Liver disease is one of the main causes of mortality in HIV-infected individuals in regions where antiretroviral therapy is widely available . The high rate of chronic hepatitis B and C in this population and the accelerated progression to liver cirrhosis characteristically seen in coinfected patients [2,3] largely explain this observation. The hepatotoxicity of several medications, including antiretroviral agents, also contributes to liver-related morbidity and mortality . On the other hand, frequent comorbidities in HIV-infected patients such as elevated alcohol consumption, insulin resistance and dislipidemias along with some antiretroviral medications may account for a relatively high rate of steatosis and steatohepatitis . Finally, HIV itself  or the consequences of intense microbial translocation from the gut to the portal tract in HIV infection [7,8] might further contribute to the increased liver disease burden seen in HIV-infected patients.
Severe liver disease in the absence of any recognizable cause has been reported in HIV-infected individuals and has mainly been grouped into two categories. First, patients presenting with persistently elevated liver enzymes, in whom liver biopsies have generally found fatty liver disease as the main reason for liver function tests abnormalities, and in whom significant liver fibrosis and cirrhosis may result from steatohepatitis . Less commonly, a second group of patients presents with severe portal hypertension, with clinical (gastrointestinal bleeding), laboratory (pancytopenia due to hypersplenism) and/or endoscopic (esophageal varices) manifestations [10–15]. Liver biopsy in these patients has revealed a variety of lesions, of which the most common is nodular regenerative hyperplasia [11,14] and hepatoportal sclerosis , almost always in the absence of advanced liver fibrosis or cirrhosis. Herein, we describe a relatively large series of cases of unexplained noncirrhotic portal hypertension (NCPH) in HIV-infected individuals in whom characteristic histological lesions in the liver and prior exposure to didanosine concur to define a new condition.
Patients and methods
This is a description of the main clinical and pathological findings of all consecutive cases of NCPH seen in HIV-positive individuals at one referral HIV clinic located in Madrid, Spain. All patients had HIV-1 infection with portal hypertension without liver cirrhosis. Although most of them presented with clinical events (variceal bleeding, portal thrombosis, ascites, etc.), a few had just unexplained persistent elevated liver enzymes with signs of hypersplenism. All underwent percutaneous liver biopsy, after giving signed informed consent, using 16–18-gauge needles. Liver sections from formalin-fixed, paraffin-embedded tissues were stained with hematoxilin–eosin, Masson trichrome, Wilder's reticulin, periodic acid-Schiff (PAS) with diastase and Perls coloration. All sections were evaluated by at least two pathologists, one of whom examined all but three specimens. In addition to conventional examinations, a careful examination of portal veins within portal tracts was carried out in all liver sections.
Routine laboratory screens for various etiologies of liver disease and for hypercoagulable disorders at the time of diagnosis included serum HCV-RNA, hepatitis B virus (HBV)-DNA, hepatitis E virus (HEV)-RNA, antinuclear antibody, antismooth muscle antibody, iron, total iron binding capacity, alpha-1 antitrypsin, ceruloplasmin, protein C activity, protein S activity, antithrombin III, anticardiolipin antibody, antiphospholipid antibody and factor V (R506Q) Leiden mutation. Case histories and comprehensive longitudinal medication lists were reviewed for each patient.
A total of 12 HIV-infected patients were diagnosed with unexplained NCPH over the past 5 years from our outpatient clinic of approximately 2600 HIV-infected individuals, giving an overall estimated rate of 0.45%. Any other underlying liver diseases were excluded. Conditions specifically investigated were hemochromatosis, alpha-1 antitripsin deficiency, autoimmune hepatitis, inherited procoagulation disorders, Wilson disease, chronic hepatitis B or C infections (including occult or seronegative infections) and alcohol abuse. Intake of medications with potential liver involvement was similarly checked and excluded such as antidepressants, opioids, antituberculous, antiepileptics and so on.
The main characteristics of the study population are recorded in Table 1. The majority of patients were white (11 of 12) and men (10 of 12); all of whom but one were men who have sex with men. The median age at diagnosis was 47 (range 28–67) years. The median time since HIV diagnosis was 11 years. All but one were on antiretroviral therapy and all treated patients had undetectable plasma HIV-RNA (<50 copies/ml). One patient had been on therapy for several years but was currently off antiretroviral medications. Three patients had current CD4 cell counts below 200 cells/μl, though nine had had past nadir CD4 cell counts below 200 cells/μl. All patients had been treated with didanosine, and median time of exposure was 53 months (range 14–43 months).
