test (LFT) results. Any patient with abnormal LFT was checked for their HIV status if HIV test was done. The two variables (abnormal LFT ands HIV test) were matched and confirmed. Pa- tients’ names and hospital numbers with confirmed abnormal LFT and HIV tests were recorded. After ethics approval was obtained, patients files were retrieved from the !21 CHBAH filling room. Data from these patients files were recorded and analysed. A data collection sheet (see appendix) was populated with all serological, histological and radio- logical investigations and documented. All HIV positive patients with deranged liver en- zymes irrespective of any preexisting conditions were included in the study, excluding those with normal LFT and negative HIV test. Medical records of patients admitted to the medical admissions ward were sought from the CHBAH filing room for the period between January 2019 to March 2020. Ethical approval was obtained to conduct this study from the Witwatersrand health research ethics commit- tee (HREC), ethic number M200315. Permission to conduct the study at CHBAH was also granted by the Chief executive officer, a medical advisory committee (MAC), and the head of the internal medicine department. The medical admissions ward at CHBAH is one of the busiest wards in the hospital, with approximately 90 patients being admitted via the ward daily and approximately 15-20% of these patients being retroviral disease positive. The study included 208 PLWHA over the age of 18 years. The age group of the study was between the ages of 18 and 76 years, with a mean average of 41.8 years. For this study, files were collected and reviewed for the age, gender, ART and regimen, CD4 and viral load, preexisting liver disease, comorbid disease and treatment, toxins including alcohol and herbal medications, as well as clinical presentation, clinical examination, hepatic biochemistry, serological markers, Imaging and tissue biopsy including liver biopsy. A statistician was consulted for data analysis. Quantitive methods were used to analyze the data. Stata 16.1 (StataCorp, College Station, Texas, USA) was used to get the means, standard deviations, and interquartile ranges. Results were reported in frequency and per- centage. Results In this study, we reviewed clinical files from patients who are HIV positive presenting with liver injury to the medical admissions ward at CHBAH. The demographics and baseline characteristics of these patients with abnormal liver function are depicted in table 1 and 2 below. The mean age was 43.7 in males and 39.9 in females. Table 1: Demographics of patients in a cohort Variable 95% CI Males 95% CI Females 95% !22 As shown in table 2 below, patients on ARVs made up 58% of the patients under the study and 88.4% (n=107), 10.7% (n=13) and 0.8% (n=1), were on 1st line, 2nd line and 3rd line respectively. The majority of patients on ARVs were females accounting for 55.3% (n=67) and males made up 44.6% (n=54). The mean CD4 count was 273 cells/ul, with a mean CD4 count for males 218.2 and fe- males 317.1 cells/ul, respectively. The viral load ranges from 0 copies per mil to 100 00000 copies/ml with a mean viral load of 429761.7 copies. Abnormal liver functions show a mean total bilirubin of 35.3, ALP of 226.7, GGT of 252, AST 174.2, ALT 103.2, and mean albumin of 29.72 and mean INR of 1.7. The mean of the GGT and