Spectrum of Kidney Diseases in Patients With Hepatitis C Virus Infection

A 10-Year Study

Shunhua Guo, MD; Meghan E. Kapp, MD; Diego M. Beltran, MD; Cesar Y. Cardona, MD; Dawn J. Caster, MD; Ronald R. Reichel, MD; Agnes B. Fogo, MD

Disclosures

Am J Clin Pathol. 2021;156(3):399-408. 

In This Article

Results

HCV+ Patients in the 10-Year Native Kidney Biopsy Specimens

There were 9,836 native kidney biopsy specimens from January 2007 to December 2016. Among them, 273 (2.8%) patients had a history of HCV+ Table 1. There were 101 (2.6%) HCV+ patients among 3,854 native kidney biopsy specimens from 2007 to 2011 and 172 (2.9%) HCV+ patients among 5,982 native kidney biopsy specimens from 2012 to 2016. The mean age of these patients was 54 years, with 186 male and 87 female patients. Ethnicity was known in 195 patients, including 102 Black and 93 White patients.

Renal Diseases in HCV+ Patients

The kidney diseases in the HCV+ patients were categorized into three groups: HCV-associated GN, other immune complex–mediated glomerular diseases, and non–immune complex–mediated kidney diseases Table 2. Among the 273 HCV+ patients from 2007 to 2016, 115 (42.1%) patients had kidney disease consistent with HCV-associated GN. Non–HCV-associated kidney diseases comprised most diagnoses (158 cases, 57.9%), including non–immune complex–mediated kidney diseases (127 cases, 46.5%) and other immune complex–mediated glomerular diseases (31 cases, 11.4%). Forty-one (40.6%) patients had HCV-associated GN among 101 HCV+ patients from 2007 to 2011 in comparison to 74 (43.0%) patients with HCV-associated GN among 172 HCV+ patients from 2012 to 2016. Comparisons between the 2007 to 2011 and 2012 to 2016 periods did not show statistically significant differences in the frequency of these groups of disease.

Five morphologic patterns of HCV-associated GN were observed: focal proliferative, diffuse mesangial proliferative, diffuse MPGN, focal or diffuse proliferative GN with crescentic lesions, and membranous patterns Table 3. Among these patterns, the most common were diffuse mesangial proliferative (50.4%) and MPGN patterns (28.7%). More biopsy specimens showed a diffuse mesangial proliferative pattern (58.1%) and fewer showed the MPGN pattern (18.9%) in the 2012 to 2016 period than the 2007 to 2011 period (36.6% and 46.3%, P = .03 and P = .01, respectively).

Cryo plugs and deposits with substructure with short fibrillary or short microtubular features are characteristic of cryoglobulinemic GN Table 4. There were more cases with cryo plugs in 2007 to 2011 (26.8%) vs 2012 to 2016 (8.1%) (P = .01). Cryoglobulinemia was not routinely tested or reported in the available clinical data. For the patients in whom the test was reported, 14 (77.8%) of 18 cases were positive in 2007 to 2011 vs 8 (36.4%) of 22 cases in 2012 to 2016 (P = .01). Characteristic substructure by EM was present in 56 (48.7%) cases of the 115 HCV-associated GN cases, including 23 (56.1%) of the 41 cases in 2007 to 2011 vs 33 (44.6%) of the 74 cases in 2012 to 2016 (P = .24).

To assess chronicity features in the kidney biopsy specimens of HCV-associated GN, global glomerulosclerosis and interstitial fibrosis were evaluated Table 5. Global glomerulosclerosis extent was similar in 2007 to 2011 (21.8% ± 21.3%) vs 2012 to 2016 (24.9% ± 21.4%) (P = .38). Interstitial fibrosis was numerically lower in 2007 to 2011 (22.0% ± 15.9%) vs 2012 to 2016 (28.8% ± 18.9%), but the difference was not statistically significant (P = .08).

