IgA Nephropathy and IgA Vasculitis (Henoch-Schönlein Purpura) Nephritis
Mark Haas
HISTORICAL BACKGROUND
IgA nephropathy first became recognized as a distinct entity in 1968, when Berger and Hinglais (1) reported a cohort of patients with persistent microscopic hematuria, episodes of macroscopic hematuria in some that were often associated with a sore throat, mild to moderate proteinuria without the nephrotic syndrome, and normal renal function in most. Renal biopsies from these patients showed varying histologic features ranging from normal to chronic glomerulonephritis, but most often focal glomerulonephritis, without typical features of acute postinfectious (poststreptococcal) glomerulonephritis. However, immunofluorescence microscopy in each case showed mesangial deposition of IgA, usually with accompanying but less intense staining for IgG and C3, and electron microscopy (EM) confirmed the presence of mesangial immune complex deposits (1,2).
Still, it is almost certain that IgA nephropathy existed long before the techniques of immunofluorescence and EM made its definitive identification possible. For example, in the early 20th century, Volhard and Fahr (3) described an acute form of focal glomerulonephritis occurring at the height of certain infections including tonsillitis. This was characterized by hematuria but not hypertension and only rarely edema and azotemia. These features, which resemble those described by Berger (1,2), appeared to distinguish this form of glomerulonephritis from acute poststreptococcal glomerulonephritis (3). Furthermore, unlike the typical course of poststreptococcal glomerulonephritis, some patients continued to have persistent or episodic hematuria, the latter often exacerbated by recurrent infections (3), and some developed a chronic stage of the disease characterized by proteinuria as well as hematuria. A similar group of patients with “acute focal nephritis” were later described by Ellis (4). In the interval between these early reports and that of Berger and Hinglais (1), a number of investigators described in autopsy and renal biopsy studies lesions of focal glomerulonephritis, most often mesangial proliferative, associated with gross and microscopic hematuria (5,6,7). It is almost certain that many of the lesions described in these reports were in fact IgA nephropathy.
Subsequent to the initial demonstration by Berger and Hinglais (1) of glomerular IgA deposits in a group of patients with primary glomerulonephritis, similar mesangial deposits of IgA were identified in renal biopsies of patients with Henoch-Schönlein purpura (HSP) nephritis (8). HSP, which had been
initially described over a decade earlier (9,10), is now classified as a form of systemic vasculitis with “IgA-dominant immune deposits affecting small vessels and typically involving the skin, gut, and glomeruli and associated with arthralgias or arthritis” (11), although renal involvement does not occur in all cases. Renal disease in Henoch-Schönlein purpura (Henoch-Schönlein nephritis), and how this may be related to primary IgA nephropathy, are addressed later in this chapter.
initially described over a decade earlier (9,10), is now classified as a form of systemic vasculitis with “IgA-dominant immune deposits affecting small vessels and typically involving the skin, gut, and glomeruli and associated with arthralgias or arthritis” (11), although renal involvement does not occur in all cases. Renal disease in Henoch-Schönlein purpura (Henoch-Schönlein nephritis), and how this may be related to primary IgA nephropathy, are addressed later in this chapter.
The initial portion (and majority) of this chapter focuses on the pathology of IgA nephropathy, including primary and secondary forms but primarily the former, with significant emphasis given to clinicopathologic correlations and potential pitfalls in diagnosis. In this context, the epidemiology, clinical features, treatment, and current understanding of the pathogenesis of primary IgA nephropathy are reviewed in some detail, although references to recent, comprehensive reviews of these latter topics are also provided.
TABLE 12.1 Frequency of primary IgA nephropathy in renal biopsy series | ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
PRIMARY IgA NEPHROPATHY
Epidemiology
IgA nephropathy is recognized as the most common form of primary glomerulonephritis in the world (12,13,14), although the prevalence of IgA nephropathy among glomerular diseases and its incidence in the general population show considerable variation among geographic regions and among regions with different renal biopsy practices. As shown in Table 12.1, IgA nephropathy accounts for approximately one third of the total renal biopsy diagnoses and nearly one half of all primary
glomerular diseases diagnosed by renal biopsy in certain Asian countries, including Japan, Mainland China, and Singapore (15,16,17,18,19,20,21). In most European and Australian studies, IgA nephropathy accounts for 10% to 20% of renal biopsy diagnoses and 20% to 30% of primary glomerular diseases (22,23,24,25,26,27,28,29,30,31,32,33,34), although in the United States and Canada, these percentages are generally lower (37,38,39,40,42). Still, IgA nephropathy has been found to be the most common primary glomerular disease in the United States among adults aged 20 to 39, accounting for 14% of renal biopsy diagnoses in this age group (41).
glomerular diseases diagnosed by renal biopsy in certain Asian countries, including Japan, Mainland China, and Singapore (15,16,17,18,19,20,21). In most European and Australian studies, IgA nephropathy accounts for 10% to 20% of renal biopsy diagnoses and 20% to 30% of primary glomerular diseases (22,23,24,25,26,27,28,29,30,31,32,33,34), although in the United States and Canada, these percentages are generally lower (37,38,39,40,42). Still, IgA nephropathy has been found to be the most common primary glomerular disease in the United States among adults aged 20 to 39, accounting for 14% of renal biopsy diagnoses in this age group (41).
TABLE 12.2 Incidence of primary IgA nephropathy | ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Geographic differences in the frequency of diagnosis of IgA nephropathy are also reflected in studies determining the incidence of IgA nephropathy as new cases per 1,000,000 population per year in different regions of the world (Table 12.2). While all of the factors underlying these geographic differences are not known, two factors appear to play significant roles: renal biopsy practices in different countries and race. In some Asian countries, most notably Japan, there are well-established screening programs to detect urinary abnormalities in children and young adults, with renal biopsies presently being performed in those individuals with persistent microscopic hematuria and/or persistent mild proteinuria (15). In a recently published study, 48% of Japanese children initially identified through the urinary screening program who subsequently underwent a renal biopsy had IgA nephropathy (15). The opposite end of the renal biopsy practice spectrum is represented by many if not most centers in the United States and Canada, as well as in New Zealand and some centers in the United Kingdom (24,26,33), where renal biopsies are often not performed in individuals (particularly adults) with microscopic hematuria and mild proteinuria (less than 0.5 or 1.0 g/d or less than 2+ on dipstick) unless there is accompanying renal insufficiency and/or hypertension (47). Geddes et al. (48) recently reported that only 5.8% of Canadian patients with biopsy-proven IgA nephropathy had proteinuria of less than 0.5 g/d, as compared with greater than 20% of IgA nephropathy patients from Scotland, Finland, and Australia. The increase in the apparent frequency of IgA nephropathy among primary glomerulonephritides in Torino, Italy, from 15.4% during the 1970s to 34.5% from 1990 to 1994 has been attributed to a change in renal biopsy practices and a resulting increase in isolated urinary abnormalities as the indication for renal biopsy (from 4% to 30%) (29).
TABLE 12.3 Frequency of IgA nephropathy in different races in the U.S. renal biopsy series | |||||||||||||||||||||||||||||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| |||||||||||||||||||||||||||||||||||||||||||
Race also appears to be an important determinant in the frequency of IgA nephropathy. Several studies in the United States, summarized in Table 12.3, have reported a considerably lower incidence of IgA nephropathy in African Americans as compared with that in Caucasians (37,38,39,40), although one study from Shelby County (Memphis), Tennessee, found similar, relatively low (less than 6 cases per million per year) incidences in Caucasian and African American children (49). Notably, IgA nephropathy also appears to be uncommon in African blacks. In a study from Natal, South Africa (35), it was found that IgA nephropathy comprised less than 1% of primary glomerular
diseases in blacks, as compared with 13% of primary glomerular diseases in Indians (see Table 12.1). Among biopsies processed at the University of Chicago from 1980 to 1994, we also found a higher frequency of IgA nephropathy in specimens from Asian Americans than from Caucasians (see Table 12.3), although the frequency in the former group was only 12%, well below that reported in studies from Japan and China (see Table 12.1), consistent with the concept that race and differences in renal biopsy practice are both major contributors to the geographic variability seen in the incidence of IgA nephropathy. A high incidence of IgA nephropathy has also been reported in Native Americans from New Mexico (38), although its incidence among the Polynesian population of New Zealand is substantially lower than among New Zealanders of European descent (24) (see Tables 12.1 and 12.3).
diseases in blacks, as compared with 13% of primary glomerular diseases in Indians (see Table 12.1). Among biopsies processed at the University of Chicago from 1980 to 1994, we also found a higher frequency of IgA nephropathy in specimens from Asian Americans than from Caucasians (see Table 12.3), although the frequency in the former group was only 12%, well below that reported in studies from Japan and China (see Table 12.1), consistent with the concept that race and differences in renal biopsy practice are both major contributors to the geographic variability seen in the incidence of IgA nephropathy. A high incidence of IgA nephropathy has also been reported in Native Americans from New Mexico (38), although its incidence among the Polynesian population of New Zealand is substantially lower than among New Zealanders of European descent (24) (see Tables 12.1 and 12.3).
Clinical Presentation and Laboratory Findings
IgA nephropathy can occur in individuals of virtually any age, from young children to the elderly (40), but occurs most commonly between the ages of 10 and 40 years (Fig. 12.1). Most studies show a male predominance, with an overall average male:female ratio of approximately 2:1 (50). However, this ratio tends to be higher in the predominantly Caucasian populations of Western Europe, Australia, and North America than in Japan (15,17,22,23,25,30,40,48,51), and in Hong Kong, a slight female predominance has been reported (52,53).