ALP are higher than the transaminases, which may suggest obstructive or infiltrative patterns compared to he- patitic. Age(years) mean (SD) 41.8(12.4) 40.1-43.5 43.7(11.5) 41.5-45.9 39.9(13.1) 37.3-42.5 Sex n(%) 208 105(50.5%) 103(49.5%) !23 Table 2: Baseline characteristics of patients in a cohort Patterns of liver injury found during this study were infiltrative, mixed, hepatocellular, and cholestatic. The patterns of liver injury are shown in table 3 below, with the most common Variable 95% CI Males 95% CI Females 95% CI ART regimen n(%) 121(58.1%) 54(44.6%) 67(55.4%) 1st line n(%) 107(88.4%) 48(44.9%) 59(55%) 2nd line n(%) 13(10.7%) 5(38.5%) 8(61.5%) 3rd line n(%) 1(0.8%) 1(100%) 0(0%) CD4 count Mean (SD) 273.1(298.3 ) 218.5-327. 7 218.2(248.4 ) 149.0-287.3 317.0(328.1 ) 225.7-351.0 Viral load Mean (SD) 429761.7(1 352218) 234152.4-6 25370.6) 433395.1(1 369645) 155879.2-7 10911 470443.1(1 361006) 190147-750 738.9 Total bili Mean (SD) 35.3(67.5) 26.1-44.6 28.2(56.1) 17.5-38.8 42.0(76.5) 27.0-56.9 Direct bili Mean (SD) 208(27.5) 19.9-35.1 20.6(43.3) 12.4-28.8 34.0(64.8) 21.4-46.7 ALP Mean (SD) 226.7(212.3 ) 197.7-255. 8 223.2(177.6 ) 189.5-256.9 231.2(244.5 ) 183.5-279.0 GGT Mean (SD) 252.1(446.2 ) 190.9-313. 3 196.9(201.8 ) 158.5-235.4 302.7(595.8 ) 186.3-419.2 AST Mean (SD) 174.2(313.2 ) 131.3-217. 1 166.1(325.9 ) 104.0-228.3 181.3(295.2 ) 123.6-239.0 ALT Mean (SD) 103.2(248.4 ) 69.3-137.2 104.1(295.7 ) 46.5-156.8 102.4(190.0 ) 65.2-139.5 Albumin Mean (SD) 29.72(8.1) 28.6-30.9 29.2(8.0) 27.6-30.8 29.9(8.4) 28.2-31.5 INR Mean (SD) 1.7(1.7) 1.5-2.0 1.6(0.8) 1.3-1.8 1.9(1.6) 1.5-2.3 !24 pattern found in the study being infiltrative, accounting for 67.8% (n=141) of all cases. Mixed pattern made up 22.1% (n=46), hepatocellular pattern at 7.7% (n=16) of the cases and the least common was cholestatic pattern with 2.4% (n=5) of the cases. Table 3: Patterns of liver injury in a cohort Etiologies found in the study are depicted in table 4 below, with Mycobacterium tuberculo- sis (MTB) infection accounting for 36% (n=75) of all cases with abnormal liver enzymes, mycobacterium avian complex/mycobacterium other than tuberculosis, 2.8% (n=6), HBV infection, 6.2% (n=13), HCV infection, 0.4% (n=1), with non-Hodgkin lymphoma and Hodgkin lymphoma accounting for 6.2% (n=13) and 1.9% (n=4) cases respectively. Drug- induced liver injury (DILI) 5.7% (n=12), RVD cholangiopathy 1.4% (n=3), and other oppor- tunistic infections (i.e., pneumocystis jiroveci pneumonia) 7.7% (n=16) were among etiolo- gies found in this study. Other metastatic non-benign lesions 5.3% (n=11), Heart failure 4.8% (n=10), biliary masses 1.4% (n=3), Alcoholic liver disease 1.4% (n=3), Hepatocellular carcinoma 0.4% (n=1), and herbal toxicity 0.4% were also amongst etiologies ascribed by treating physicians. In 26.4% (n=55) of cases, the diagnosis was either unknown, or the treating physician ascribed no diagnosis. The liver injury patterns in relation to etiologies are also depicted in table 4, with the infiltra- tive pattern found to be the most common, irrespective of the etiology. MTB infection ac- counted for 33.3% (n=47), HBV infection at 5.6% (n=8), HCV infection at 0.7% (n-1), my- cobacterium other than tuberculosis at 4.2% (n=6) of the infiltrative pattern, with other op- portunistic infections accounting for 9.2% (n=13), non-Hodgkin lymphoma at 7.8% (n=11), Hodgkin