Hepatitis C virus (HCV) is a major health problem in hemodialysis patients, which leads to significant morbidity and mortality through progressive hepatic fibrosis or cirrhosis. Percutaneous liver biopsy is the gold standard to stage hepatic fibrosis. However, it is an invasive procedure with postbiopsy complications. Because uremia may significantly increase the risk of fatal and nonfatal bleeding events, the use of noninvasive means to assess the severity of hepatic fibrosis is particularly appealing to hemodialysis patients.
1. Introduction
Hepatitis C virus (HCV) infection, which may result in fibrosis, cirrhosis, hepatic decompensation, and hepatocellular carcinoma (HCC), is a leading cause of chronic liver disease in patients receiving hemodialysis
[1][2][3][4]. In addition to a solid link to liver-related morbidities, HCV infection is associated with a high risk of cardiovascular and infectious-related hospitalization and mortality in hemodialysis patients
[5]. In contrast, the health-related outcomes are significantly improved once HCV is eradicated with effective antiviral treatment
[6][7][8][9][10]. Because nearly all patients can successfully clear HCV infection with a short course of potent and safe direct-acting antivirals (DAAs), they are particularly relevant to practitioners and hemodialysis patients in moving toward HCV microelimination by 2030
[11][12][13][14][15][16][17][18].
Although the introduction of DAAs has tremendously advanced HCV care, accurate staging of hepatic fibrosis remains essential for therapeutic and prognostic purposes. The presence of cirrhosis may affect the treatment duration, the need for ribavirin (RBV) coadministration, and the sustained virologic response (SVR) rates in certain groups of patients
[19][20][21]. Furthermore, information about the severity of hepatic fibrosis can efficiently help clinicians determine the surveillance strategies for portal hypertension and HCC before and after viral cure
[22][23][24][25]. Currently, percutaneous liver biopsy is the gold standard to stage hepatic fibrosis. However, it is an invasive procedure with poor patient acceptance. Because platelet dysfunction significantly affects hemostasis in kidney failures, hemodialysis patients with HCV infection have a risk of bleeding complications ranging from 1.3% to 5.9%, which is much higher than the risk of nonfatal bleeding of 0.16% in nonuremic patients
[26][27][28][29]. In addition, the biopsy specimens are prone to sampling and interpretation variability
[30]. The use of noninvasive means to assess hepatic fibrosis in hemodialysis patients with HCV infection is appealing to healthcare providers, particularly in monitoring disease evolution over time.
2. Biochemical Index
2.1. Aspartate Transaminase (AST) to Alanine Transaminase (ALT) Ratio (AAR)
An elevated AAR has been known to suggest cirrhosis in nonuremic HCV patients, with a positive predictive value (PPV) and specificity of 100% when the cut-off value is ≥1 [31]. Ustündag et al. assessed the AAR features in 49 hemodialysis patients with HCV infection who underwent liver biopsy. They found that the AAR increased with more severe hepatic fibrosis (0.36 ± 0.17, 0.67 ± 0.17, and 0.86 ± 0.07 in patients with no fibrosis, mild fibrosis, and moderate fibrosis) [32].
Schmoyer et al. assessed the diagnostic power of the AAR in predicting significant hepatic fibrosis (≥F2), according to the METAVIR scores in hemodialysis patients with HCV infection, which revealed that the area under the receiver operating characteristic (AUROC) was only 0.59. The PPV and negative predictive value (NPV) were 27.0% and 92.3% at a cut-off value of 0.70 [33]. Because the AAR is designed to predict cirrhosis, which is seldom seen in hemodialysis patients, applying the AAR in predicting ≥F2 is of limited clinical utility.
2.2. AST-to-Platelet Ratio Index (APRI)
Wai et al. correlated various biochemical parameters with the stage of hepatic fibrosis in 270 nonuremic patients with HCV infection. They found that the levels of platelet count, AST, ALT, and alkaline phosphatase (ALP) were highly associated with patients with ≥F2 and cirrhosis (F4). A novel biochemical index, APRI, was developed by amplifying the different effects of the platelet count and AST level on hepatic fibrosis stage [34]. The AUROCs were 0.88 and 0.94 in predicting HCV patients with a fibrosis stage of ≥F2 and F4, respectively. The sensitivity, specificity, PPV, and NPV for ≥F2 were 91%. 47%, 61%, and 86% with a cut-off value of 0.5, and 41%, 95%, 88%, and 64% with a cut-off value of 1.5. In addition, the sensitivity, specificity, PPV, and NPV for F4 were 89%, 75%, 38%, and 98% with a cut-off value of 1.0, and 57%, 93%, 57%, and 98% with a cut-off value of 2.0. Using these cut-off values, the clinicians can correctly diagnose 51% and 81% of patients with a fibrosis stage of ≥F2 and F4 by the APRI without requiring liver biopsy. The APRI has been widely applied in clinical practice because it is simple, readily available, and validated in meta-analyses [35].