TABLE 12.4 Presenting clinical syndromes in children and adults with IgA nephropathy | ||||||||||||||||||||||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| ||||||||||||||||||||||||||||||||||||
The two major clinical presentations of primary IgA nephropathy are asymptomatic urinary abnormalities and macroscopic hematuria, the former being most common in adults and the latter in children (Table 12.4). When macroscopic hematuria is a presenting symptom, this frequently occurs at the time of or 1 to 2 days following an infectious illness, most commonly pharyngitis or tonsillitis but occasionally gastroenteritis, pneumonia, or urinary tract infection (23,50,54). This presentation thus differs from that of acute poststreptococcal glomerulonephritis in which macroscopic hematuria is typically observed 1 to 2 weeks after an episode of pharyngitis or tonsillitis and 3 to 6 weeks after a skin infection (55), although rare cases of IgA nephropathy presenting several weeks following a streptococcal pharyngitis have been reported (56). Recurrent macroscopic hematuria is not uncommon over the course of the disease. This is often associated with infections and may occur in patients who did not initially present with macroscopic hematuria. Unilateral or bilateral abdominal pain, in some instances related to intestinal vasculitis, may be present and can be severe, giving the false impression that the accompanying hematuria is secondary to passage of a renal stone (54,57). Hypertension is noted in approximately 25% of patients at presentation, with this fraction increasing with increasing age, and develops in an additional 25% of patients over time. When hypertension is present at presentation in children and young adults, it is often accompanied by proteinuria and renal insufficiency (50). Table 12.5, which is taken from an exhaustive review of the literature published by Ibels and Gyory in 1994 (50), summarizes clinical features at presentation from 82 series of patients with IgA nephropathy.
The majority of patients with IgA nephropathy present with some degree of proteinuria, although this is often mild and nephrotic syndrome is relatively uncommon in most series (see Tables 12.4 and 12.5). In the original series of Berger (2), in which only 4 of 55 patients were hypertensive and only 3 had renal insufficiency, most patients had only mild proteinuria (less than 1 g/d) and none had nephrotic syndrome. Likewise, in children, who most often present with macroscopic and/or microscopic hematuria with normal renal function, fewer than 50% have proteinuria of greater than 1 g/1.73 m2/24 h or
greater than 1+ on dipstick and fewer than 10% have proteinuria in the nephrotic range (≥3.5 g/1.73 m2/24 h) (15,31,58,59,60). Proteinuria of greater than 1 g/d or greater than 1+ by dipstick at the time of renal biopsy is also uncommon in most Japanese series (e.g., (17,61,62)), most likely reflecting the considerable fraction of IgA nephropathy patients identified through that country’s urinary screening program. By contrast, in most North American studies as well as in other studies focusing on patients with significant (but not severe) glomerular alterations on renal biopsy, the majority of patients have proteinuria of greater than 1 g/d (40,51,63,64,65), and in two large series from the United States, 47% and 48%, respectively, of patients with primary IgA nephropathy had proteinuria of ≥2 g/d (40,51). As is discussed below, a small number of patients will present with an overlap of IgA nephropathy and minimal change disease or perhaps more correctly minimal change disease with incidental mesangial IgA deposits. Identification of these patients is important as their proteinuria is steroid responsive and most have an excellent prognosis (66,67,68,69).
greater than 1+ on dipstick and fewer than 10% have proteinuria in the nephrotic range (≥3.5 g/1.73 m2/24 h) (15,31,58,59,60). Proteinuria of greater than 1 g/d or greater than 1+ by dipstick at the time of renal biopsy is also uncommon in most Japanese series (e.g., (17,61,62)), most likely reflecting the considerable fraction of IgA nephropathy patients identified through that country’s urinary screening program. By contrast, in most North American studies as well as in other studies focusing on patients with significant (but not severe) glomerular alterations on renal biopsy, the majority of patients have proteinuria of greater than 1 g/d (40,51,63,64,65), and in two large series from the United States, 47% and 48%, respectively, of patients with primary IgA nephropathy had proteinuria of ≥2 g/d (40,51). As is discussed below, a small number of patients will present with an overlap of IgA nephropathy and minimal change disease or perhaps more correctly minimal change disease with incidental mesangial IgA deposits. Identification of these patients is important as their proteinuria is steroid responsive and most have an excellent prognosis (66,67,68,69).
TABLE 12.5 Clinical and laboratory features at presentation or biopsy in patients with IgA nephropathy | ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
The frequency of renal insufficiency at the time of initial presentation or diagnosis likewise varies between different studies. This frequency is greater in adults than in children and in North American series compared with those from Asia and Europe. In our series from Chicago (40) and that of Radford et al. (51) from the Mayo Clinic, in which nearly 50% of patients had proteinuria of 2 g or more at the approximate time of renal biopsy, the majority of patients had initial serum creatinine levels greater than 1.2 mg/dL, and the mean serum creatinine among 220 patients in our series was 2.2 ± 1.9 (SD) mg/dL. By contrast, in the German study of Bogenschutz et al. (70) in which only 88 (39%) of 226 patients had proteinuria of greater than 1 g/d at the time of biopsy, only 34% of patients had an initial serum creatinine greater than 1.2 mg/dL, and in the Japanese series of Koyama (17) in which 97 of 322 (30%) patients presented with greater than 1+ proteinuria by dipstick, only 23% of patients had an initial serum creatinine greater than 1.3 mg/dL. IgA nephropathy rarely presents with acute renal failure in children or young adults, although this occurs more often in the elderly (see Table 12.4), perhaps due in part to the frequent presence of an underlying chronic renal disease (e.g., due to essential hypertension) in this population. Severe, crescentic forms of IgA nephropathy may present with acute renal failure (71), similarly to other forms of crescentic glomerulonephritis. Reversible acute renal failure, which may be severe and even require dialysis, can occasionally occur in association with episodes of macroscopic hematuria resulting from IgA nephropathy. The glomerular histopathologic lesions in such cases are often relatively mild (72,73,74,75,76). This acute renal failure is felt to result from injury to tubular epithelium and/or tubular obstruction by red blood cell casts (Fig. 12.2). Interestingly, patients with IgA nephropathy who are receiving anticoagulant drugs may be more susceptible to this form of acute renal failure (77).
Serum IgA levels tend to be higher in patients with IgA nephropathy than in healthy subjects, with significant elevations in up to 50% of patients (23,78,79). However, the elevation in serum IgA levels is not a consistent enough finding for this to be
useful as a diagnostic marker for IgA nephropathy (49,78,79). Serum IgA levels also have no prognostic significance as a marker for disease severity or progression (80,81,82). Serum levels of IgA-fibronectin aggregates were initially reported to be elevated in 93% of patients with IgA nephropathy or Henoch-Schönlein nephritis, as compared with fewer than 20% of patients with other glomerular diseases (83). However, in a subsequent study with a greater number of patients involving a larger number of patients with IgA nephropathy or Henoch-Schönlein nephritis, only 38% of serum samples from these patients had elevated levels of IgA-fibronectin aggregates, although 48% of patients had an elevated level on at least one occasion (79). Serum levels of IgA-fibronectin aggregates clearly cannot be used as an alternative to a renal biopsy for the diagnosis of IgA nephropathy.
useful as a diagnostic marker for IgA nephropathy (49,78,79). Serum IgA levels also have no prognostic significance as a marker for disease severity or progression (80,81,82). Serum levels of IgA-fibronectin aggregates were initially reported to be elevated in 93% of patients with IgA nephropathy or Henoch-Schönlein nephritis, as compared with fewer than 20% of patients with other glomerular diseases (83). However, in a subsequent study with a greater number of patients involving a larger number of patients with IgA nephropathy or Henoch-Schönlein nephritis, only 38% of serum samples from these patients had elevated levels of IgA-fibronectin aggregates, although 48% of patients had an elevated level on at least one occasion (79). Serum levels of IgA-fibronectin aggregates clearly cannot be used as an alternative to a renal biopsy for the diagnosis of IgA nephropathy.
 FIGURE 12.2 IgA nephropathy presenting with acute renal failure. Photomicrograph of the medulla shows collecting tubules packed with red blood cells, apparently producing urinary obstruction. (Hematoxylin-eosin [H&E] stain, magnification ×200.) |
Serum levels of total complement, C3, and C4 are within the normal range or are elevated (mainly C4) in the vast majority of patients with IgA nephropathy and are only rarely decreased (23,81). The normal levels of C3 in IgA nephropathy contrast with those of acute postinfectious glomerulonephritis, in which C3 levels are often depressed (84). Still, there is serologic evidence of complement activation in IgA nephropathy, in the form of activated forms of C3 (85,86,87). Monoclonal antibodies developed against neoantigens expressed on C3 breakdown products (C3b, iC3b, C3d, and C3dg) that are formed following C3 activation have been detected in 30% to 50% of plasma samples from patients with IgA nephropathy (85,86). In one study, such evidence of C3 activation was found in 75% of adult and 57% of pediatric IgA nephropathy patients when serial plasma samples were analyzed (85). Mean plasma C3a levels have also been reported to be significantly higher in patients with IgA nephropathy than in healthy individuals or those with non-immunemediated renal diseases (87). The significance of this apparent C3 activation on the course of IgA nephropathy remains unclear. While studies from one laboratory have shown a correlation between high plasma levels of activated C3 (as detected by one of the monoclonal antibodies noted above), but not those of C3a, and deterioration of renal function (86,87), another laboratory found no association between levels of neoantigen from the same C3 breakdown products (albeit detected using a different monoclonal antibody) and the degree of proteinuria or the presence of chronic renal insufficiency (85). Elevated urinary levels of interleukin-6 (IL-6), a cytokine secreted by both mesangial and tubular epithelial cells, or an elevated ratio of the urinary concentration of IL-6 to that of epidermal growth factor (EGF) has also been reported to correlate with clinical and histologic progression of IgA nephropathy (88,89).
Correlation of Clinical Parameters With Clinical Outcomes
A considerable number of clinical parameters and laboratory findings have been reported to correlate with clinical outcomes in patients with IgA nephropathy. These range from results of highly specialized assays performed only in a limited number of laboratories, such as those discussed in the previous paragraph, to routine parameters measured in most or nearly all patients suspected of having kidney disease. In IgA nephropathy, as in a number of other glomerular diseases, three of the following parameters are most consistently and reproducibly found to be correlated with disease progression: glomerular filtration rate (GFR) or serum creatinine level at the time of presentation or renal biopsy, severity of proteinuria, and presence or absence of hypertension. As shown in Table 12.6, which is taken from a review by D’Amico (96) and summarizes results from 23 studies from four continents examining actuarial renal survival in relatively large cohorts of nonselected adults with biopsy-proven IgA nephropathy, reduced GFR (or elevated serum creatinine), severe proteinuria (defined in different studies as greater than 1, ≥2, ≥3 g/d, or nephrotic range), and hypertension were found to be independent predictors of progression to endstage renal disease (ESRD) by multivariate analysis in nearly all studies in which such analysis was performed. In the majority of these studies, the multivariate analysis included histopathologic parameters (which will be discussed later) as well as multiple clinical parameters, some but not all of which are listed in Table 12.6. In a number of studies, including a German study of 239 patients (70) and a study of 148 patients from the Mayo Clinic (50), serum creatinine at the time of the renal biopsy was found to be a strongest independent predictor of renal survival. By contrast, Bartosik et al. (63) found that among 298 patients from Toronto, the mean arterial pressure and severity of proteinuria over time were the most important prognostic indicators of deterioration of renal function. Furthermore, time-averaged proteinuria during follow-up is of greater predictive value than proteinuria at the time of renal biopsy with respect to rates of disease progression (110), perhaps, at least in part, because factors other than active glomerular injury contribute to proteinuria.