Pathologic features of HCV-associated GN with focal proliferative, diffuse mesangial proliferative, or diffuse MPGN patterns were analyzed together Table 6. Again, more patients with cryo plugs were found in the biopsy specimens with these patterns in 2007 to 2011 (31.4%) than in 2012 to 2016 (8.0%) (P = .02). IgM-dominant IF staining was similar in the two eras. EM demonstrated that deposits often coexisted in several locations in the same case. The most common locations were mesangial (88.7%) and subendothelial (63.9%). Focal subepithelial deposits were not uncommonly found (26.8%) in the patients with proliferative patterns, in addition to being seen in membranous pattern cases, which by definition showed dominant subepithelial deposits. The extent of mesangial and intramembranous deposits was similar in the two eras. In contrast, there were more subendothelial deposits in 2007 to 2011 (77.1%) than in 2012 to 2016 (56.4%) (P = .04), and there were more subepithelial deposits in 2012 to 2016 (37.1%) than in 2007 to 2011 (8.6%) (P = .01). Characteristic substructure of deposits found by EM was only numerically more frequent in 2007 to 2011 (62.9%) than in 2012 to 2016 (48.4%) (P = .17). One case in 2013 with a diffuse mesangial proliferative pattern showed dominant IgA deposits with 2+ granular diffuse mesangial and segmental capillary loop staining for IgA and C3, as well as trace to 1+ focal segmental granular mesangial and capillary loop staining for IgM, κ and λ light chains. EM showed intramembranous and subendothelial deposits and occasional mesangial deposits, which demonstrated short fibrillary and short microtubular substructures. Therefore, the biopsy specimen was classified as HCV-associated GN rather than primary IgA nephropathy.

Nine cases showed crescentic lesions with focal or diffuse mesangial proliferative or membranoproliferative patterns: 3 (7.3%) of 41 HCV-associated GN cases in 2007 to 2011 and 6 (8.1%) of 74 HCV-associated GN cases in 2012 to 2016 Table 7. The crescentic lesions were focal (combined cellular, fibrocellular, and fibrous crescents involving less than half of total glomeruli) in seven cases and diffuse (combined crescents involving equal to or over half of total glomeruli) in two cases. Six of these cases with crescents also showed focal fibrinoid necrosis of the glomerular tuft. IF showed IgM dominance in seven of the cases with crescents, with one case showing IgG dominance and one case with C3 dominance. The case with C3 dominance showed 2+ mesangial and segmental capillary loop C3, trace mesangial IgG, trace mesangial and segmental capillary loop IgM staining by IF with diffuse mesangial proliferation and segmental endocapillary proliferation, diffuse crescents (7/13 glomeruli) and focal fibrinoid necrosis by LM, and scattered intramembranous, subendothelial, and mesangial deposits that showed short fibrillary substructure by EM. ANCA serologies were negative in this patient. With these features, this case was classified as HCV-associated GN rather than C3 glomerulopathy or pauci-immune crescentic GN. Characteristic substructure of deposits was revealed by EM in seven (77.8%) of these cases with crescents. ANCA serologies were available in three cases, two of which were negative and one was positive (see Discussion). Three of these crescentic cases had cryoglobulinemia results reported with two positive. Four of these cases had reported serum complement C3 and C4 levels, and all of them were low.

Nine cases of HCV-associated GN showed a membranous pattern: 3 (7.3%) of 41 in 2007 to 2011 and 6 (8.1%) of 74 in 2012 to 2016 Table 8. Most were accompanied by focal to diffuse mesangial proliferation with dominant IgM staining by IF. In addition to subepithelial deposits, eight cases showed mesangial deposits, six showed subendothelial deposits, and seven showed intramembranous deposits by EM. Three of these cases showed substructure of deposits by EM. IF staining for PLA2R was performed on the four cases from 2014 with one positive and three negative. For the five cases before 2014, three cases had the frozen tissue available for PLA2R IF staining, which showed two positive and one negative. One case with positive PLA2R staining showed diffuse global capillary loop and segmental mesangial granular staining for IgG, IgA, and IgM by IF, as well as scattered mesangial and subendothelial deposits in addition to subepithelial deposits by EM with short fibrillary and microtubular substructure. The second PLA2R-positive case showed diffuse global mesangial and segmental capillary loop granular staining for IgG and IgM by IF and subepithelial, intramembranous, and mesangial deposits by EM. The third PLA2R-positive case showed a fibrocellular crescent and fibrinoid necrosis in one of nine glomeruli, diffuse granular capillary loop staining for IgG and mesangial staining for IgM by IF, and frequent subepithelial and scattered subendothelial and mesangial deposits by EM. ANCA test result was not available in this patient. These features suggested a secondary etiology of the membranous pattern GN.