lymphoma at 1.4% (n=2), DILI made up 1.4% (n=2), hepatocellular carcinoma at 0.7% (n=1), other metastatic non-benign lesions at 7.8% (n=11), heart failure at 6.3% Patterns Total (n) Percentage (%) Infiltrative 141 67.8 Mixed 46 22.1 Hepatitic 16 7.7 Cholestatic 5 2.4 !25 (n=9), alcoholic liver disease at 1.4% (n=2) and non-alcoholic fatty liver disease, autoim- mune hepatitis contributed 0.7% (n=1) each, of all infiltrative pattern cases. With liver injury patterns, MTB was found in 33.3% (n=47), 43.4% (n=20), 43.7% (n=7), and 20% (n=1) of the infiltrative, mixed, hepatocellular, and cholestatic patterns respective- ly. HBV infection accounted for 5.6% (n=8) of infiltrative, 12.5% (n=2) of hepatocellular, and 4.3% (n=2) of mixed patterns. Non-Hodgkin lymphoma accounted for 7.8% (n=11) of infiltrative, 2.1% (n=1) mixed, and 6.2% (n=1) hepatocellular patterns. DILI accounted for 15.2% (n=7) of mixed, 1.4% (n=2) of infiltrative, 12.5% (n=2) of hepatocellular, and 20% (n=1) of the cholestatic patterns. RVD cholangiopathy made up 40% (n=2) cholestatic and 2.1% (n=1) of mixed patterns. Of the patients with DILI, 66.6%(n=8) were on antitubercu- losis therapy (rifampicin, isoniazid, ethambutol and pyrazinamide), with 16.6%(n=2) on 1st line ART (tenofovir, lamivudine and efavirenz), whilst the remainder had no known possible drug culprits. A significant number of patients, 34.1% (n=71), contributed to at least one-third of patients in the study, wherein the diagnosis was either unknown or not directly related to the liver injury. Those were, Other opportunistic infections accounting for 7.7% (n=16), other diag- noses made up 13% (n=27), and no diagnosis at 13.4% (n=28). The typical pattern in this group of patients was also infiltrative. Other opportunistic infec- tions accounted for 9.2% (n=13) of infiltrative patterns, 12.5% (n=2) of hepatocellular pat- terns, and lastly 2.1% (n=1) of mixed patterns. Other diagnoses accounted for 14.1% (n=20) of infiltrative patterns, 13% (n=6) mixed patterns, and lastly, 6.2% (n=1) of cholesta- tic pattern. No diagnosis accounted for 12% (n=17) of infiltrative patterns, 17.4% (n=8) and 18.7% (n=3) of the cholestatic patterns. !26 Table 4: Percentages of patterns found in different etiologies Etiology Total n(%) Infiltrative pattern n(%) Mixed pattern n(%) Hepatocellul ar pattern n(%) Cholestatic pattern n(%) MTB 75(36%) 47(33.3%) 20(43.4%) 7(43.7%) 1(20%) No diagnosis 28(13.4%) 17(12%) 8(17.4%) 3(18.7%) 0(0%) Other diagnoses 27(13%) 20(14.1%) 6(13%) 1(6.2%) 0(0%) Other OI 16(7.7%) 13(9.2%) 1(2.1%) 2(12.5%) 0(0%) HBV 13(6.2%) 8(5.6%) 2(4.3%) 2(12.5%) 1(20%) NHL 13(6.2%) 11(7.8%) 1(2.1%) 1(6.2%) 00%) DILI 12(5.7%) 2(1.4%) 7(15.2%) 2(12.5%) 1(20%) Other metastatic NBL 11(5.3%) 11(7.8%) 0(0%) 0(0%) 0(0%) Heart failure 10(4.8%) 9(6.3%) 1(2.1%) 0(0%) 0(0%) MAC/MOTT 6(2.8%) 6(4.2%) 0(0%) 0(0%) 0(0%) HL 4(1.9%) 2(1.4%) 2(4.3%) 0(0%) 0(0%) RVD cholangiopathy 3(1.4%) 0(0%) 1(2.1%) 0(0%) 2(40%) Biliary masses 3(1.4%) 0(0%) 2(4.3%) 0(0%) 1(20%) ALD 3(1.4%) 2(1.4%) 1(2.1%) 0(0%) 0(0%) HCV 1(0.4%) 1(0.7%) 0(0%) 0(0%) 0(0%) NAFLD/NASH 1(0.4%) 1(0.7%) 0(0%) 0(0%) 0(0%) !27 A total of 21.6% (n=45) of biopsies were done, as depicted in Table 5 below. Fifteen, 7.2% were bone marrow biopsies, 5.8% (n=12) liver biopsies, 4.3% (n=9) lymph node biopsies, 2.4% (n=5) lymph node fine needle aspirations (FNA), 0.9% (n=2) skin