As shown in Table 12.6, a number of additional demographic and clinical features have been found to correlate with an increased risk of progression to ESRD in some but not all studies, often by univariate but not multivariate analysis. These include male sex, older age at presentation or diagnosis, and the absence of any history of macroscopic hematuria, although notably in studies from the Mayo Clinic, younger age at diagnosis and female sex were found to be independent risk factors for disease progression (51). It is of interest that the absence of macroscopic hematuria has been found to be correlated with shortened renal survival, at least by univariate analysis, in the majority of studies in which this has been examined. Active crescentic lesions on renal biopsy have been linked to episodes of macroscopic hematuria in IgA nephropathy (111); however, overall the presence of macroscopic hematuria tends to be more frequently associated with mild histologic lesions. In our study of over 200 patients (mostly adults) with IgA nephropathy, 17 of 38 (45%) patients with normal glomerular histology had macroscopic hematuria, as compared with 35 of 104 (34%) patients with moderate histologic lesions and 15 of 68 (22%) patients with the most severe histologic lesions in terms of activity and chronicity (40). We have more recently found similar results in children with IgA nephropathy: 22 of 27 (81%) whose renal biopsy showed normal glomerular histology had at least one episode of macroscopic hematuria prior to biopsy, as compared with 14 of 25 (56%) children with diffuse proliferative glomerulonephritis and 0 of 4 with advanced chronic glomerulonephritis.
One factor that has not been shown to correlate with clinical outcome is race, at least among patients from the same geographic region. Lau et al. (112) recently reported that clinical features at presentation and 5-year renal survival among Caucasian and African American children with IgA nephropathy from Tennessee were not significantly different except for a higher incidence of macroscopic hematuria among Caucasians. In IgA nephropathy patients (mostly adults) from Chicago, race and ethnicity (white, black, Hispanic, Asian) were not significant prognostic indicators of renal survival by univariate analysis, although the numbers of Hispanic and Asian patients were small in this study (40).
TABLE 12.6 Statistical analysis of the predictive value, as risk factors for ESRF, of some clinical features in 23 studies selected from the literature | ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Pathologic Findings
Gross Pathology
The kidneys of most patients with IgA nephropathy are likely to appear grossly normal in the absence of changes related to chronic glomerulonephritis with associated tubulointerstitial scarring and/or to hypertensive vascular disease. A case of diffuse proliferative and crescentic IgA nephropathy has been reported in a pig-tailed macaque that was euthanized 1 year after experimental coinfection with simian immunodeficiency virus and Mycobacterium bovis; both of the animal’s kidneys were found to be swollen and enlarged (113), suggesting that similar gross changes in the kidneys may occur in cases of human IgA nephropathy that show similarly severe histologic changes.
Light Microscopy and Histopathologic Classification
In Berger’s original series of patients with IgA nephropathy, the majority of renal biopsies showed focal glomerulonephritis, although a significant number showed chronic glomerulonephritis and approximately 10% were histologically normal (2). Subsequent studies have confirmed that while focal or diffuse mesangial proliferative glomerulonephritis are the most common histologic lesions seen on renal biopsies from patients with IgA nephropathy, there is a wide degree of histologic variability in this disease, ranging from no detectable histologic lesion to diffuse proliferative and crescentic glomerulonephritis, much as is the case with lupus nephritis (17,40,50,80,97,114,115,116,117,118).
Because of the highly diverse histologic presentation of IgA nephropathy, a number of histologic classification systems have been devised and tested for their value in predicting clinical outcomes, most often actuarial renal survival. These classification systems are of two basic types: semiquantitative and single grade. Semiquantitative scoring systems, such as those employed by Alamartine et al. (97), Kobayashi et al. (61), and Radford et al. (51), assign a semiquantitative grade (e.g., 0, 1, 2, or 3 corresponding to absent, mild, moderate, or severe) to each of a number of parameters comprising the glomerular, tubular, interstitial, and vascular compartments and generate a sum score for each compartment as well as a total histologic score by adding the sum scores for each compartment. The number of individual scores in such systems generally ranges from 10 to 20. Semiquantitative scoring systems have the clear advantage of identifying specific morphologic changes (or combinations of these) that are best correlated with clinical outcome and/or with each other. However, despite generally good correlations between total histologic scores and renal failure rates (51,61,97), semiquantitative scoring systems are felt by many pathologists to be too time consuming to be used routinely in busy renal biopsy practices. Still, abbreviated versions of these, focusing on histologic changes in a single compartment (e.g., tubulointerstitial changes) or only on chronic changes within each compartment, have been used at some centers and found to show good correlation with clinical outcomes (52,106).
Single-grade systems combine various glomerular and tubulointerstitial features, in most instances with emphasis on the former, into a relatively small number (most often five) of histologic grades (40,114,115,116,119). Of these, the three most widely used are those of Lee et al. (114), Haas (40), and modifications of different versions of the World Health Organization (WHO) and closely related International Society of Nephrology/Renal Pathology Society (ISN/RPS) classification for lupus nephritis (see Chapter 14). The classification of Lee et al. (114) was based on the previously published classification system of Meadow et al. (120) for HSP IgA vasculitis nephritis (see section on IgA vasculitis later in this chapter), whereas that of Haas (40) borrowed features of the Lee et al. and WHO lupus nephritis classifications, as well as from findings of D’Amico et al. (92) with respect to the influences of multiple individual histologic parameters on renal survival in a large cohort of Italian IgA nephropathy patients. Both of the latter classification systems have been independently verified to be clinically relevant (63,103,106,116,118), although it has not been shown that the system alone is superior either to a semiquantitative tubulointerstitial score (106) or to a model based on the mean arterial pressure and urinary protein excretion averaged over time during 2 to 3 years of observation (63) in predicting clinical progression of IgA nephropathy. Another potential shortcoming of both the Lee et al. (114) and Haas (40) classifications is that both can be vague with respect to terminology, limiting their reproducibility. For example, the Lee et al. (114) schema does not clearly distinguish between mesangial sclerosis and segmental glomerulosclerosis (121), while that of Haas (40) does not distinguish between mesangial and endocapillary hypercellularity. Furthermore, while both of these classifications have five individual histologic subclasses, the majority of cases of IgA nephropathy fall within just one or two of these subclasses (63,121,122), limiting their overall clinical utility.
THE OXFORD CLASSIFICATION AND ITS CLINICAL APPLICABILITY
Documentation of the clinical relevance of the histologic classifications of Lee et al. (114) and of Haas (40), as well as of other classifications (52,116,117,118,119), has been based on retrospective studies, most often at single centers, using the development of ESRD as their primary end point. However, rates of progression to ESRD from the time of diagnosis (biopsy) are largely related to irreversible damage already present at the time of biopsy, as well as compensatory mechanisms (e.g., glomerular hyperfiltration) that result from this underlying nephron loss. This is borne out by the finding that the strongest and most consistent histologic predictors of the development of ESRD in IgA nephropathy are tubular atrophy/interstitial fibrosis (TA/IF) and, to a somewhat lesser extent, glomerular sclerosis (93,96). By contrast, time-averaged proteinuria, the strongest clinical correlate of the rate of decline of renal function in IgA nephropathy (110,121), reflects in large part active glomerular injury. Time-averaged proteinuria during follow-up is of greater predictive value than proteinuria at a single time point (e.g., the time of biopsy) with respect to rates of disease progression (110), perhaps, at least in part, because factors other than active glomerular injury contribute to proteinuria at any given time point. Still, there are different elements of active glomerular injury (e.g., mesangial proliferation, endocapillary proliferation, glomerular necrosis, and crescent formation) that may have rather different implications with respect to prognosis and/or potential response to immunosuppressive therapy and may affect clinical parameters such as proteinuria very differently. As such, there remains a potentially important role for histologic assessment in guiding the clinician’s approach to the patient with IgA nephropathy, provided the right histologic parameters are evaluated.
With this in mind, an international committee comprising of pathologists and nephrologists from four continents first convened in 2004 for the purpose of developing a new, evidenced-based, international consensus histologic classification for IgA nephropathy. A number of features distinguished
the approach of this group from that used in the development of prior histologic classifications of IgA nephropathy (123,124):
the approach of this group from that used in the development of prior histologic classifications of IgA nephropathy (123,124):
The pathologists scored different histologic parameters from biopsies of 265 closely monitored patients with active (i.e., with proteinuria, without advanced chronic renal insufficiency) IgA nephropathy. The parameters for this scoring were agreed upon and criteria for different lesions (e.g., mesangial hypercellularity) defined at a meeting of the participating pathologists, and subsequently each biopsy was scored by a minimum of three pathologists and by four or five in 83% of cases.
The parameters scored by the pathologists included a comprehensive list of glomerular, tubulointerstitial, and vascular findings with no preconceived determination as to the relative importance of any of these. The inclusion of individual histologic parameters in the final classification was determined by three factors: reproducibility among the pathologists, independence from other parameters, and value as a predictor of clinical outcome.
The 265 cases included biopsies from 206 adults and 59 children and were from 17 different centers based in eight countries and four continents. Thus, the aim was to develop a classification that would be applicable worldwide. The participating nephrologists determined a minimum set of clinical data that had to be available for all patients included in the study. This included follow-up for a minimum of 3 years (and/or until an end point of ESRD or ≥50% decline in estimated GFR [eGFR]) with a sufficient number of determinations of eGFR during this interval to allow for determination of the slope of eGFR versus time, as a measure of the rate of renal functional decline.