2.3. Fibrosis Index Based on Four Parameters (FIB-4)
FIB-4, an index that combined four biochemical parameters including age, AST, ALT, and platelet count, was initially developed to predict the severity of hepatic fibrosis in 832 patients with HCV and human immunodeficiency virus (HIV) coinfection
[36]. The AUROC in predicting a fibrosis stage of ≥F3 was 0.765. The NPV and PPV were 90% and 65% at cut-off values of <1.45 and >3.25, and 87% of patients with a FIB-4 score outside 1.45 to 3.25 can avoid fibrosis staging with liver biopsy. Subsequently, Vallet-Pichard et al. validated the performance of FIB-4 in 847 patients with HCV monoinfection, which yielded an AUROC of 0.85 in predicting a fibrosis stage of ≥F3. The NPV and PPV at cut-off values of <1.45 and >3.25 were 94.7% and 82.1%, and 72.8% patients with a fibrosis stage of <F3 or ≥F3 can be correctly classified with FIB-4
[37].
2.4. King’s Score and Fibrosis Index
In addition to the AAR, APRI, and FIB-4 indices, Schmoyer et al. applied King’s score and the Fibrosis index, which have been tested in nonuremic patients with HCV, to diagnose the severity of hepatic fibrosis in hemodialysis patients with HCV infection
[38][39]. The AUROC of King’s score in predicting a fibrosis stage of ≥F2 was 0.69, which was inferior to the AUROC of 0.79 to predict a similar stage of hepatic fibrosis in nonuremic patients. The cut-off value of a King’s score of 6.9 had an NPV of 87.3% for a fibrosis stage of F2, compared to that of 12.3 with an NPV of 77% in nonuremic patients
[33][38].
3. Serological Index
3.1. FibroTest
In early 2000, the MULTIVIRC group developed a novel index, named FibroTest, to grade the severity of hepatic fibrosis in patients with HCV infection. They combined α2 macroglobulin, haptoglobin, apolipoprotein A1, γ-glutamyl transpeptidase, total bilirubin, age, and sex into a regression model to reach a final score ranging from 0.00 to 1.00
[40]. The same group further confirmed that only 87 of 537 (16.2%) patients included in another prospective study had discordant results for fibrosis stage between FibroTest and liver biopsy. Furthermore, kidney failure did not significantly contribute to inconsistent fibrosis results
[41]. Although the diagnostic accuracy of FibroTest showed promise to assess hepatic fibrosis in HCV, a small-scaled study conducted by the same group that enrolled 50 hemodialysis patients with HCV infection, showed that the AUROCs of FibroTest were only 0.47 and 0.66 in predicting patients with a fibrosis stage of ≥F2 and ≥F3, respectively
[42].
3.2. Hyaluronic Acid (HA)
HA is a chief extracellular matrix (ECM) component and continues to deposit in the liver in response to hepatic inflammation, leading to hepatic fibrosis or cirrhosis
[43]. Serum levels of HA correlate with the severity of hepatic fibrosis in nonuremic patients with HCV infection, particularly in those with advanced liver diseases
[44][45][46][47]. A cut-off value of 60 ng/mL had NPVs of 93% and 99% for patients with ≥F3 and F4, respectively, while a cut-off value of 72 ng/mL had a PVV of 100% for those with F4
[46][47].
3.3. Tyrosine-Lysine-Leucine 40 Kilodalton (YKL-40)
YKL-40, also known as chitinase-3-like protein 1 (CHI3L1), is a glycoprotein expressed and secreted by various cells, including macrophages, chondrocytes, fibroblast-like synovial cells, hepatic stellate cells, and vascular smooth muscle cells. The serum levels of YKL-40 correlate with the severity of hepatic fibrosis of various etiologies
[48]. In nonuremic patients with HCV infection, Saitou et al. demonstrated that the AUROCs of YKL-40 were 0.809 and 0.795 in predicting a fibrotic stage of ≥F2 and F4
[49]. The PPV and NPV were 80% and 79% in predicting patients with ≥F2 hepatic fibrosis when the cut-off level of YKL-40 was 186.4 ng/mL, and were 73% and 78% in predicting patients with cirrhosis when the cut-off level was 284.8 ng/mL.