biopsies, 0.5% (n=1) colon biopsy and 0.5% (n=1) biliary biopsy. These results were used to assign pos- sible etiology by treating physicians. However, apart from liver biopsy, the other biopsy re- sults had no significant impact as confirmatory to the etiology of liver injury. Table 5: Biopsy in a cohort Of the total 21.6% (n=45) of biopsies done, the liver biopsy contributed 5.8% (n=12), as shown in Table 6. Two, 16.6% of total liver biopsies done had features of steatohepatitis, with 16.6% (n=2) having features of metastatic non-benign lesions, 25.1% (n=3) with fea- tures of DILI and 16.6% (n=2) having features in keeping with RVD cholangiopathy, while 25.1% (n=3) of liver biopsies results were of unknown diagnosis Autoimmune hepatitis 1(0.4%) 1(0.7%) 0(0%) 0(0%) 0(0%) Herbal toxicity 1(0.4%) 0(0%) 1(2.1%) 0(0%) 0(0%) HCC 1(0.4%) 1(0.7%) 0(0%) 0(0%) 0(0%) Biopsy Total cases (n) Percentage (%) Bone marrow aspiration and biopsy 15 7.2 Liver biopsy 12 5.8 Lymph node biopsy 9 4.3 Lymph node FNA 5 2.4 Skin biopsy 2 0.9 Colon biopsy 1 0.5 Biliary biopsy 1 0.5 Total biopsies 45 21.6 !28 Table 6: Liver biopsy results features Discussion This study demonstrated that liver injury is common in PLWHA in both males and females. The main problem with HIV and AIDS is immunosuppression. Consequently, those infect- ed are at risk of multi-organ involvement, including liver injury from the HIV itself or various etiologies (10). Liver function tests measure synthetic, biliary, and intracellular functions. Abnormal LFTs are characterized by elevations above the upper limit of normal in ALT, AST, ALP, GGT, and bilirubin. A study in Nigeria by Ejimele et al. found that 85.5% of their cases were classified as he- patocellular patterns (6). This differs from our study, which shows that the most common pattern of liver injury is an infiltrative pattern, then a mixed pattern, followed by a hepato- cellular pattern, and lastly cholestatic pattern of liver injury. An infiltrative pattern in which the liver is invaded by a non-hepatic substance, often resulting in a liver injury similar to a cholestatic pattern with elevation in ALP and GGT, was the most common pattern in this study. A mixed pattern in which there is both cholestatic and hepatic enzyme, as well as elevation in bilirubin, was the second most common pattern of liver injury, followed by he- patocellular pattern, in which the primary injury is to the hepatocytes with elevation in ALT and AST with or without elevation in total bilirubin and lastly cholestatic pattern of injury in which injury is to the bile ducts with elevations in ALP and GGT, with or without elevations in Bilirubin and this was the least common pattern of liver injury. Price et al., 2010 indicated in their study that the spectrum of liver disease shifted in the era of ART to concomitant HCV infection, HBV infection, DILI, alcohol abuse, and NAFLD (17), while Andotsakos et al., 2020 confirmed similar liver conditions in PLWHA. However, Liver biopsy results Total (n) Percentage (%) Steatohepatitis 2 16.6 Metastatic NBL 2 16.6 DILI 3 25.1 RVD cholangiopathy 2 16.6 Unknown 3 25.1 !29 in this study, MTB infection, HBV infection, lymphomas, MOTT, and DILI were among the significant groups implicated as causes of liver injury, with a large group of either unknown diagnosis or diagnosis not directly related to liver injury, 