The scoring of the slides and the analysis of the pathologic and clinical data took approximately 4 years, with consensus working meetings of the full group of participating pathologists and nephrologists held at Oxford University in 2005 and at the conclusion of the study in 2008 (hence the name Oxford Classification). Three noninterdependent histologic parameters were found to be significant predictors of both the rate of renal functional decline and survival from ESRD or ≥50% decline in eGFR by univariate analysis, as well as an independent predictor of one or both of these outcomes in a multivariate analysis including initial eGFR, mean arterial pressure, and proteinuria (123): mesangial hypercellularity in ≤50% versus greater than 50% of glomeruli, presence or absence of segmental sclerosis in one or more glomeruli, and TA/IF in ≤25% versus greater than 25% of the renal cortical tissue present. Furthermore, the presence of TA/IF in greater than 50% of the cortex was associated with significantly worse outcomes (both the rate of eGFR decline and ESRD or ≥50% decline in eGFR) than TA/IF in 26% to 50%. One additional histologic parameter—endocapillary hypercellularity in ≥1 glomerulus—was not significantly correlated with the above outcomes by multivariate analysis but was strongly correlated with response to immunosuppressive therapy. Among patients whose biopsies showed endocapillary hypercellularity, those who were treated with immunosuppressive agent(s) had a significantly lower rate of eGFR decline than those who were not (123). Notably, a number of other morphologic parameters, including crescents and vascular lesions, were not found to be significant predictors of clinical outcomes or response to therapy. However, it should be emphasized that patients who progressed to ESRD within 12 months of biopsy and those with an initial eGFR of less than 30 mL/min/1.73 m2 were excluded from the Oxford cohort (123); this would have excluded substantial fractions of those patients with very advanced chronic disease and rapidly progressive disease with large numbers of crescents.
 FIGURE 12.3 IgA nephropathy with minimal histologic lesion (Oxford Classification score M0 E0 S0 T0). The glomeruli are normocellular and histologically unremarkable. There is minimal TA, IF, or interstitial inflammation. (Periodic acid-Schiff [PAS] stain, ×400.) |
Thus, the outcome of this large international study was the recommendation that the primary classification of biopsies with IgA nephropathy include four histologic parameters, M, E, S, and T (illustrated in Figs. 12.3, 12.4, 12.5, 12.6, 12.7, 12.8, 12.9, 12.10, 12.11, 12.12, 12.13, 12.14, 12.15, 12.16):
M0 or M1, indicating mesangial hypercellularity (≥4 cells in one or more mesangial areas) in ≤50% versus >50% of glomeruli
E0 or E1, indicating endocapillary hypercellularity in zero versus one or more glomeruli
S0 or S1, indicating segmental sclerosis in zero versus one or more glomeruli
T0, T1, or T2, indicating TA/IF in ≤25%, 26% to 50%, or >50% of renal cortex, respectively
 FIGURE 12.4 IgA nephropathy with mesangial proliferation (M1 E0 S0 T0). The glomeruli show a mild increase in mesangial matrix, and the glomerulus at left, which was representative of the majority of the glomeruli on this biopsy, shows a segmental increase in mesangial cellularity. (PAS, ×400.) |
 FIGURE 12.5 IgA nephropathy with diffuse mesangial proliferative glomerulonephritis (M1 E0 S0 T0). There is a diffuse increase in mesangial matrix and cellularity, without endocapillary proliferation, segmental glomerulosclerosis, crescent formation, or TA/IF. (PAS, ×200.) |
Studies examining clinicopathologic correlations in IgA nephropathy using the Oxford (MEST) Classification as well as other morphologic parameters are reviewed in detail later in this chapter (see Prognosis and Clinicopathologic Correlations).
Immunohistologic Findings
GLOMERULI
In the original series of patients with IgA nephropathy described by Berger and Hinglais (1,2), it was noted that the glomeruli showed deposits of IgA by immunofluorescence in the intercapillary (mesangial) areas. Berger (2) noted that in only 5 of 55 cases were the deposits large enough to be clearly evident by light microscopy, although on retrospective analysis of silver-stained sections he was able to find small mesangial deposits in a larger number of cases. The mesangial IgA deposits were accompanied by C3 and in all but 2 of 55 cases by IgG (2).
 FIGURE 12.6 IgA nephropathy with diffuse mesangial proliferative glomerulonephritis (M1 E0 S0 T0). There is a diffuse increase in mesangial matrix and cellularity, without endocapillary proliferation or segmental glomerulosclerosis. There is mild chronic change, more than in Figure 12.5, but the degree of TA/IF is less than 25%. (PAS, ×200.) |
 FIGURE 12.7 IgA nephropathy with mesangial proliferation (M1 E0 S0 T0). Note that expanded mesangial areas do not stain black with the methenamine silver stain, this staining is typical of immune complex deposits rather than mesangial matrix. EM (see Fig. 12.22) confirmed the presence of extensive mesangial deposits. (Jones methenamine silver stain, ×400.) |
Following these original reports, the immunohistology of IgA nephropathy has been examined in multiple case series together comprising many hundreds of patients. It is now well accepted that the diagnosis of IgA nephropathy depends on the demonstration, by immunofluorescence or immunohistochemistry, of glomerular staining that is IgA dominant or IgA codominant, involving the mesangium with or without staining of peripheral capillary walls, in a patient without evidence of systemic lupus erythematosus (SLE) or Henoch-Schönlein
purpura (125,126). As noted in Table 12.7, which summarizes results of 13 studies comprising over 2200 patients, over 40% of cases of IgA nephropathy contain mesangial deposits of IgG and over 50% contain mesangial deposits of IgM, although in each case, the overall intensity of staining for IgA was at least as great as that for either of the other immunoglobulins. IgA deposited in the glomeruli of patients with IgA nephropathy is predominantly of the IgA1 subclass. In different studies, IgA2 has been reported to be present (in lesser amount than IgA1) in between 0% and 53% of biopsies, although in the majority of studies, the presence of glomerular IgA2 is uncommon (128,133,134,135,136,137,138,139,140). Most studies indicate that the IgA present in the glomerular deposits is polymeric, based on the absence of secretory component and the presence of J chain within the deposits (134,135,138,141), as will be discussed in more detail below in the section dealing with the pathogenesis of IgA nephropathy.
purpura (125,126). As noted in Table 12.7, which summarizes results of 13 studies comprising over 2200 patients, over 40% of cases of IgA nephropathy contain mesangial deposits of IgG and over 50% contain mesangial deposits of IgM, although in each case, the overall intensity of staining for IgA was at least as great as that for either of the other immunoglobulins. IgA deposited in the glomeruli of patients with IgA nephropathy is predominantly of the IgA1 subclass. In different studies, IgA2 has been reported to be present (in lesser amount than IgA1) in between 0% and 53% of biopsies, although in the majority of studies, the presence of glomerular IgA2 is uncommon (128,133,134,135,136,137,138,139,140). Most studies indicate that the IgA present in the glomerular deposits is polymeric, based on the absence of secretory component and the presence of J chain within the deposits (134,135,138,141), as will be discussed in more detail below in the section dealing with the pathogenesis of IgA nephropathy.
 FIGURE 12.8 IgA nephropathy resembling focal segmental glomerulosclerosis (M0 E0 S1 T0). The glomerulus at left shows perihilar segmental sclerosis with associated hyalinosis as well as an increase in mesangial matrix; however, neither glomerulus shows mesangial or endocapillary hypercellularity. A small number of red blood cells are seen within a tubular lumen above the segmentally sclerotic glomerulus. (H&E, ×400.) |
 FIGURE 12.9 IgA nephropathy with mesangial hypercellularity and segmental glomerulosclerosis (M1 E0 S1 T0). The glomerulus shows an increase in mesangial cellularity and matrix, as well as segmental sclerosis with capsular adhesion. While a small number of double contours of the glomerular basement membrane are present, there is no definite endocapillary proliferation. (PAS, ×400.) |
 FIGURE 12.10 IgA nephropathy with focal endocapillary proliferation (M0 E1 S0 T0). The glomerulus at left is unremarkable; however, that at right shows segmental endocapillary hypercellularity, as well as a small cellular crescent between 1 and 2 o’clock. TA/IF is present, but overall, this was less than the 25% cutoff for T1 according to the Oxford Classification. (H&E, ×200.) |
 FIGURE 12.11 IgA nephropathy with endocapillary proliferation (M0 E1 S0 T0). There is segmental endocapillary proliferation (4 to 6 o’clock) but no mesangial hypercellularity. There are, however, mesangial immune complex deposits that stain red on this trichrome stain. (Masson trichrome stain, ×400.) |
With respect to complement components, C3 is detected in glomeruli in greater than 90% of cases of primary IgA nephropathy (see Table 12.7), as are the C3 breakdown products C3c and C3d (125,131,142). Early components
of the classical complement pathway, namely, C1q and C4 (53,129,131,132,142), are absent in most cases of IgA nephropathy, although C4 has been reported in up to 30% of cases, more often those with deposits containing multiple immunoglobulins rather than IgA alone (143). However, in many if not most cases, this C4 deposition probably indicates activity of the lectin pathway of complement activation rather than that of the classical pathway (144). In the lectin pathway, C4 becomes activated independently of C1 following the binding of mannose-binding lectin to carbohydrate ligands (145). Properdin, a component of the alternative pathway, has been reported to be present in nearly all cases in some studies (132,143), although in less than 50% in others (129). Terminal components of complement C5, C6, and C9 as well as neoantigens of the C5b-9 membrane attack complex (MAC) have also been detected by immunofluorescence in glomeruli of patients with IgA nephropathy, and dual immunostaining studies have shown colocalization of MAC neoantigens and IgA (but interestingly not IgG or IgM) at identical sites in the mesangium and peripheral capillary walls (131).
of the classical complement pathway, namely, C1q and C4 (53,129,131,132,142), are absent in most cases of IgA nephropathy, although C4 has been reported in up to 30% of cases, more often those with deposits containing multiple immunoglobulins rather than IgA alone (143). However, in many if not most cases, this C4 deposition probably indicates activity of the lectin pathway of complement activation rather than that of the classical pathway (144). In the lectin pathway, C4 becomes activated independently of C1 following the binding of mannose-binding lectin to carbohydrate ligands (145). Properdin, a component of the alternative pathway, has been reported to be present in nearly all cases in some studies (132,143), although in less than 50% in others (129). Terminal components of complement C5, C6, and C9 as well as neoantigens of the C5b-9 membrane attack complex (MAC) have also been detected by immunofluorescence in glomeruli of patients with IgA nephropathy, and dual immunostaining studies have shown colocalization of MAC neoantigens and IgA (but interestingly not IgG or IgM) at identical sites in the mesangium and peripheral capillary walls (131).