Cirrhosis was excluded in all 11 patients in whom a liver biopsy was performed. In one of them, however, pieces were too fragmented to provide valuable material for interpretation. Liver stiffness assessment in this patient, however, excluded cirrhosis. In the remainder, three had nodular regenerative hyperplasia, whereas the other seven had morphological features of hepatoportal sclerosis. The most frequent histological finding in these individuals was the paucity of portal small veins in more than one-third of portal tracts (Fig. 1c). Occasionally, the portal veins were replaced by small thin-walled slit-like channels (re-cannalization) (Fig. 1d). In four cases, we identified a lesion of the small portal veins characterized by fibrous obliteration with marked thickening of the small portal vein wall, along with partial or total occlusion of the lumen (sclerosing portal venopathy) (Fig. 1b). The median diameter of the affected portal veins was 100 μm. Complete absence of portal vein branches was recognized in multiple portal tracts (more than 60% of them in four cases). Focal dilatation of sinusoids and portal fibrosis were frequent features. Mild-to-moderate degrees of steatosis were noted in the adjacent liver parenchyma in nearly one-third of patients, but features of severe steatosis or steatohepatitis were absent in all instances. Although areas of hepatocyte regeneration were recognized in many cases, clear features of nodular regenerative hyperplasia were found in only three patients.
All patients but two had persistently mild-to-moderate elevations in liver enzymes for at least 2 years before diagnosis; in the other two patients, aspartate aminotransferase/alanine aminotransferase (AST/ALT) was intermittently elevated. Following diagnosis and after a median follow-up of 31 months (range, 19–57 months), all patients remained alive. However, portal vein thrombosis developed in six patients. All individuals who experienced variceal bleeding underwent variceal band ligation procedures and β-blocker therapy was initiated, re-bleeding occurring in one patient. Transjugular intrahepatic portal shunting (TIPS) was performed in three patients and it is currently planned for another patient.
We have described a series of HIV-infected patients with severe portal hypertension in the absence of significant impairment in the hepatic synthetic function. The majority of these individuals showed unique features in the liver biopsy, the most remarkable of which was massive absence of portal veins and focal sclerosis of the small portal vein branches. Prior prolonged exposure to didanosine was uniformly recognized in all cases. We hypothesize that a primary injury of the portal vessel endothelium by this adenosine analogue could play a central pathogenic role in this condition.
Cases of NCPH have been reported in HIV-negative individuals as a result of exposure to chemotherapeutic agents, mainly purine analogues (e.g., azathioprine, 6-mercaptopurine or 6-thioguanine) [16–19], bacterial infections, genetic coagulation disorders and/or autoimmune diseases [20,21]. Nodular regenerative hyperplasia has been the most common histological finding in cases associated with thiopurine treatment . Interestingly, histopathological changes often appear after months to years of drug exposure and are dose-dependent . The primary damage seems to occur in the liver vasculature, involving the central venules at an early stage. Thereafter, impairment of vascular flow may lead to diffuse hepatocyte hyperplasia and nodule formation.
Exposure to abacavir and didanosine, both of which are purine analogues, has recently been shown to be associated with increased cardiovascular risk in HIV-infected patients . These nucleoside analogues may cause systemic endothelial damage throughout still unclear mechanisms. An enhancement of pro-inflammatory mediators has been reported in patients taking these drugs , which might create a pro-thrombotic state and precipitate cardiovascular events . An alternative hypothesis has proposed that purine analogues could interfere with the pro-inflammatory signaling molecules adenosine triphosphate (ATP) and diphosphate (ADP) present in vascular endothelial cells, leading to vascular damage . In our patients with NCPH, we hypothesize that inflammatory changes in the small portal vessels could have occurred as first step, preceding the fibrotic wall lesions we recognized at the time biopsies were made. Interestingly, although all our patients had been exposed to didanosine, generally for long periods, only four of them had received abacavir.
Thrombophilic abnormalities have been reported in some HIV-infected individuals with NCPH, notably protein S deficiency [14,27], the most common underlying abnormality. However, none of our patients showed any significant coagulation alteration, which may predispose to portal vessel thrombosis. A role for repeated episodes of pylephlebitis due to microbial translocation from the gut must also be considered. However, vascular hepatic damage due to bacterial infections tends to affect larger vessels instead of the endothelium of the portal veins. Alternatively, inflammatory phenomena within the portal circulation could be enhanced due to the increased microbial translocation characteristically seen in HIV-infected persons. Although evidence for this observation derives from measuring DNA sequences encoding bacterial ribosomal 16S RNA (16S rDNA) and lipopolysaccharides in the bloodstream, it is reasonable to believe that concentration of bacterial gut products may be even higher in the portal vasculature .
