Chapter 30 Kidney Involvement in Systemic Lupus Erythematosus
The kidney is the most commonly involved major internal organ in patients with systemic lupus erythematosus (SLE). It is important to recognize that lupus kidney disease is not expressed as a simple phenotype. Indeed, there are extraordinarily broad spectra of both clinical manifestations and pathologic categories of lupus kidney disease. Not surprisingly, no single paradigm can be offered to define the pathogenesis of all forms of lupus nephritis. Most patients with renal involvement have immune complex–mediated glomerulonephritis, but extraglomerular tubulointerstitial inflammation and vasculopathy are also relatively common components of lupus nephropathy. Vasculopathy is mostly due to inadequately controlled hypertension, but SLE-related microvascular thrombosis and rarely vasculitis can contribute significantly to lupus renal disease.
Depending on the populations studied, nephritis occurs in about 50 to 75% of patients with SLE. In those affected, nephritis characteristically appears within the first year after diagnosis of SLE. Careful screening tests are critical because most patients present with asymptomatic urine abnormalities, such as hematuria or proteinuria, or with new onset or worsening hypertension. Nocturia (due to loss of renal concentrating capacity) and/or foamy urine (due to proteinuria) are common initial manifestations of renal involvement but are rarely recognized unless the patient is specifically queried by the clinician. Proteinuria reflects the extent of involvement of peripheral glomerular capillary loops and tends to increase incrementally from cases of mesangial to endocapillary proliferative to membranous forms of lupus nephropathy. The latter involves virtually all glomerular capillary loops and is characteristically accompanied by heavy, nephrotic range (>3.5 g/day) proteinuria.
Glomerular hematuria, recognized as dysmorphic erythrocytes in urine sediment, is common in lupus nephritis; it is usually accompanied by proteinuria except in early mesangial nephropathy where it is often an isolated finding. Full-blown nephritic syndrome (hematuria with cellular casts and variable proteinuria) is seen in 30 to 40% of patients, while rapidly progressive glomerulonephritis (doubling or more of serum creatinine within a 3-month period) is rare and accounts for less than 10% of initial presentations. Hypocomplementemia (especially C3) and anti-DNA antibodies are commonly found in proliferative forms of nephritis.
Classic cases of proliferative and membranous forms of lupus nephritis have distinct clinical presentations (Table 30.1). However, it is important to keep in mind that overlapping and mixed classes may coexist and that the classes are not static. Indeed, transitions among the various forms of lupus nephritis are common over time. The clinical presentation does not always predict the underlying histologic class of nephritis. This is especially true in treated patients where therapy may modify both the clinical and pathologic findings. In general, patients with mesangial nephritis have small amounts of proteinuria (<1 g/day) with hematuria but typically no cellular casts. Patients with proliferative nephritis have hypertension, nephritic urine sediment with various degrees of proteinuria (often at nephrotic range), low C3 and typically high titers of anti-DNA antibodies, whereas patients with membranous glomerulopathy have proteinuria often at nephrotic range but otherwise bland urine sediments; C3 tends to be normal, and anti-DNA antibodies when present are usually found in low titers.
Careful laboratory monitoring is essential for early detection of lupus nephritis. Urinalysis is the most important and effective method to detect and monitor disease activity in lupus nephritis but special efforts are usually necessary to obtain accurate urinary microscopy. A fresh clean catch, unrefrigerated, second-morning urine sediment should be stained and examined in the clinic, if possible. Alternatively, specimens should be flagged for expeditious processing in the service laboratory to minimize spuriously negative results due to breakdown of cellular casts. Hematuria (usually microscopic, rarely macroscopic) indicates inflammatory glomerular or tubulointerstitial disease. Erythrocytes are fragmented or misshaped (dysmorphic). Granular and fatty casts reflect proteinuric states, whereas red blood cell, white blood cell, and mixed cellular casts reflect nephritic states (Fig. 30.1). Broad and waxy casts reflect chronic renal failure.