 FIGURE 12.12 Mesangial and endocapillary proliferative IgA nephropathy (M1 E1 S0 T0) resembling membranoproliferative glomerulonephritis. There is an increase in silver-positive mesangial matrix, hyperlobulation, and mesangial and endocapillary hypercellularity. Several double contours of the glomerular basement membrane are apparent. (Jones methenamine silver, ×400.) |
 FIGURE 12.13 IgA nephropathy with diffuse mesangial and endocapillary hypercellularity; the glomerulus at right shows a segmental fibrocellular crescent with segmental sclerosis of the tuft adjacent to the crescent (M1 E1 S1 T0). There is less than 25% TA/IF. (PAS, ×200.) |
 FIGURE 12.14 IgA nephropathy with mesangial hypercellularity, segmental glomerulosclerosis, and TA/IF (M1 E0 S1 T1). The glomerulus present shows segmental mesangial hypercellularity and segmental sclerosis. There is moderate TA/IF, overall estimated to involve greater than 25% of the renal cortical tissue present. (Masson trichrome stain, ×400.) |
 FIGURE 12.15 Advanced chronic IgA nephropathy with residual mesangial hypercellularity (M1 E0 S0 T2). There is TA/IF, with a nonspecific mononuclear cell inflammatory infiltrate in the sclerotic interstitium. Arterioles present (upper right) show hypertensive changes. One of the two glomeruli shown is globally sclerotic, while the other shows mild and segmental mesangial hypercellularity. (H&E, ×200.) |
Lai et al. (53,148) have studied in some detail the light chain composition of glomerular IgA deposits in patients with IgA nephropathy from Hong Kong. They found that 29 of 45 (64%) biopsies showed predominant staining for lambda light chains by immunofluorescence, while only 1 of
45 biopsies (2%) showed kappa predominance; the remaining 15 (33%) cases showed equivalent staining for kappa and lambda. Seventeen (38%) of the biopsies showed staining for lambda in the absence of kappa (148). Smith (130) has noted similar findings in Native Americans and, to a lesser degree, Caucasians, from the Southwestern United States. Others, however, have found exclusively lambda staining less frequently, if at all. In 813 cases of primary IgA nephropathy, we have examined at three separate institutions, 11% showed glomerular staining for lambda alone, 1% for kappa alone, 86% for both kappa and lambda, and 2% for neither kappa nor lambda. Jennette (125) found glomerular staining for lambda alone in only 3% of cases, although the intensity of lambda staining exceeded that for kappa in an additional 64%. Suzuki et al. (139) found staining for both kappa and lambda in each of 191 cases examined, although they too commented that the intensity of staining for lambda was usually equal to or stronger than that for kappa.
45 biopsies (2%) showed kappa predominance; the remaining 15 (33%) cases showed equivalent staining for kappa and lambda. Seventeen (38%) of the biopsies showed staining for lambda in the absence of kappa (148). Smith (130) has noted similar findings in Native Americans and, to a lesser degree, Caucasians, from the Southwestern United States. Others, however, have found exclusively lambda staining less frequently, if at all. In 813 cases of primary IgA nephropathy, we have examined at three separate institutions, 11% showed glomerular staining for lambda alone, 1% for kappa alone, 86% for both kappa and lambda, and 2% for neither kappa nor lambda. Jennette (125) found glomerular staining for lambda alone in only 3% of cases, although the intensity of lambda staining exceeded that for kappa in an additional 64%. Suzuki et al. (139) found staining for both kappa and lambda in each of 191 cases examined, although they too commented that the intensity of staining for lambda was usually equal to or stronger than that for kappa.
 FIGURE 12.16 Severe chronic and active IgA nephropathy (M1 E1 S0 T2). As shown by the Masson trichrome stain, there are severe TA and IF, overall involving more than 50% of the cortical tissue present. There is also a persistent active glomerulonephritis with mesangial and endocapillary hypercellularity, and one glomerulus with a circumferential, largely cellular crescent (lower right). (Masson trichrome stain, ×200.) |
TABLE 12.7 Glomerular immunofluorescence/immunohistochemical findings in primary IgA nephropathy | ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Mesangial deposits of fibrinogen are also seen in the majority of cases of IgA nephropathy (129,131,132). By immunofluorescence, this mesangial fibrinogen staining has a granular appearance, in contrast to the nonspecific pseudolinear staining that is often observed with direct immunofluorescence staining for fibrinogen, the latter mainly in glomerular capillary loops (Fig. 12.17). Fibrinogen staining is also encountered in active crescentic and necrotizing lesions; in these instances, the distribution of the staining corresponds to that of the crescent or area of necrosis, and the staining has a blotchy or globular rather than a granular pattern.
The prototypical pattern of glomerular IgA deposits in IgA nephropathy, as originally described by Berger (2), is that of global and diffuse staining limited to mesangial areas (Fig. 12.18). The intensity of IgA staining is quite variable but is always at least equal to, and in the great majority of cases is greater than, that for any other immunoglobulin (i.e., IgG, IgM) present. In occasional cases, most often with relatively low-intensity IgA staining, this mesangial staining may be more segmental (Fig. 12.19). In a significant subset of cases, no more than 40% in most studies (127,128,131,132,150) but as high as 70% in one (149), deposits of IgA are seen segmentally in peripheral glomerular capillary walls in addition to globally in the mesangium (Fig. 12.20). In most biopsies showing peripheral capillary IgA staining (not including nonspecific pseudolinear staining that may be seen in some cases), EM confirms the presence of subendothelial, subepithelial, and/or intramembranous deposits (127). The potential prognostic implications of such peripheral capillary deposits are discussed below (see Prognosis and Clinicopathologic Correlations).
EXTRAGLOMERULAR DEPOSITS
Specific staining of extraglomerular structures within the kidney is uncommon in IgA nephropathy, although tubular casts and, in cases where there is moderate to heavy proteinuria, protein
resorption droplets within tubular epithelial cells typically stain for IgA and serve as convenient internal controls for IgA immunofluorescence. IgA staining within arterioles and/or small interlobular arteries has been reported in approximately 5% of renal biopsies showing IgA nephropathy (128,151); however, the deposits typically stop abruptly at the glomerular hilum. Although the vascular deposits of IgA are granular and usually accompanied by deposits of C3, their presence does not correlate with specific histopathologic changes in the vessels or with a greater incidence of hypertension or renal insufficiency (151).
resorption droplets within tubular epithelial cells typically stain for IgA and serve as convenient internal controls for IgA immunofluorescence. IgA staining within arterioles and/or small interlobular arteries has been reported in approximately 5% of renal biopsies showing IgA nephropathy (128,151); however, the deposits typically stop abruptly at the glomerular hilum. Although the vascular deposits of IgA are granular and usually accompanied by deposits of C3, their presence does not correlate with specific histopathologic changes in the vessels or with a greater incidence of hypertension or renal insufficiency (151).
 FIGURE 12.17 Glomerular fibrinogen deposition in IgA nephropathy. Immunofluorescence shows fairly global, predominantly granular mesangial staining for fibrinogen, with weaker pseudolinear staining in glomerular capillary walls. (Fluorescein isothiocyanate-conjugated [FITC] anti-human fibrinogen, ×400.) |
 FIGURE 12.18 IgA nephropathy. Immunofluorescence shows global, granular to globular staining for IgA limited to the mesangium. (FITC antihuman IgA, ×400.) |
Granular deposits of IgA in blood vessels of the superficial dermis of normal-appearing skin have been reported in a high fraction (up to 89% in one study (152)) of patients with IgA nephropathy, without systemic manifestations of HSP (152,153,154,155). These vascular IgA deposits are accompanied by C3 and fibrin in most cases, by IgM in the majority, and less often by IgG and C1q (151,152). C1q was seen, albeit most often weakly, in approximately 50% of IgA-positive skin biopsies examined by Thompson et al. (152) but in none of 45 skin biopsies reported by Baart de la Faille-Kuyper et al. (153). The latter group also found that among 262 patients with various nephropathies undergoing a biopsy of normal-appearing skin, 35 of 45 patients with vascular IgA deposits on the skin biopsy also had renal biopsy findings (including mesangial IgA deposits) indicative of IgA nephropathy (24 cases) or Henoch-Schönlein nephritis (11 cases), a level of specificity suggesting a potential diagnostic utility for skin biopsy in cases of suspected IgA nephropathy where a renal biopsy could not be safely performed. A similar conclusion was reached by Lee et al. (154), who found IgA in dermal vessels in 7 of 11 (64%) patients with IgA nephropathy as compared with 19 of 161 (12%) patients with SLE and other immune-mediated disorders. However, this level of specificity was not found by Hirbec et al. (155), who reported dermal IgA staining in 50% of patients with IgA nephropathy and 30% of patients with other immune complex-mediated glomerular diseases, or by Thompson et al. (152), who found cutaneous vascular deposits of IgA in 4 of 5 patients with immune complex-mediated glomerular diseases, 6 of 15 patients with non-immune-mediated renal diseases, and 5 of 7 healthy individuals.