Kidney diseases other than HCV-associated GN in HCV+ patients were divided into other immune complex–mediated glomerular diseases and non–immune complex–mediated kidney diseases. The former includes IgA nephropathy, idiopathic membranous GN, postinfectious GN, lupus nephritis, pauci-immune crescentic GN, and fibrillary GN Table 9. Five cases had changes consistent with idiopathic membranous GN with characteristic LM and IF findings, and EM showed only subepithelial deposits with no features to suggest a secondary etiology. PLA2R stain was performed on four cases with tissue available, with two positive. Fibrillary GN was diagnosed in 3 (3.0%) of 101 HCV+ patients in 2007 to 2011 and 6 (3.5%) of 172 HCV+ patients in 2012 to 2016.

The non–immune complex–mediated kidney diseases Table 10 predominantly comprised diabetic nephropathy (19.8%) and arterionephrosclerosis (12.5%). Other diseases included obstruction/reflux nephropathy, acute or chronic interstitial nephritis, HIV-associated nephropathy, acute tubular injury, light chain cast nephropathy or light chain deposition disease, thrombotic microangiopathy, focal segmental glomerulosclerosis (tip lesion type or not otherwise specified type), and others.

Clinical Data

Many cases had only limited clinical information for analysis (Supplemental Table 1; all supplemental materials can be found at American Journal of Clinical Pathology online). The available data showed higher serum creatinine at biopsy of HCV-associated GN in the 2012 to 2016 group than in the 2007 to 2011 group (P = .04), while proteinuria was not statistically different (P = .25). Serum creatinine and proteinuria in the patients with other immune complex–mediated glomerular diseases and non–immune complex–mediated kidney diseases did not differ significantly in 2007 to 2011 vs 2012 to 2016. The HCV-associated GN group numerically had more proteinuria than the other immune complex–mediated diseases and the non–immune complex–mediated kidney diseases groups, but the differences were not statistically significant (P = .39 and P = .06, respectively). Serum creatinine levels at biopsy of the three groups were not statistically different. Most cases had hematuria, but its severity was usually not reported.

Cryoglobulinemia testing results were available in 41 HCV+ patients, among which 23 (56.1%) were positive. Twenty-two (95.7%) showed HCV-associated GN. One case with positive cryoglobulinemia showed diabetic nephropathy, with IF showing only weak nonspecific staining in sclerosed glomeruli and EM showing no deposits. Positivity for cryoglobulinemia did not necessarily correlate with the presence of cryo plugs seen by LM, IF, and EM or vice versa: of the 23 cryoglobulinemia-positive patients, only 7 (30.4%) had cryo plugs identified; of the 17 cases with cryo plugs identified, only 7 were reported cryoglobulinemia test positive, 1 reported negative, and 9 were not reported.

Sixty cases had follow-up information either from electronic medical records or returned follow-up forms (Supplemental Table 2). The average clinical follow-up duration was 3 years. Fourteen patients were deceased without cause of death detailed. Twenty-seven (45%) patients reached end-stage kidney disease, with 25 (42%) on dialysis or with kidney transplant. Serum creatinine in the patients not on dialysis or transplant was only numerically lower in the HCV-associated GN group (mean, 3.2 mg/dL) than in the non–HCV-associated kidney disease group (mean, 4.3 mg/dL) (P value not significant). The two groups had a similar mean HCV RNA titer but with marked variability. Eleven of the 21 patients with HCV-associated GN with follow-up information had known antiviral therapy, and in 6 (54.5%), HCV RNA became negative; 1 showed no response to therapy (interferon with ribavirin, then telaprevir-boceprevir); 2 patients could not finish the treatment due to side effects; and 2 patients had an unknown posttherapy HCV result. Twelve of the 39 patients with non–HCV-associated kidney diseases had known antiviral therapy, 7 (58.3%) had HCV RNA turn negative after treatment, and 5 patients had unknown posttherapy HCV results. Only two cases with antiviral therapy had both pre- and posttherapy proteinuria/renal function results available, inadequate to assess the specific effect of HCV therapy on proteinuria or renal function.

processing....