Renal function evaluation should include estimation of glomerular filtration rate. This can be accomplished by formulas, such as the Cockroft-Gault formula1 or Modification of Diet in Renal Disease study (MDRD) formula,2,3 or from timed (usually 24-hour) creatinine clearance. Indeed, many clinical laboratories around the world are incorporating formula-based estimates of glomerular filtration rate (GFR) concurrently with measures of serum creatinine. Methods of estimating proteinuria vary among institutions with some strongly preferring 24-hour collections and others preferring the simpler protein/creatinine ratio on spot urine samples.4,5
In the absence of objective urinary abnormalities, renal biopsy has little value and ordinarily should not be performed. It rarely establishes the diagnosis of lupus, but is necessary for classifying renal pathology; this information that has important prognostic implications. In fact, it is recognized that knowledge of pathologic findings is a powerful impetus for clinicians to prescribe aggressive intervention, particularly when results of common clinical tests may be less compelling.1
Early biopsy (before treatment) is indicated in patients with nephritic urine sediment, glomerular hematuria with proteinuria higher than 0.5 to 1.0g/day, low C3 and/or positive anti-ds DNA, or proteinuria higher than 1.0 to 2.0g/day (especially if C3 is low and/or positive anti-ds DNA). Patients with clinical and laboratory evidence of severe lupus nephritis, including nephritic or nephrotic syndrome, azotemia, and hypertension, may not require a renal biopsy prior to treatment with cytotoxic drugs. On the other hand, in patients with concomitant serologic abnormalities (i.e., low C3, positive anti-ds DNA) or patients who had previous immunosuppressive treatment may be candidates for renal biopsy even with more subtle findings. A repeat biopsy during or after treatment is indicated for unexplained worsening of proteinuria (e.g., >2g/day increase if non-nephrotic at baseline or >50% increase if nephrotic), unexplained worsening of renal function (e.g., reproducible >33% increase in serum creatinine, representing a 25% decrease in GFR), or persistent glomerular hematuria with proteinuria higher than 2g/day or proteinuria higher than 3g/day (especially if C3 is decreased).
Following a succession of versions of the World Health Organization (WHO) classification of lupus nephritis, a novel approach has been recently been promulgated in an attempt to provide a more concise description of various lesions and classes of lupus nephritis1 (Table 30.2). From the major histologic classes, class IV nephritis is the most common (approximately 40%), while classes III and V follow with an approximate frequency of 25% and 15%, respectively (see Figs. 30.2–30.5). Transformation from one class to another can occur, both spontaneously and as a result of treatment. The additional features of glomerular disease activity (potentially reversible) and sclerosis (irreversible damage) should be considered in each class of lupus nephritis. This is done by a semiquantitative analysis (on a 0 to 3+ scale) of specific histologic features of activity and sclerosis. This information is captured by either indicating the predominant feature as a subheading (a, active; s, sclerosing; or a/s, mixed active and sclerosing) to each major class or in the widely used checklists of activity and chronicity indices (Table 30.2).1
|Histologic Classification||Activity and Chronicity Indices|
|Class I||Minimal mesangial proliferative LN||Activity index (lesions are scored 0 to 3+ with maximum score of 24 points)|
|Class II||Mesangial proliferative LN||Hypercellularity: endocapillary proliferation compromising glomerular capillary loops|
|Class III||Focal LN||Leukocyte exudation: polymorphonuclear leukocytes in glomeruli|
|Class III (A)||Active lesions: focal proliferative LN||Karyorrhexis/fibrinoid necrosis (weighted x2): necrotizing changes in glomeruli|
|Class III (A/C)||Active and chronic lesions: focal proliferative an sclerosing LN||Cellular crescents (weighted x2): layers of proliferating epithelial cells and monocytes lining|
|Class III (C)||Chronic inactive lesions with glomerular scars: focal sclerosing LN||Hyaline deposits: eosinophilic and PAS-positive materials lining (wire loops) or filling (hyaline thrombi) capillary loops|
|Class IV||Diffuse LN||Interstitial inflammation: infiltration of leukocytes (predominantly mononuclear cells) among tubules|
|Class IV-S (A)||Active lesions: diffuse segmental LN||Chronicity index (lesions are scored 0 to 3+ with maximum score of 12 points)|
|Class IV-G (A)||Active lesions: diffuse global LN||Glomerular sclerosis: collapse and fibrosis of capillary tufts|
|Class IV-S (A/C)||Active and chronic lesions: diffuse segmental proliferative and sclerosing LNActive and chronic lesions: diffuse global proliferative and sclerosing LN||Fibrous crescents: layers of fibrous tissue lining capsule|
|Class IV-S (C)||Chronic inactive lesions with scars: diffuse segmental sclerosing LN||Tubular atrophy: thickening of tubular basement membranes, tubular epithelial degeneration, with separation of residual tubules|
|Class IV-G (C)||Chronic inactive lesions with scars: diffuse global sclerosing LN||Interstitial fibrosis: deposition of collagenous connective tissue among tubules|
|Class V||Membranous LN|
|Class VI||Advanced sclerosis LN|
LN, lupus nephritis.