 FIGURE 12.19 IgA nephropathy. Mesangial staining for IgA is much less intense than that in Figure 12.18 and not completely global. This biopsy was from a young adult male with a history of gross hematuria during an upper respiratory infection, persistent microscopic hematuria, very mild proteinuria, and normal renal function. Glomeruli were histologically unremarkable. Immunofluorescence showed only the modest mesangial staining for IgA and C3. Small mesangial electron-dense deposits were seen by EM, with no other deposits and no abnormalities of the glomerular basement membranes. He was diagnosed with mild IgA nephropathy (Oxford Classification M0 E0 S0 T0). (FITC anti-human IgA, ×400.) |
 FIGURE 12.20 IgA nephropathy. There is granular staining for IgA in the mesangium and in some peripheral glomerular capillary walls. (FITC anti-human IgA, ×400.) |
Electron Microscopic Findings
The diagnosis of IgA nephropathy is typically made on the basis of the immunofluorescence and light microscopic findings, together with the clinical data, and it is only in rare cases that EM is needed for an actual diagnosis of IgA nephropathy (156). An example of the latter might involve a patient with a positive antinuclear antibody (ANA), a mesangial proliferative glomerulonephritis, and immunofluorescence showing moderate to strong mesangial staining for IgA, IgG, IgM, and C3, with weak staining for C1q. In this case, EM may be helpful in distinguishing between IgA nephropathy and mesangial (class II)
lupus nephritis (see Differential Diagnosis, below). Still, there are a number of fairly characteristic ultrastructural features of IgA nephropathy, and some studies suggest that certain EM features may have prognostic significance.
lupus nephritis (see Differential Diagnosis, below). Still, there are a number of fairly characteristic ultrastructural features of IgA nephropathy, and some studies suggest that certain EM features may have prognostic significance.
TABLE 12.8 Ultrastructural location of glomerular immune complex deposits in primary IgA nephropathy and IgA vasculitis (Henoch-Schönlein purpura) nephritis | |||||||||||||||||||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| |||||||||||||||||||||||||||||||||
Table 12.8 shows the locations of electron-dense, immune complex deposits within glomeruli in 756 cases of IgA nephropathy examined by EM. In our experience, failure to detect deposits by EM is very rare in IgA nephropathy (0.5% of cases). This has also been the case in most other studies (58,115,119,127,146,157), although in some studies, the absence of deposits by EM has been reported in up to 25% of cases (50,114,147). Virtually all cases of IgA nephropathy show mesangial deposits within expanded mesangial regions. The most typical location of mesangial deposits is in the paramesangium (i.e., the interface of the glomerular basement membrane with the mesangial matrix), such as shown in Figure 12.21. Deposits located deeper in the mesangial matrix, such as are seen in Figure 12.22, have been postulated by some investigators to indicate more chronic lesions (158). Very large mesangial deposits, described as hemispherical (159), may be present and cause prominent bulging of the mesangium, resembling a capillary with a large subendothelial deposit (Fig. 12.23). Mesangial deposits in IgA nephropathy typically show uniform electron density without specific substructure, although occasionally deposits may show areas of electron lucency, microvesicular change, or even membrane-like structures indicative of partial resorption (Fig. 12.24) (160). Collagen fibrils have been noted within the expanded mesangial matrix in some relatively advanced cases (158).
 FIGURE 12.21 IgA nephropathy. A: Drawing of a glomerular capillary and adjacent mesangium showing mesangial immune deposits (black), especially immediately beneath the paramesangial basement membrane. B: Electron micrograph showing paramesangial electron-dense deposits within a mildly expanded mesangium. A small subendothelial deposit is also present (lower right portion of the capillary loop). Glomerular basement membranes are unremarkable, and the majority of podocyte foot processes appear intact. (Uranyl acetate and lead citrate stain, original magnification ×6300.) |
In our experience, deposits involving the glomerular capillary wall (i.e., subendothelial, subepithelial, and/or intramembranous) in addition to the mesangium are seen in approximately 30% of cases of IgA nephropathy. These findings are similar to those reported in most other series, which found peripheral capillary wall deposits in between 20% and 40% of cases (127,146,161), although the frequency of such deposits in some series is only 10% to 20% (50,162). Subendothelial deposits are seen more commonly than subepithelial deposits in our series and that of Lee et al. (146), although the opposite was observed by Yoshimura et al. (127). When subendothelial deposits are present in IgA nephropathy, they are often quite segmental and appear to extend from paramesangial deposits (Fig. 12.25). Occasionally, there are subendothelial deposits associated with duplication of the glomerular basement membrane (Fig. 12.26), although when present, this tends to be
more segmental than in type I membranoproliferative glomerulonephritis (MPGN). Infrequently, there are large subendothelial deposits resembling those seen in lupus nephritis (Fig. 12.27). The presence of large, hump-like subepithelial deposits suggests the possibility of an IgA-dominant postinfectious glomerulonephritis rather than IgA nephropathy, as discussed below (see Differential Diagnosis).
more segmental than in type I membranoproliferative glomerulonephritis (MPGN). Infrequently, there are large subendothelial deposits resembling those seen in lupus nephritis (Fig. 12.27). The presence of large, hump-like subepithelial deposits suggests the possibility of an IgA-dominant postinfectious glomerulonephritis rather than IgA nephropathy, as discussed below (see Differential Diagnosis).
 FIGURE 12.22 IgA nephropathy. A large amount of mesangial electron-dense deposit is present within the expanded mesangium. No subendothelial or subepithelial deposits are present, however. Light microscopy of this biopsy is shown in Figure 12.7. (Uranyl acetate and lead citrate, original magnification ×3800.) |
Electron-dense deposits have been noted in preglomerular arterioles in some cases of IgA nephropathy and rarely in the basement membrane of the Bowman capsule (163,164). However, deposits in peritubular capillaries, small arteries, or tubular basement membranes are not seen.
Structural abnormalities of the glomerular basement membranes are relatively common in IgA nephropathy and may be so pronounced that they suggest the possibility of Alport syndrome. The most common abnormality is thinning of the peripheral capillary basement membranes and in particular the lamina densa, which may be widespread or localized (159,165,166,167). Berthoux et al. (165) found glomerular basement membrane thinning (mean thickness 208 vs. 365 nm in normal controls and 344 nm in cases of IgA nephropathy without thinning) in 23 (39%) of 59 biopsies examined from patients with IgA nephropathy and notably in 14 (67%) of 21 biopsies from females with IgA nephropathy. Others have noted a significantly lower mean glomerular basement membrane thickness in patients with IgA nephropathy than in normal controls (269 vs. 339 nm; Ref. (168)). A significant association between glomerular basement membrane thinning and hematuria was noted by some investigators (159,165) but not by others (166,168). Segmental splitting of the glomerular basement membrane, resembling that seen in Alport syndrome or in cases of severe glomerular hyperfiltration (169), is also not an uncommon finding in IgA nephropathy (159,164,166,167,170). This change may be associated with the presence of subendothelial or intramembranous immune complex deposits but may also be seen in cases in which deposits are limited to the mesangium, as in Figure 12.28, which also shows thinning of the glomerular basement membrane. Other less common alterations of the glomerular basement membranes that may be seen segmentally in IgA nephropathy include widening, lysis of the lamina densa, and rarely frank disruption (159,166,167).
 FIGURE 12.23 IgA nephropathy. There are prominent mesangial electron-dense deposits, including a large deposit (upper left) with associated bulging of the mesangium into the urinary space. Podocyte foot processes show a significant degree of effacement. (Uranyl acetate and lead citrate, original magnification ×5000.) |
Differential Diagnosis
The diagnosis of IgA nephropathy is usually straightforward, as there are relatively few glomerular diseases in which dominant or codominant IgA deposits occur (Table 12.9). HSP nephritis, which will be reviewed in detail later in this chapter, cannot be distinguished from IgA nephropathy on a renal biopsy, and this distinction must be made on clinical grounds. Lupus nephritis may in some cases show glomerular IgA staining by immunofluorescence that is as or more intense than staining for IgG and IgM. In most instances, however, these cases can be readily distinguished from IgA nephropathy clinically and serologically. Rare cases have been reported in which a patient who developed SLE had a renal biopsy showing a mesangial proliferative or membranous lesion with predominant IgA deposits months to several years prior to developing clinical and serologic manifestations of SLE (171). However, such cases of latent lupus nephritis typically show C1q deposits by immunofluorescence and tubuloreticular inclusions by EM (172), whereas these findings are quite uncommon in IgA nephropathy. Cases of C1q nephropathy (173,174) and
human immunodeficiency virus (HIV)-associated lupus-like glomerulonephritis (175,176) may also exhibit strong glomerular IgA staining, but like lupus nephritis also show prominent C1q staining.
human immunodeficiency virus (HIV)-associated lupus-like glomerulonephritis (175,176) may also exhibit strong glomerular IgA staining, but like lupus nephritis also show prominent C1q staining.
 FIGURE 12.24 Chronic IgA nephropathy. This electron micrograph shows several largely resorbed mesangial deposits. The deposits contain electron-lucent areas, as well as microvesicular and membrane-like structures. (Uranyl acetate and lead citrate, original magnification ×17,000.) |
IgA-dominant postinfectious glomerulonephritis has been reported to occur following infections with staphylococci, including Staphylococcus aureus, methicillin-resistant S. aureus (MRSA), and S. epidermidis (177,178,179,180,181), and less commonly after nonstaphylococcal infections (182). A number of these lesions have been reported in patients with underlying diabetic nephropathy (178,180,182). The light microscopic appearance of these cases ranges from diffuse proliferative and exudative glomerulonephritis typical of acute postinfectious glomerulonephritis to mild mesangial proliferative glomerulonephritis, the latter possibly representing a resolving lesion. In the acute phase, these lesions are clearly identifiable as postinfectious not only histologically but also by the presence of coarse granular IgA deposits by immunofluorescence (Fig. 12.29) and large, hump-shaped subepithelial deposits by EM (178,180). Patients with acute poststaphylococcal glomerulonephritis, like those acute poststreptococcal lesions but unlike most patients with IgA nephropathy, are also frequently hypocomplementemic (178). In the resolving phase, the differentiation of these lesions from IgA nephropathy may be very difficult if not impossible, especially as unlike the IgG deposits in poststreptococcal glomerulonephritis, the IgA deposits in the poststaphylococcal lesions appear to persist beyond the acute phase of the disease. While not a definitive finding, the presence on EM of multiple partially or largely resorbed subepithelial deposits, particularly in the mesangial “notch” or “waist” region (Fig. 12.30), suggests a resolving postinfectious lesion (180,183,184).