HIV-infected patients with NCPH may experience potentially life-threatening gastrointestinal bleeding episodes. Therefore, early diagnosis of this condition is important, as preventive measures can be undertaken, especially assessing the presence of esophageal varices and performing primary prevention of bleeding. Moreover, given that portal vein thrombosis is a frequent complication, the benefit of anticoagulation warrants further investigation. It should, however, be used cautiously and may be limited to patients with portal thrombosis with a minimal risk of variceal bleeding.
It must be noted that the characteristic obstructive portal venopathy identified in these HIV-infected patients might also appear superimposed to liver damage resulting from other conditions. In that situation, its recognition might be rather difficult. Given that alcohol abuse, fatty liver disease and especially chronic hepatitis C affect a substantial proportion of the HIV population [29,30], the condition we are reporting here as didanosine-associated portal damage should be suspected when clinical, laboratory and endoscopic signs of severe portal hypertension appear in patients in whom no or only mild liver parenchymal damage is evident .
In summary, we report here a series of cases of NCPH in HIV-infected individuals, in whom prior prolonged exposure to didanosine is a common finding and in whom obliteration of portal veins is the most distinctive histological feature in the liver. On 29 January 2010, the US Food and Drug Administration updated the labeling for didanosine incorporating this information. Although the proven antiviral effect of the drug may outweigh its potential side effects, it is clear that when other drug options are available, it may be worthwhile to consider them at front.
The present work was funded in part by grants from Fundación Investigación y Educación en SIDA (IES), Red de Investigación en SIDA (RIS, project ISCIII-RETIC RD06/006) and the European NEAT project. We would like to thank Dr Marion Peters (UCSF, San Francisco, California, USA) for critical reading and suggestions.
E.V., I.M. and V.S. designed the study. A.M. and A.C. were the pathologists who interpreted the liver biopsies. E.V., I.M., P.B., S.A. and V.S. were the clinicians who identified the patients and recorded the clinical and demographic information. E.V., I.M. and V.S. wrote the manuscript. All authors revised the current submission and their suggestions were discussed and incorporated before final submission, which was approved by all authors.
1. Weber R, Sabin C, Friis-Møller N, Reiss P, El-Sadr WM, Kirk O, et al
. Liver-related deaths in persons infected with the HIV: the D:A:D study. Arch Intern Med 2006; 166:1632–1641.
2. Thio C, Seaberg E, Skolasky R, Phair J, Visscher B, Muñoz A, et al
. HIV-1, hepatitis B virus, and risk of liver-related mortality in the Multicenter AIDS Cohort study (MACS). Lancet 2002; 360:1921–1926.
3. Benhamou Y, Bochet M, Di Martino V, Charlotte F, Azria F, Coutellier A, et al
. Liver fibrosis progression in HIV and hepatitis C virus coinfected patients. Hepatology 1999; 30:1054–1058.
4. Soriano V, Puoti M, Garcia-Gasco P, Rockstroh JK, Benhamou Y, Barreiro P, McGovern B. Antiretroviral drugs and liver injury. AIDS 2008; 22:1–13.
5. Sulkowski M, Mehta S, Torbenson M, Afdhal NH, Mirel L, Moore RD, Thomas DL. Hepatic steatosis and antiretroviral drug use among adults coinfected with HIV and hepatitis C virus. AIDS 2005; 19:585–592.
6. Blackard J, Sherman K. HCV/HIV co-infection: time to re-evaluate the role of HIV in the liver? J Viral Hepat 2008; 15:323–330.
7. Brenchley J, Price D, Schacker T, Asher TE, Silvestri G, Rao S, et al
. Microbial translocation is a cause of systemic immune activation in chronic HIV infection. Nat Med 2006; 12:1365–1371.
8. Balagopal A, Philp F, Astemborski J, Block TM, Mehta A, Long R, et al
. HIV-related microbial translocation and progression of hepatitis C. Gastroenterology 2008; 135:226–233.
9. Ingiliz P, Valantin M, Duvivier C, Medja F, Dominguez S, Charlotte F, et al
. Liver damage underlying unexplained transaminase elevation in HIV-1 monoinfected patients on antiretroviral therapy. Hepatology 2009; 49:436–442.
10. Maida I, Nuñez M, Rios MJ, Martín-Carbonero L, Sotgiu G, Toro C, et al
. Severe liver disease associated with prolonged exposure to antiretroviral drugs. J Acquir Immune Defic Syndr 2006; 42:177–182.
11. Mallet V, Blanchard P, Verkarre V, Vallet-Pichard A, Fontaine H, Lascoux-Combe C, Pol S. Nodular regenerative hyperplasia is a new cause of chronic liver disease in HIV-infected patients. AIDS 2007; 21:187–192.
12. Schiano T, Kotler D, Ferran E, Fiel MI. Hepatoportal sclerosis as a cause of noncirrhotic portal hypertension in patients with HIV. Am J Gastroenterol 2007; 102:2536–2540.