34.1% (n=71). Jha et al., 2017 im- plicated opportunistic infections as the cause of liver injuries with 6.2% (n=9) cases of tu- berculosis infection and 5.6% (n=8) cases of HCV infection as a hepatocellular pattern in contrast to the findings of this study where MTB infection, MOTT, and HBV infections are the most common opportunistic infections found in PLWHA leading to majority infiltrative, then mixed, hepatocellular and cholestatic pattern of liver injuries (18). In patients (PLWHA) presenting to CHBAH with liver injury, it is essential to recognize the pattern of liver injury bearing in mind the most common (infiltrative) with a high index of suspicion for MTB infection as it is the most common etiology of liver injury. Though it might differ from other parts of the world, in our study, MTB was the most common etiology of liver disease, with 33.3% (n=47) cases accounting for infiltrative pattern, 43,4% (n=20) accounting for mixed pattern, 43.7% (n=7) accounting for hepatocellular pattern, and last 20% (n=1) accounting for a cholestatic pattern of liver injury. Irrespective of the pattern of liver injury, MTB infection was found to be the most common etiology for liver injury in PLWHA at CHBAH at the time of this study. This is likely due to infiltration or replacement of the liver by a granulomatous process of MTB infection. Though prior studies (13) in other parts of the world have found that drug-associated he- patotoxicity and viral hepatitides are the most common cause of liver disease in PLWHA, in this study, HBV infection accounted for only 6.2% (n=13) of all cases of liver injury. The typical pattern in HBV infection was also infiltrative. HCV infection accounted for only 0.7% (n=1) of the cases, leading to an infiltrative pattern. This also shows that infections leading to liver disease in PLWHA present with an infiltra- tive pattern of liver injury. This is important to note when presented with PLWHA with ab- normal liver enzymes at CHBAH. Drug-induced liver injury (DILI) has also been implicated in several previous studies as one of the common etiologies of liver disease in PLWHA; however, in this study, DILI only contributed 5.7% (n=12) of cases mostly presenting as a mixed pattern of liver injury. Liver injuries may be because of anti-TB drugs and ARTs. !30 Liver biopsies remain the gold standard for diagnosing and staging liver fibrosis; however, 5.7% (n=12) of liver biopsies were performed in this study. Nine, 74.9% (n=9) of the liver biopsies were diagnostic, with three, 25.1% (n=3) diagnoses unknown. None of the liver biopsies done were diagnostic of MTB infection. An audit done in the Gastroenterology department at CHBAH showed that histology was concordant with the diagnosis ascribed by the treating physician, posing the question, is it necessary? Strength of the study The study ascribes all patterns of liver injuries and recognizes the common etiology of PLWHA in the South African context where TB is endemic. Limitations Missing information due to the retrospective nature of the study. The study was conducted only at CHBAH, and most of our patients with TB and HIV are managed at the primary health care and district level. Recommendation All healthcare workers should be educated on the importance of screening all opportunistic infections and monitoring and managing liver abnormalities in PLWHA. Conclusion