Sources: Histologic classification, adapted from International Society of Nephrology/renal Pathology Society (ISN/RPS) 2003 classification of LN, in Weening JJ, D’Agati VD, Schwartz MM, Seshan SV, Alpers CE, Appel c lupus erythemato-GB, et al. The sus revisited. Kidney Int 2004;65:521-30; indices adapted from Austin HA, Boumpas DT, Vaughan EM, Balow JE. Predicting renal outcomes in severe lupus nephritis: contributions of clinical and histologic data. Kidney Int 1994;45:544-50.
Fig. 30.3 A, Class II, mesangial proliferative lupus nephritis; glomerular capillary loops are mostly patent and of normal thickness, but the tuft shows increased mesangial cellularity and matrix (PAS stain). B, Ultrastructure of mesangial immune complex deposits (green arrows), which are typical of class-II lupus nephritis. CL, capillary lumen.
(Micrograph courtesy of Sharda Sabnis, MD, Armed Forces Institute of Pathology, Washington, DC.)
Fig. 30.4 A, Class III, focal proliferative lupus nephritis. A segmental area of solidification is observed (black arrow); this area shows fibrinoid necrosis and karyorrhexis with an early cellular crescent forming along Bowman’s capsule (hematoxylin and eosin stain). B, Class IV, diffuse proliferative lupus nephritis. The glomerulus is globally involved with endocapillary proliferation that compromises most of the capillary loops, and extensive fibrinoid necrosis and karyorrhexis are evident (hematoxylin and eosin stain). C, Class IV, diffuse proliferative lupus nephritis. The glomerulus shows irregular changes among different segments; wire loop lesions and hyaline thrombi (black arrow) represent massive subendothelial and intraluminal deposits of immune complexes; other tufts show variable degrees of proliferation and mesangial expansion. D, Ultrastructure of subendothelial immune complex deposits (green arrows) characteristic of both class III and IV lupus nephritis. CL, capillary lumen.
(Micrograph courtesy of Sharda Sabnis, MD, Armed Forces Institute of Pathology, Washington, DC.)
Fig. 30.5 A, Class V, membranous lupus nephritis; the capillary loops are nearly uniformly thickened with only a modest expansion of mesangial structures (periodic acid–Schiff stain). B, Ultrastructure of subepithelial immune complex deposits (white asterisks) characteristic of class V, membranous lupus nephritis. CL, capillary lumen.
(Micrograph courtesy of Sharda Sabnis, MD, Armed Forces Institute of Pathology, Washington, DC.)
As described above, it is important to establish the level of GFR that corresponds to the level of serum creatinine for baseline reference in any given patient with lupus nephritis. Following this, serum creatinine can be used to measure change in renal function, since it is the main variable to account for change in GFR. In clinical practice, changes in renal function are more important than absolute values of renal function and significant reproducible changes in serum creatinine (e.g., 25 to 33% increase) are of concern—even if the change occurs within the normal population range of serum creatinine.