13. Maida I, García-Gascó P, Sotgiu G, Rios MJ, Vispo ME, Martin-Carbonero L, et al
. Antiretroviral-associated portal hypertension: a new clinical condition? Prevalence, predictors and outcome. Antivir Ther 2008; 13:103–107.
14. Saifee S, Joelson D, Braude J, Shrestha R, Johnson M, Sellers M, et al
. Noncirrhotic portal hypertension in patients with HIV infection. Clin Gastroenterol Hepatol 2008; 6:1167–1169.
15. Kovari H, Ledergerber B, Peter U, Flepp M, Jost J, Schmid P, et al
. Association of noncirrhotic portal hypertension in HIV-infected persons and antiretroviral therapy with didansoine: a nested case-control study. Clin Infect Dis 2009; 49:626–635.
16. Shastri S, Dubinsky M, Poordad F, Vasiliauskas EA, Geller SA. Early nodular hyperplasia of the liver occurring with inflammatory bowel diseases in association with thioguanine therapy. Arch Pathol Lab Med 2004; 128:49–53.
17. Seiderer J, Zech C, Diebold J, Schoenberg SO, Brand S, Tillack C, et al
. Nodular regenerative hyperplasia: a reversible entity associated with azathioprine therapy. Eur J Gastroenterol Hepatol 2006; 18:553–555.
18. Gisbert J, González-Lama Y, Maté J. Thiopurine-induced liver injury in patients with inflammatory bowel disease: a systematic review. Am J Gastroenterol 2007; 102:1518–1527.
19. de Boer N, van Bodergraven A, Jharap B, de Graaf P, Mulder C. Drug insight: pharmacology and toxicity of thiopurine therapy in patients with IBD. Nat Clin Pract Gastroenterol Hepatol 2007; 4:686–694.
20. Nakanuma Y, Hoso M, Sasaki M, Terada T, Katayanagi K, Nonomura A, et al
. Histopathology of the liver in noncirrhotic portal hypertension of unknown aetiology. Histopathology 1996; 28:195–204.
21. Hillaire S, Bonte E, Denninger M, Casadevall N, Cadranel JF, Lebrec D, et al
. Idiopathic noncirrhotic intrahepatic portal hypertension in the west: a re-evaluation in 28 patients. Gut 2002; 51:275–280.
22. Vernier-Massouille G, Cosnes J, Lemann M, Marteau P, Reinisch W, Laharie D, et al
. Nodular regenerative hyperplasia in patients with inflammatory bowel disease treated with azathioprine. Gut 2007; 56:1404–1409.
23. D.A.D. Study Group, Sabin CA, Worm SW, Weber R, Reiss P, El-Sadr W, Dabis F, et al. Use of nucleoside reverse transcriptase inhibitors and risk of myocardial infarction in HIV-infected patients enrolled in the D:A:D study: a multicohort collaboration
24. Kuller L, Tracy R, Belloso W, De Wit S, Drummond F, Lane HC, et al
. Inflammatory and coagulation biomarkers and mortality in patients with HIV infection. PLoS Med 2008; 5:e203.
25. SMART/INSIGHT; DAD Study Groups. Use of nucleoside reverse transcriptase inhibitors and risk of myocardial infarction in HIV-infected patients
26. Goicoechea M, McCutchan A. Abacavir and increased risk of myocardial infarction. Lancet 2008; 372:803–804.
27. Mallet V, Varthaman A, Lasne D, Viard JP, Gouya H, Borgel D, et al
. HIV-associated nodular regenerative hyperplasia is linked to portal obstructive venopathy and autoimmune protein S deficiency. Hepatology 2007; 46(Suppl):576A.
28. Jiang W, Lederman M, Hunt P, Sieg SF, Haley K, Rodriguez B, et al
. Plasma levels of bacterial DNA correlate with immune activation and the magnitude of immune restoration in persons with antiretroviral-treated HIV infection. J Infect Dis 2009; 199:1177–1185.
29. Koziel M, Peters M. Viral hepatitis in HIV infection. N Engl J Med 2007; 356:1445–1454.
30. Soriano V, Puoti M, Sulkowski M, Cargnel A, Benhamou Y, Peters M, et al
. Care of patients coinfected with HIV and hepatitis C virus: 2007 updated recommendations from the HCV-HIV International Panel. AIDS 2007; 21:1073–1089.
31. Vispo E, Maida I, Barreiro P, Moreno V, Soriano V. Upper gastrointestinal bleeding may unmask didanosine-associated portal hepatopathy in HIV/HCV co-infected patients. HIV Clin Trials 2008; 9:440–444.