Liver injuries are common in PLWHA, with the infiltrative pattern being the most common and MTB being a common etiology. It is vital to screen opportunistic infections, including TB, HBV, and HCV, in PLWHA to prevent, monitor, and manage liver abnormalities to pre- vent mortality associated with the liver disease among PLWHA. !31 References 1. United Nations Programme on HIV/AIDS. Global HIV and AIDS statistics. Geneva: UNAIDS; 2018. Available from: https://www.unaids.org/sites/default/files/media_as- set/unaids-data-2018_en.pdf 2. Statistics South Africa. Statistical Release P0302: Mid-year population estimates 2018. Pretoria: Stats SA; 2018. Available from: http://www.statssa.gov.za/publica- tions/P0302/P03022018.pdf 3. Sherman KE, Rockstroh J, Thomas D. Human Immunodeficiency Virus and Liver Disease: An Update. Hepatology. 2015 Dec;62(6):1871-82. 4. Hoffmann CJ, Charalambous S, Thio CL, Martin DJ, Pemba L, Fielding KL, et al. Hepatotoxicity in an African Antiretroviral Therapy Cohort: The Effect of Tuberculo- sis and Hepatitis B. AIDS. 2007 Jun 19;21(10):1301-8. 5. Kaspar MB, Sterling RK. Mechanisms of liver disease in patients infected with HIV. BMJ Open Gastroenterol. 2017 Oct 26;4(1):e000166. 6. Ejilemele AA, Nwauche CA, Ejele OA. Pattern of abnormal liver enzymes in HIV pa- tients presenting at a Nigerian Tertiary Hospital. Niger Postgrad Med J. 2007 Dec; 14(4):306-9. 7. Ganesan M, Poluektova LY, Kharbanda KK, Osna NA. Liver as a target of human immunodeficiency virus infection. World J Gastroenterol. 2018 Nov 14;24(42):4728- 37. 8. Qin F, Jiang J, Qin C, Huang Y, Liang B, Xu Y, et al. Liver damage in patients living with HIV on antiretroviral treatment with normal baseline liver function and without HBV/HCV infection: an 11-year retrospective cohort study in Guangxi, China. BMJ Open. 2019 Apr 2;9(4):e023140. !32 https://www.unaids.org/sites/default/files/media_asset/unaids-data-2018_en.pdf https://www.unaids.org/sites/default/files/media_asset/unaids-data-2018_en.pdf http://www.statssa.gov.za/publications/P0302/P03022018.pdf http://www.statssa.gov.za/publications/P0302/P03022018.pdf 9. Cullen JM. Mechanistic classification of liver injury.Toxicol Pathol. 2005;33(1):6–8. 10. Sterling RK, Chiu S, Snider K, Nixon D. The prevalence and risk factors for abnor- mal liver enzymes in HIV-positive patients without hepatitis B or C coinfections. Dig Dis Sci. 2008 May;53(5):1375–82. 11. European Association for the Study of the Liver. EASL Clinical Practice Guidelines: Drug-induced liver injury. J Hepatol. 2019 Jun;70(6):1222-1261. 12. Jong E, Conradie F, Berhanu R, Black A, John MA, Meintjes G, et al. Consensus statement: management of drug-induced liver injury in HIV-positive patients treated for TB. S Afr J HIV Med. 2013;14(3):113-9. 13. Rockstroh J, editor. Liver disease in the HIV patient [Internet]. New York: Infectious Disease Advisor - Haymarket Media, Inc.; 2017 [cited 2020 Apr 15]. Available from: https://www.infectiousdiseaseadvisor.com/home/decision-support-in-medicine/ infectious-diseases/liver-disease-in-the-hiv-patient/ 14. Dika IE, Harding JJ, Abou-Alfa GK. Hepatocellular carcinoma in patients with HIV. Curr Opin HIV AIDS. 2017 Jan;12(1):20-5. 15. Kia L, Beattie A, Green RM. Autoimmune hepatitis in patients with human immun- odeficiency virus (HIV): Case reports of a rare, but important diagnosis with thera- peutic implications. Medicine (Baltimore). 