Viral hepatitis-associated glomerulonephritis

1. What types of viral hepatitis are associated with glomerulonephritis?

Chronic carriers of hepatitis B (HBV; 400 million people worldwide) or hepatitis C (HCV; 170 million people worldwide) are at increased risk of developing a variety of glomerular diseases linked directly to active viral replication. By definition, a carrier state for these viruses would be characterized by a positive quantitative polymerase chain reaction (PCR) assay.

For HBV, the common serologic markers used for identification of a chronic carrier state includes the presence of circulating hepatitis B surface antigen (HbsAg) and in some cases hepatitis B e antigen (HbeAg) and the absence of hepatitis B surface antibody (HbsAb). For HCV, the most common initial screening test is the enzyme-linked immunosorbent assay (ELISA) for hepatitis C IgG antibodies. In stark contrast to the importance of HbsAb as an indicator of previous exposure and their absence during a chronic carrier state for HBV, the presence of HCV ELISA antibodies usually means the presence of active viremia, since this HCV antibody is non-neutralizing.

Importantly, here is a small subgroup of patients (5% to 7%) who do not mount an antibody response to HCV, yet are chronic carriers by PCR. This is especially true for patients with end-stage kidney disease; therefore if a kidney lesion is suspicious for HCV related-glomerulonephritis, then a PCR should be done even if the ELISA is negative.

2. Are the usual measurements of the glomerular filtration rate (GFR) and proteinuria applicable to patients with chronic hepatitis?

We know that the serum creatinine concentration results from the metabolism of creatine, which is a nitrogenous organic acid stored in muscles, which functions as an energy-providing catalyst. There are two sources of creatine: endogenous liver production and exogenous oral ingestion. About half the creatine comes from each source, which is extremely important when it comes to evaluating kidney function in patients with chronic hepatitis. Most patients who are chronic carriers of either hepatitis B or C have significant inflammatory injury to the liver and may even have various stages of cirrhosis. Creatine production in the liver will be significantly reduced, and this will lead to a lower serum creatinine. In the presence of cirrhosis, the synthesis of creatine can be reduced by more than 50%, and the oral intake of creatine will also be reduced as a result of malnutrition in these patients. The typical serum creatinine is usually 0.5 to 0.6 mg/dL in patients with cirrhosis, which is well below the normal for the general population (1.2 mg/dL for women and 1.5 mg/dL for men). Patients with liver disease of any cause, not just HCV or HBV, can have significant loss of GFR with a serum creatinine level within “the normal range” for the general population.

For this reason, all physicians managing patients with chronic hepatitis should be cautious in interpreting kidney function using serum creatinine and should consider monitoring kidney function with either a 24-hour creatinine clearance or an estimated GFR (eGFR). The eGFR is calculated using the Chronic Kidney Disease Epidemiology Collaboration (CKD EPI) formula, and is always included on all automated reports that have a serum creatinine.

The use of cystatin C to measure the GFR in place of the serum creatinine may be of value, especially in patients with cirrhosis, since it is not affected by muscle mass, bilirubin levels, gender, or age. However, cystatin C is affected by albumin levels and by an inflammatory state, both of which are present in cirrhotic patients. While creatinine-based formulas have overestimated GFR in cirrhotic patients, cystatin-C-based formulas have consistently underestimated the GFRs. Studies have introduced a modified CKD-EPI formula that incorporates both the serum creatinine and cystatin C levels together, and this has shown to be superior in estimating the GFR compared to either creatinine-based or cystatin-C-based formulas alone.

In addition, all patients who are chronic hepatitis carriers should be screened for proteinuria. A random urine protein/creatinine ratio or a 24-hour urine protein collection is an equally valid way to quantify the degree of proteinuria. In patients with severe jaundice, the colorimetric dipstick may give false readings for protein because of the bilirubin pigment in the urine, so every positive urine dipstick test for protein should be confirmed with a quantitative measurement.