4. Radiological Index
Transient Elastography (TE, FibroScan)
TE is a noninvasive tool that assesses hepatic fibrosis by measuring liver stiffness
[50]. To date, TE has been extensively validated in nonuremic patients with HCV infection, with a diagnostic power at least equivalent to various biochemical and serological indices. The AUROCs in predicting a fibrosis stage of ≥F2, ≥F3, and F4 are 0.83, 0.90, and 0.95, respectively
[51]. The PPV and NPV are 95% and 48% at a cut-off value of 7.1 kilopascal (kPa), 87% and 81% at a cut-off value of 9.5 kPa, and 77% and 95% at a cut-off value of 12.5 kPa to predict patients with ≥F2, ≥F3, and F4. About 5% of patients, of whom most are obese, may fail to reach reliable results. Appling the XL probe may improve the diagnostic yield of TE in obese patients.
5. Clinical Application of Noninvasive Indices for Hepatic Fibrosis in Hemodialysis Patients with HCV Infection
In medical institutions where TE is readily available and accessible, directly measuring liver stiffness by this sonography-based technique can offer an excellent diagnostic yield, to predict the severity of hepatic fibrosis and avoid invasive liver biopsy in up to 90% of hemodialysis patients with HCV infection
[52]. Because liver stiffness measurement is not a synonym of liver fibrosis, patient-related confounding factors that may alter the liver stiffness, such as heart failure-induced hepatic congestion, hepatic necroinflammatory reaction, digestion state, etc., should be taken into consideration with TE
[53]. Therefore, patients are recommended to receive TE at fasting state and after hemodialysis. Magnetic resonance (MR) elastography, an advanced technique that can comprehensively assess the stiffness of the whole liver, has been shown to yield higher diagnostic accuracy than TE in nonuremic patients
[54]. Although TE has been shown to perform better than biochemical or serological indices in hemodialysis patients with HCV infection, studies aiming at the clinical performance of MR elastography are awaited in this special population.
For medical institutions where TE is unavailable, the APRI can be the choice to stage hepatic fibrosis because of its ease of use and access. Three independent studies, which adopted liver biopsy as the reference standard, have shown that the AUROCs of the APRI were 0.80 to 0.84 in predicting a fibrosis stage of ≥F2 and ≥F3. An APRI of 0.40 and 0.95 can have NPVs of 85% to 93% and PPVs of 66% to 82% to predict F2
[55][56][57]. Because the percentage of hemodialysis patients with a fibrosis stage of ≥F3 is limited, the clinical value of the APRI would be more focused on the high NPV for F3. Current data indicate that an APRI of <0.55 can exclude a fibrosis stage of ≥F3 in 99% of hemodialysis patients with HCV infection
[55].
Although the FIB-4 index has superior diagnostic accuracy to the APRI to predict the severity of hepatic fibrosis in nonuremic HCV patients, the performance of FIB-4 in hemodialysis patients with HCV infection is not ideal or stable
[33][58][59]. Furthermore, it does not recommend King’s score or the Fibrosis index in these patients, based on the poor diagnostic performance
[33].
Because most hemodialysis patients present with systemic inflammation/fibrosis from nonhepatic origins, all serological indices targeting ECM dynamics, including FibroTest, hyaluronic acid, and YKL-40, are not recommended in clinical practice to predict the stage of hepatic fibrosis in hemodialysis patients with HCV infection.
To date, data regarding the application to monitor the evolution of fibrotic changes following antiviral treatment are scarce. In nonuremic patients with HCV who achieve SVR with antiviral therapy, studies have shown that the APRI, FIB-4, and TE had low diagnostic accuracies in assessing the evolution of hepatic fibrosis
[60][61]. Current evidence does not favor the use of the APRI to monitor the evolution of hepatic fibrosis because the diagnostic accuracy of the APRI in hemodialysis patients with HCV infection seemed to decrease once SVR was achieved with antiviral therapy, compared to the pretreatment status
[56]. The clinical performance of TE or MR elastography to follow the evolution of hepatic fibrosis needs further investigation.
This entry is adapted from the peer-reviewed paper 10.3390/diagnostics12102282