2017 Feb;96(7):e6011. 16. Pillay B, Ramdial PK, Naidoo DP. HIV-associated Large-Vessel Vasculopathy: A Review of the Current and Emerging Clinicopathological Spectrum in Vascular Sur- gical Practice. Cardiovasc J Afr. 2015 Mar-Apr;26(2):70-81. 17. Price JC, Thio CL. Liver disease in HIV-infected individuals. Clinical gastroenterolo- gy and Hepatology, volume 8, issue 12, 2010, pages 1002-1012. 18. Shiferaw MB, Tulu KT, Zegeye AM, Wubante AA. Liver enzymes abnormalities among HAART experienced and HAART naive HIV-1 infected patients at Debre Ta- bor Hospital, North West Ethiopia: A comparative cross-sectional study. 2016 19. Jha RK, Sah SK. Prevalence and clinical spectrum of liver disease in N Nepalese HIV-seropositive patients undergoing Antiretroviral therapy: A cross-sectional hospi- tal Bases study. AIDS Res Treat. 2017. 20. Murali AR, Carey WD, editors. Liver Test Interpretation - Approach to the Patient with Liver Disease: A Guide to Commonly Used Liver Tests [Internet]. Ohio: Cleve- land Clinic – Center for Continuing Education; 2017 [cited 2020 Apr 15]. Available from: https://www.clevelandclinicmeded.com/medicalpubs/diseasemanagement/he- patology/guide-to-common-liver-tests/ !33 https://www.infectiousdiseaseadvisor.com/home/decision-support-in-medicine/infectious-diseases/liver-disease-in-the-hiv-patient/ https://www.infectiousdiseaseadvisor.com/home/decision-support-in-medicine/infectious-diseases/liver-disease-in-the-hiv-patient/ https://www.infectiousdiseaseadvisor.com/home/decision-support-in-medicine/infectious-diseases/liver-disease-in-the-hiv-patient/ https://www.clevelandclinicmeded.com/medicalpubs/diseasemanagement/hepatology/guide-to-common-liver-tests/ https://www.clevelandclinicmeded.com/medicalpubs/diseasemanagement/hepatology/guide-to-common-liver-tests/ https://www.clevelandclinicmeded.com/medicalpubs/diseasemanagement/hepatology/guide-to-common-liver-tests/ 21. Debes JD, Bohjanen PR, Boonstra A. Mechanism of accelerated liver fibrosis pro- gression during HIV infection. J Clin Transl Hepatol. 2016 Dec 28;4(4):328-35. 22. Naseer M, Dailey FE, Juboori AA, Samiullah S, Tahan V. Epidemiology, determi- nants, and management of AIDS cholangiopathy: A review. World J Gastroenterol. 2018 Feb 21;24(7):767–74. 23. Puoti M, Nasta P, Gatti F, Matti A, Prestini K, Biasi L, et al. HIV-related Liver Dis- ease: ARV Drugs, Coinfection, and Other Risk Factors. J Int Assoc Physicians AIDS Care (Chic). 2009;8(1):30-42. 24. Cai J, Osikowicz M, Sebastian G. Clinical significance of elevated liver transami- nases in HIV-infected patients. AIDS. 2019 Jul 1;33(8):1267-82. 25. Lodenyo H, Schoub B, Ally R, Kairu S, Segal I. Hepatitis B and C virus infections and liver function in AIDS patients at Chris Hani Baragwanath Hospital, Johannes- burg. East Afr Med J. 2000 Jan;77(1):13–5. 26. Ocama P, Katwere M, Piloya T, Feld J, Ohio KC, Kambugu A, et al. The spectrum of liver diseases in HIV-infected individuals at an HIV treatment clinic in Kampala, Uganda. Afr Health Sci. 2008 Mar;8(1):8-12. 27. Pacheco A, Perazzo H, Cardoso S; Fonseca MJ, Griep R, Lotufo P, et al. HIV infec- tion is an independent risk factor for liver steatosis: a study in HIV mono-infected patients compared to uninfected paired controls and associated risk factors [ab- stract]. Amsterdam: 22nd International AIDS Conference; 23-27 Jul 2018. Available from: https://onlinelibrary.wiley.com/doi/full/10.1002/jia2.25148 28. Jadeja RN, Devkar RV, Nammi S. Oxidative Stress in Liver Diseases: Pathogene- sis, Prevention, and Therapeutics. Oxid Med Cell Longev. 