3. What types of glomerular lesions are seen with a chronic hepatitis B carrier state?

The most common glomerular pathology seen with HBV carriers is membranous nephropathy (MN; Table 35.1 ). These patients can also develop an immune complex form of polyarteritis nodosa, membranoproliferative glomerulonephritis (MPGN), IgA nephropathy, and focal segmental glomerulosclerosis (FSGS).

4. How does HBV cause each of these types of kidney disease?

The most common lesion seen worldwide with chronic HBV is MN. This lesion is a result of the formation of immune complexes in the sub-epithelial space of the basement membrane of the glomerulus forming “spikes” on electron microscopy similar to idiopathic MN, and demonstrates the classic “granular” immunofluorescence pattern with IgG and C3. The immune complex located in the sub-epithelial space most likely represents a sequential deposition of antigen followed by attachment by antibodies (as opposed to the circulating antigen–antibody complex being deposited in the sub-epithelial space). The only antigen that consistently fits the size limitation to deposit in this area of the basement membrane is HbeAg. Consistent with that observation is the fact that the vast majority of patients with chronic HBV who develop MN are positive for the HbsAg.

For HBV carriers who are HbsAg positive but HbeAg negative, the kidney lesion is usually not MN but MPGN, because the HbsAg is too large to deposit in the sub-epithelial space and is restricted to the sub-endothelial location. The subsequent local inflammation leads to cellular proliferation and histologically appears as Type I MPGN.

IgA nephropathy may result from three separate etiologies in HBV patients:

• 1.

  Since the liver is responsible for the metabolism of immunoglobulin A (IgA), in the presence of significant liver injury, the circulating levels of IgA increase and may deposit in the kidney. These complexes first embed in the mesangium, but may also be in the peripheral capillary loops. Clinically significant kidney disease from this “secondary” deposition of IgA in the glomeruli in any patient with liver disease regardless of etiology is extremely rare.

• 2.

  IgA antibodies have been shown to co-localize with HBV antigens in the mesangium, indicating HBV can directly cause IgA nephropathy.

• 3.

  Because HBV is present worldwide, especially in Asia where IgA is extremely prevalent, it is possible that HBV and IgA can occur independent of each other in certain populations.

Polyarteritis nodosa is a necrotizing vasculitis of medium-sized vessels that is not a direct form of glomerulonephritis. Rather, it affects the inflow circulation of the kidney and, as a result of fibrinoid necrosis and thrombosis of the larger muscular arteries of the kidney, there is severe hypertension, glomerular ischemia, and progressive kidney failure. The etiology for the vasculitis is the presence of HbsAg-antibody-immune complexes in the blood vessel walls, causing local complement activation and necrosis. The presence or absence of HbeAg is not important in the genesis of this lesion.

Finally, reports have shown a higher than expected incidence of FSGS in patients with chronic HBV. Because there are no immune complexes in the kidney with FSGS, it is not clear why this pathologic entity develops. Some studies have shown that HBV may enter and deposit in kidney tissue and through unknown mechanisms lead to FSGS.

5. Can you determine the specific kidney disease without a kidney biopsy in patients with chronic HBV?

Ultimately, a kidney biopsy is needed to be certain of the diagnosis. However, among the different possibilities, both MN and FSGS are associated with the nephrotic syndrome while MPGN and IgA present with a nephritic urinary sediment. Polyarteritis nodosa (PAN) does not cause changes in the urinalysis, since the lesion is a vasculitis in larger blood vessels extrinsic to the glomerulus.

When the diagnosis is MN on biopsy, the question arises as to whether this could be idiopathic MN occurring by chance in a patient with HBV. Treatment options are contingent on which one is the cause. Patients with idiopathic MN are predominantly positive for anti-phospholipase A2 receptor antibodies compared to mostly negative results in HBV MN. However, there is enough overlap in published studies to raise a question about the predictive accuracy of this assay.

In spite of HBV-related immune complexes being present in MN, MPGN, and PAN, systemic complement activation is not seen, and C3 and C4 levels remain normal. Low complement levels should raise concern for unsuspected co-infection with HCV.