 FIGURE 12.25 IgA nephropathy. There are mesangial electron-dense deposits that extend into the adjacent glomerular capillary as subendothelial deposits. (Uranyl acetate and lead citrate, original magnification ×3800.) |
Autopsy studies as well as studies of donor kidney biopsies have demonstrated incidental mesangial IgA deposits in a significant fraction of cases, ranging from 4% in an autopsy study from Singapore (185) to 16% in a recent Japanese study of perioperative renal transplant biopsies involving mainly living donors with no overt signs or symptoms of renal disease (186). While a retrospective analysis of these latter cases showed mild microscopic hematuria as well as mild mesangial hypercellularity and glomerular macrophage infiltration in a subset of kidneys with mesangial IgA (and more often those with both IgA and C3) (186), it remains questionable as to whether even the latter cases truly represent an extremely mild form of IgA nephropathy. Regardless, perhaps the greatest consequence of incidental mesangial IgA deposits from the standpoint of the renal pathologist involves interpretation of renal biopsies with such deposits and a superimposed glomerular disease. Minimal change disease with mesangial IgA deposits, presenting as steroid-responsive nephrotic syndrome, has already been noted (65,66,67,68), and likewise, cases of idiopathic membranous nephropathy with mesangial IgA deposits have been reported (187,188,189). These latter biopsies, of which an example is shown in Figure 12.31, show changes of membranous nephropathy
with variable thickening of glomerular capillary loops histologically, often with basement membrane “spikes” evident on silver stains, granular immunostaining for IgG along glomerular capillary loops, and numerous subepithelial deposits by EM. In addition, there is an increase in mesangial matrix and an increase in mesangial cellularity, often mild and segmental, as well as mesangial deposits by EM corresponding to the mesangial immunostaining for IgA (187,188,189).
with variable thickening of glomerular capillary loops histologically, often with basement membrane “spikes” evident on silver stains, granular immunostaining for IgG along glomerular capillary loops, and numerous subepithelial deposits by EM. In addition, there is an increase in mesangial matrix and an increase in mesangial cellularity, often mild and segmental, as well as mesangial deposits by EM corresponding to the mesangial immunostaining for IgA (187,188,189).
 FIGURE 12.26 IgA nephropathy. Subendothelial electron-dense deposits associated with duplication of the glomerular basement membrane. (Uranyl acetate and lead citrate, original magnification ×5000.) |
More important from a clinical management standpoint are cases of ANCA-associated necrotizing and crescentic glomerulonephritis (ANCA-GN) with mesangial IgA (190,191,192,193,194). The high rate of responsiveness of such lesions to immunosuppressive therapy (e.g., corticosteroids plus cyclophosphamide) render it important that they be recognized and distinguished from necrotizing and crescentic forms of IgA nephropathy that generally have a poorer response to therapy (19,195,196). Serologically, indirect immunofluorescence assays for ANCA are positive in approximately 15% of cases of IgA nephropathy and a higher fraction of cases of IgA nephropathy with crescents; nearly all of these positives are P-ANCA without associated seropositivity for antimyeloperoxidase or anti-proteinase 3 (197). Histologically, cases of ANCA-GN with mesangial IgA show mild or absent mesangial or endocapillary hypercellularity, both in glomeruli with crescents (Fig. 12.32) and in those without, although this is often most evident in the latter glomeruli. By contrast, mesangial and/or mesangial plus endocapillary hypercellularity is prominent in cases of IgA nephropathy with crescents and/or necrotizing glomerular lesions (Figs. 12.13, 12.33, and 12.34) (190).
 FIGURE 12.27 IgA nephropathy. There is a large amount of mesangial electron-dense deposit and a segmental, large subendothelial deposit resembling a “wire loop” of lupus nephritis. (Uranyl acetate and lead citrate, original magnification ×3000.) |
Pathogenesis of Primary IgA Nephropathy
In considering the pathogenesis of primary IgA nephropathy, four sets of questions need to be addressed:
What are the properties of the IgA that is deposited in the glomeruli (and more specifically, the mesangium), and are there properties of this IgA that are unique to patients with IgA nephropathy?
What is/are the source(s) of the mesangial IgA?
What are the mechanisms involved in mesangial IgA deposition, and do specific properties of the IgA and/or the mesangium in patients with IgA nephropathy promote this deposition?
What are the events involved in the glomerular response to IgA deposition that lead to glomerulonephritis, and what role(s) do specific properties of the IgA and/or mesangium in patients with IgA nephropathy play in promoting these events?
Consideration of these issues best begins with a brief review of the structure, subclasses, and synthesis of IgA in humans.
 FIGURE 12.28 Glomerular basement membrane structural abnormalities in IgA nephropathy. This glomerular capillary shows both thinning (left) and splitting (right) of the GBM. Foot processes are effaced overlying the area of GBM splitting. Paramesangial electron-dense deposits are present. (Uranyl acetate and lead citrate, original magnification ×8000.) |
Structure and Production of Human IgA
There are two subclasses of human IgA, IgA1 and IgA2, and both can exist as monomers (mIgA) or polymers (pIgA), the latter mainly dimers. IgA dimers contain the 21 kDa linking protein known as the J chain (198). The main structural difference between these subclasses is the presence in IgA1 of an 18-amino-acid hinge region located between the CH1 and CH2 portions of the molecule (199,200). This hinge region is composed of proline, serine, and threonine residues and carries O-linked carbohydrate side chains that have been documented to be present on two of the serine and three of the threonine residues (201) (Fig. 12.35). These O-glycans consist of a core N-acetylgalactosamine (GalNAc) that in most instances is linked to galactose and/or sialic acid (198,200). It is worth noting here that unlike humans, nonprimate mammals including rodents have only a single type of IgA that lacks a hinge region and is therefore structurally closer to human IgA2 than IgA1. As will become evident below, this limits the usefulness of rodent models in our understanding of the pathogenesis of IgA nephropathy and, in particular, the glomerular deposition of IgA.
There are two main sites of IgA production. Most IgA is produced at mucosal sites; the great majority of this becomes contained within mucosal secretions and very little enters the systemic circulation. The transepithelial transport of IgA from submucosal plasma cells into the lumen is accomplished via the polymeric immunoglobulin receptor (pIgR), which is expressed on the basolateral membrane and binds J chain-containing immunoglobulins (IgA and IgM). After binding, the receptor-immunoglobulin complex is internalized and the immunoglobulin is ultimately secreted into the lumen, still retaining a portion of the pIgR known as the secretory component. As the J chain is essential for this transepithelial transport of IgA, essentially all mucosally derived IgA, including both IgA1 and IgA2, is pIgA (200,202). The remainder of human IgA is produced mainly in the bone marrow and secreted into the blood. The great majority of this marrow-derived IgA is monomeric IgA1 (mIgA1) (198).
TABLE 12.9 Major differential diagnosis of dominant/codominant mesangial IgA deposits | ||||||||||||||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| ||||||||||||||||||||||||||||
Properties of Circulating and Mesangial IgA in Patients With IgA Nephropathy
Total serum IgA concentration is elevated in one third to one half of the adults with IgA nephropathy and a somewhat higher percentage of children (54,203). However, high
concentrations of serum IgA do not necessarily lead to mesangial IgA deposition, as evidenced by the lack of such deposits in most patients with IgA multiple myeloma (204,205). Indeed, the IgA in the serum of patients with IgA nephropathy shows a number of atypical features. First, the increase in serum IgA, when present, is accounted for by an increase in pIgA1, with levels of mIgA1 and of IgA2 being within the normal range (200,206). Second, there is an elevated ratio of lambda to kappa light chains in this pIgA1, particularly in patients with active disease (207,208). Third, there is altered O-glycosylation (galactosylation and sialylation) in the hinge region of this IgA1, as demonstrated both by altered lectin binding and mass spectroscopy, with no abnormalities of the N-linked sugars noted (209,210).
concentrations of serum IgA do not necessarily lead to mesangial IgA deposition, as evidenced by the lack of such deposits in most patients with IgA multiple myeloma (204,205). Indeed, the IgA in the serum of patients with IgA nephropathy shows a number of atypical features. First, the increase in serum IgA, when present, is accounted for by an increase in pIgA1, with levels of mIgA1 and of IgA2 being within the normal range (200,206). Second, there is an elevated ratio of lambda to kappa light chains in this pIgA1, particularly in patients with active disease (207,208). Third, there is altered O-glycosylation (galactosylation and sialylation) in the hinge region of this IgA1, as demonstrated both by altered lectin binding and mass spectroscopy, with no abnormalities of the N-linked sugars noted (209,210).
 FIGURE 12.29 IgA-dominant postinfectious glomerulonephritis. Immunofluorescence for IgA shows staining in the mesangium and peripheral glomerular capillary walls, with a “starry sky” pattern and segmental distinct deposits corresponding to subepithelial “humps.” (FITC anti-human IgA, ×400.) |
 FIGURE 12.30 IgA-dominant postinfectious glomerulonephritis (resolving phase). Electron micrograph shows subepithelial deposits in the “notch” region underlying the mesangium (lower portion of figure) and occasional mesangial deposits. (Uranyl acetate and lead citrate, original magnification ×5000.) |
 FIGURE 12.31 Superimposed membranous and IgA nephropathies. Methenamine silver stain shows glomerular basement membrane “spikes” typical of membranous nephropathy (best seen between 12 and 2 o’clock), with segmental mesangial hypercellularity (lower left). (Jones methenamine silver stain, ×1000.) |
 FIGURE 12.32 ANCA-associated crescentic glomerulonephritis with mesangial IgA. Portions of the glomerular tuft not directly adjacent to the hilum or to the crescent are not hypercellular. (PAS, ×400.) |
These features of serum IgA in patients with IgA nephropathy correlate well with those of IgA deposited in the mesangium. As noted earlier, the glomerular deposits of IgA in these patients consist predominantly of IgA1 (128,133,134,135,136,137,138,139,140), and much of this is polymeric based on the
presence of J chain (134,135,138,141) (see Immunohistologic Findings). The intensity of immunostaining for lambda light chains also exceeds that for kappa in the majority of cases of IgA nephropathy (130,139,148,211). Furthermore, when the O-glycosylation of IgA1 eluted from kidneys (nephrectomy or autopsy specimens, or combined from a large number of biopsy specimens) with IgA nephropathy was studied using lectin binding and mass spectrometry, there was decreased galactosylation and sialylation in the hinge region that was even more pronounced than that seen in serum IgA1 from IgA nephropathy patients (212,213). These findings suggest that the most abnormally O-glycosylated pIgA1 molecules are those most likely to be deposited in the glomeruli, and indeed, fractions of IgA1 from IgA nephropathy patients enriched in molecules with underglycosylated hinge regions correspond to those fractions showing the greatest degree of glomerular accumulation in the perfused rat kidneys (214).
presence of J chain (134,135,138,141) (see Immunohistologic Findings). The intensity of immunostaining for lambda light chains also exceeds that for kappa in the majority of cases of IgA nephropathy (130,139,148,211). Furthermore, when the O-glycosylation of IgA1 eluted from kidneys (nephrectomy or autopsy specimens, or combined from a large number of biopsy specimens) with IgA nephropathy was studied using lectin binding and mass spectrometry, there was decreased galactosylation and sialylation in the hinge region that was even more pronounced than that seen in serum IgA1 from IgA nephropathy patients (212,213). These findings suggest that the most abnormally O-glycosylated pIgA1 molecules are those most likely to be deposited in the glomeruli, and indeed, fractions of IgA1 from IgA nephropathy patients enriched in molecules with underglycosylated hinge regions correspond to those fractions showing the greatest degree of glomerular accumulation in the perfused rat kidneys (214).