2017;2017:8341286. 29. Hulgan T, Gerschenson M. HIV, and Mitochondria: more than just drug toxicity. J infect Dis .2012 Jun 15; 205(12):1769-71. 30. Rodrigues EM Filho, Fernandes R, Susin R, Fior B. Immune reconstitution inflam- matory syndrome as a cause of autoimmune hepatitis and acute liver failure. Rev Bras Ter Intensiva. 2017 Jul-Sep;29(3):382-5. 31. Robinson MW, Harmon C, O'Farrelly C. Liver immunology and its role in inflamma- tion and homeostasis. Cell Mol Immunol. 2016 May;13(3):267-76. 32. Androutsakos T, Schina M, Pouliakis A, Kontos A, Sipsas N, Hatzis G. Causative factors of liver fibrosis in HIV-infected patients. A single-center study. BMC Gas- troenterology. 2020 Apr 6;20(1):91. 33. Cohen J, Torres C. HIV-associated cellular senescence: A contributor to accelerated aging. Aging Res Rev. 2017 Jul;36:117-124. !34 https://www.ncbi.nlm.nih.gov/pubmed/?term=Lodenyo%252525252525252525252525252525252520H%25252525252525252525252525252525255BAuthor%25252525252525252525252525252525255D&cauthor=true&cauthor_uid=10944831 https://www.ncbi.nlm.nih.gov/pubmed/?term=Schoub%252525252525252525252525252525252520B%25252525252525252525252525252525255BAuthor%25252525252525252525252525252525255D&cauthor=true&cauthor_uid=10944831 https://www.ncbi.nlm.nih.gov/pubmed/?term=Ally%252525252525252525252525252525252520R%25252525252525252525252525252525255BAuthor%25252525252525252525252525252525255D&cauthor=true&cauthor_uid=10944831 https://www.ncbi.nlm.nih.gov/pubmed/?term=Kairu%252525252525252525252525252525252520S%25252525252525252525252525252525255BAuthor%25252525252525252525252525252525255D&cauthor=true&cauthor_uid=10944831 https://www.ncbi.nlm.nih.gov/pubmed/?term=Segal%252525252525252525252525252525252520I%25252525252525252525252525252525255BAuthor%25252525252525252525252525252525255D&cauthor=true&cauthor_uid=10944831 https://onlinelibrary.wiley.com/doi/full/10.1002/jia2.25148 Annexure 1: Ethics approval letter !35 Annexure 2: Medical advisory committee approval letter !36 Annexure 3: Head of Department of internal medicine approval letter !37 ! Annexure 4: Data collection sheet !38 1. Data collection sheet Patient study number Age 18-25 26-35 36-45 46-55 56-65 66-75 >75 Gender M F ART Y N ART Regimen First line Second line Third line CD4 (cells/ul) Viral load (copies/ml) Risk factors: Preexisting liver disease (Y/N) Viral hepatitis NAFLD Autoimmune NBL other Comorbid disease (Y/N) !39 TB Dyslipidaemia Other Comorbid treatment (Y/N) ATT Other Alcohol Mild Moderate Severe Other toxins Traditional medications Other Clinical presentation Jaundice Pruritus Changes in stool and urine colour Abdominal distension Liver failure Incidental Other Clinical examination findings Jaundice Hepatology Portal hypertension Liver failure Other Hepatic biochemistry LFTs ALT AST ALP !40 INR Albumin Glucose Serological markers Hep BsAg P/N/U Hep C Ab P/N/U Hep E Ab P/N/U ANA P/N/U ALKM1 P/N/U Imaging (if available) Ultrasound Fatty liver Microabscesses Cirrhosis Portal HPT HCC Other CT scan abdomen Fatty liver Microabscesses Cirrhosis Portal HPT HCC Other MRCP Obstruction Dilated ducts Other Histology (if available) !41