 FIGURE 12.33 IgA nephropathy with fibrinoid necrosis and crescent formation. The methenamine silver stain shows fibroid necrosis with associated disruption of the glomerular basement membrane and associated early crescent formation, most evident in lower left quadrant of this glomerulus. In contrast to the case with ANCA-associated crescentic glomerulonephritis (see Fig. 12.32), the glomerular tuft shows segmental mesangial and endocapillary hypercellularity. (Jones methenamine silver, ×400.) |
 FIGURE 12.34 IgA nephropathy with segmental fibrinoid necrosis. In addition to the segmental necrotizing lesion, this glomerulus also shows mesangial hypercellularity and matrix expansion. (H&E, ×400.) |
The abnormal O-glycosylation of IgA1 is posttranslational, as the primary structure of the hinge region of IgA1 is normal in patients with IgA nephropathy (199). There is also no evidence that this abnormal glycosylation is due to post-secretory degradation of the O-linked glycans (209). Evidence for a functional defect in core 1, β1,3-galactosyltransferase 1 (C1GALT1), the enzyme that catalyzes addition of galactose to O-linked GalNac, was reported in peripheral blood B cells from patients with IgA nephropathy (200), and a recent study of Serino et al. (215) identified a possible molecular mechanism underling the reduced C1GALT1 activity. These investigators identified a microRNA (miR-148b) within human peripheral blood mononuclear cells (PBMCs) that acts to reduce levels of C1GALT1 mRNA and protein within these cells. Furthermore, Serino et al. (215) showed that PBMCs from patients with IgA nephropathy had significantly higher levels of miR-148b than PBMCs from health control subjects, whereas cells from patients with HSP nephritis, membranoproliferative GN type I, and focal segmental glomerulosclerosis (FSGS) did not. Transfecting cells from patients with IgA nephropathy with an inhibitor (a mutated form) of miR-148b restored expression of C1GALT1 mRNA and protein to normal levels. Furthermore, transfecting an IgA1-producing B-cell line with a molecular mimic of miR418b increased production of galactose-deficient IgA1, whereas transfection with the inhibitor reduced production of galactose-deficient IgA (215). While the basis for the increased levels of miR-148b remains unknown, these findings open a potential new avenue for the treatment of IgA nephropathy (215,216).
 FIGURE 12.35 O-glycosylation of the hinge region of IgA1. A: Structure of monomeric IgA1. There are two α1 heavy chains and two κ or λ light chains. The heavy chains contain three constant region domains (CH1, CH2, and CH3) and a variable (VH) region; the 18-amino-acid hinge region is located between the CH1 and CH2 domains. B: Amino acid sequence of the hinge region, indicating those serine and threonine residues that have been shown to be O-glycosylated. (Modified from Barratt J, Feehally J, Smith AC. Pathogenesis of IgA nephropathy. Semin Nephrol 2004;24:197-217.) |
Source(s) of the IgA Within Glomerular Deposits
Because the IgA deposited in the glomeruli of patients with IgA nephropathy is mainly pIgA1 and under normal conditions mucosal IgA is almost exclusively polymeric while bone marrow-derived IgA is nearly all monomeric, and noting the common association between infections involving mucosal surfaces and the onset of symptoms in IgA nephropathy, a number of investigators hypothesized that the mesangial IgA in patients with IgA nephropathy is mucosally derived (217,218,219). Furthermore, a number of studies have demonstrated increased serum levels of IgA antibodies directed against food and other mucosal-associated antigens (220,221,222). However, more recent evidence strongly suggests that abnormalities in IgA immune responses to mucosal antigens, including a loss of oral tolerance to these antigens and a shift in production of pIgA1 from plasma cells associated with mucosal tissues to those in the
bone marrow, most likely underlie the presence of circulating, abnormally glycosylated pIgA1 that ultimately becomes deposited in the glomeruli (200,206,223,224,225,226,227). A major portion of this evidence can be summarized as follows:
bone marrow, most likely underlie the presence of circulating, abnormally glycosylated pIgA1 that ultimately becomes deposited in the glomeruli (200,206,223,224,225,226,227). A major portion of this evidence can be summarized as follows:
Patients with IgA nephropathy have increased numbers of plasma cells containing pIgA and IgA1 in the bone marrow, and bone marrow cultures from these patients produce increased amounts of pIgA1 (as well as mIgA1) compared with age- and sex-matched controls (225,226,227).
By contrast, numbers of mucosal-associated plasma cells producing pIgA are not elevated, and in some instances have been found to be decreased, in patients with IgA nephropathy (228,229), and levels of IgA and of IgA1 are not increased in the saliva of patients with IgA nephropathy who have elevated serum IgA and IgA1 levels (206).
Patients with IgA nephropathy, unlike healthy controls, respond to intranasal antigen challenge with influenza virus with increased levels of IgA antibodies in the circulation (230), providing an explanation (in the form of abnormal handling of mucosal antigens and breakdown of oral tolerance) other than spillage of mucosal IgA into the circulation for the presence of IgA antibodies against food and other mucosal environmental antigens in the serum of patients with IgA nephropathy (220,221,222).
It remains unclear why the abnormal production of IgA1 by the bone marrow of patients with IgA nephropathy involves pIgA1, which is more characteristic of mucosal secretions, as well as mIgA1. Perhaps, the apparent dysregulation of the “mucosa-bone marrow axis” of IgA production associated with IgA nephropathy somehow also involves J chain expression, although this is speculation.
Mesangial Deposition of IgA in IgA Nephropathy
The mesangial deposits of IgA in IgA nephropathy could potentially result from deposition of circulating immune complexes, binding of IgA to specific antigen(s) planted in the mesangium, and/or binding of IgA (or more specifically, abnormally glycosylated pIgA1) to specific receptors expressed on mesangial cells in the absence of antigen.
IgA nephropathy is commonly described as an immune complex-mediated glomerular disease. It was initially hypothesized that the IgA deposited in the mesangium is derived from circulating immune complexes (219,231,232,233). Support for this hypothesis comes from observations in both humans and experimental models. In humans, episodes of macroscopic hematuria and other clinical and histologic features of active glomerulonephritis in some patients (mainly children) with IgA nephropathy have been found to be associated with an increase in circulating IgA immune complexes (219,232,234), although such evidence of active disease has also been found to be associated with increases in serum pIgA not in the form of immune complexes (235). The frequent recurrence of mesangial IgA deposits in renal allografts of recipients whose original disease was IgA nephropathy (236) (also see Renal Transplantation in Patients With IgA Nephropathy, below) is also consistent with a scenario of mesangial deposition of circulating immune complexes, although nephritogenic aggregates of abnormal IgA not configured as immune complexes with bound antigen could also explain the recurrence. The frequent presence of IgA deposits in dermal blood vessels of patients with IgA nephropathy has also been taken as evidence for this scenario, although as noted previously (see Immunohistologic Findings) such deposits are not specific for IgA nephropathy and may be seen in a number of different renal diseases and even in normal individuals (152,155). In a mouse model, intravenous injection of pIgA-containing immune complexes produced a mesangial pattern of IgA and C3 immunostaining with focal glomerulonephritis and associated hematuria (237).
In patients with IgA nephropathy without circulating IgA immune complexes, it has been hypothesized that such complexes form in situ, perhaps with antigens planted in the mesangium (231). A number of potential antigens have been proposed to be involved in the pathogenesis of IgA nephropathy, including food antigens (gliadin, casein, ovalbumin, bovine serum albumin, and others) (220,221,222,238,239), viral antigens (cytomegalovirus, hepatitis B surface antigen) (240,241,242), and bacterial antigens (e.g., Haemophilus parainfluenzae, S. aureus cell envelope antigen) (243,244,245). While findings with food and viral antigens have generally not proven to be consistent or highly reproducible (231,246,247,248,249,250), those with bacterial antigens may be somewhat more promising. For example, Koyama et al. (244) reported colocalization of S. aureus cell envelope antigen (“possible adhesion”) and IgA by immunofluorescence in 79 of 116 (68%) renal biopsy specimens from patients with IgA nephropathy and 6 of 10 specimens from patients with HSP nephritis but in none of 76 specimens lacking glomerular immune complex deposits. This antigen was found, however, in 3 of 10 biopsy specimens with lupus nephritis (244). Recently, Schmitt et al. (251) demonstrated tissue deposits of IgA-binding regions (IgA-BR) of staphylococcal M proteins in the majority of renal biopsy specimens from patients with both IgA nephropathy and HSP nephritis, although there was no correlation between the presence of IgA-BR and clinical or histopathologic findings.
Related posts:
Development of the Kidney
Membranous Glomerulonephritis
Renal Involvement in Polyarteritis Nodosa, Kawasaki Disease, Takayasu Arteritis, and Giant Cell Arteritis
Pauci-Immune and Antineutrophil Cytoplasmic Autoantibody-Mediated Crescentic Glomerulonephritis and Vasculitis
Renal Disease Caused by Hypertension
Pyelonephritis and Other Direct Renal Infections, Reflux Nephropathy, Hydronephrosis, Hypercalcemia, and Nephrolithiasis
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