Measurement of 24-hour protein excretion is the gold standard, although this method is cumbersome for patients and fraught with collection errors. Collections of urine containing creatinine concentrations that deviate significantly from population averages for males (∼20 mg/kg/day) or females (∼15 mg/kg/day) should raise suspicions about the adequacy of the urine collection. Spot urine protein/creatinine is a simpler method to estimate the severity of proteinuria, and is increasingly popular form of monitoring proteinuria.9–11 In general, the numeric ratio approaches the number of grams per day of proteinuria. For example, if the protein-to-creatinine ratio is 2.0, the 24-hour protein excretion is approximately 2.0 g/day.
Resolution of active urine sediment is a feature of renal remission, but to be clinically meaningful has to be reproducible and sustained for several weeks. Reappearance of cellular casts with significant proteinuria is an early and reliable predictor of renal relapse and in most patients usually precedes rises in anti-DNA titers or decreases in C3 by several weeks1.
Anti-DNA antibodies and C3 and C4 complement components are useful in monitoring activity of lupus nephritis and in guiding treatment. In general, changes in anti-DNA titers are more valuable than their absolute values. Patients with rising titers of anti-DNA antibodies warrant close monitoring for evidence of lupus activity. Complement has an important role in the pathogenesis of LN. Traditional measures of complement activity, such as CH50, C3, and C4, have low sensitivity and specificity because plasma levels reflect the result of the dynamic state of complement synthesis and consumption, both of which are increased during inflammation. C3 is more useful clinically than C4 because C4 deficiency is common in lupus patients, values of C3 and C4 are rarely discordant, and C3 levels correlate best with renal histology on repeat renal biopsies.
Activation of the complement system is characterized by the generation of activated breakdown products of precursor molecules. Complement breakdown products may be a better index of complement activation than factor levels, and there is a good rationale to use them as markers of disease activity. However, the available studies show conflicting results with markers of the classic, alternative, or final common pathways, showing correlation with activity in some but not in other studies. Some of this may result from methodologic differences, such as the use of plasma versus serum and differences in the definition of disease activity. Further work and large-scale trials are needed in this area to help further define appropriate complement split products for assessing lupus disease activity and to determine whether any of these can be used as a reliable biomarker. Plasma and urinary cytokines or chemokines or urinary podocytes may reflect lupus activity, but these tests are not used in routine clinical practice at present (Fig. 30.5).
Numerous demographic and clinical variables can affect prognosis. Careful assessment of the unique combinations of such risk factors in individual patients is essential to optimize long-term outcomes. Patient characteristics associated with bad outcomes include black race,13,14 azotemia, anemia, antiphospholipid syndrome,15,16 failure to respond to initial immunosuppressive therapy, and flares with worsening in renal function.1 Combinations of severe active (crescents and fibrinoid necrosis) with marked chronic changes (moderate to severe tubulointerstitial fibrosis and tubular atrophy, such as chronicity index >3) are particularly ominous.8,18 In general, patients with a greater number of risk factors carry worse prognoses, are less likely to respond to therapy, tend to respond more slowly, and thus, need more aggressive treatment.
The immunosuppressive treatment of proliferative lupus nephritis consists of a period of intensive immunosuppression (induction) followed by a longer period of less intensive maintenance therapy. Despite a consensus on the general approach, there are wide-ranging opinions on the details of either of these treatments.
The cornerstone of treatment of lupus nephritis is corticosteroid therapy. For induction it is used as high dose daily treatment (prednisone 0.5 to 1.0 mg/kg/d) or as bolus intravenous therapy (methylprednisolone 0.5 to 1.0 g for 1 to 3 days), most commonly in combination with other immunosuppressive drugs. There have been no controlled clinical trials proving the benefit of corticosteroids over supportive therapy; nonetheless, the long clinical experience amply demonstrates the value of steroids in the management of patients with lupus nephritis.
Lupus nephritis is the only major organ manifestation for which effective immunosuppressive treatment has been established in controlled clinical trials.19–32 However, many of these studies have been limited by generic problems, including small number of patients, diverse racial mixes and socioeconomic backgrounds, and relatively short follow-up. Advances in adjunctive treatments over time, such as the use of angiotensin antagonists to reduce proteinuria (an independent risk factor for progressive renal dysfunction) may have improved the overall outlook for patients with lupus nephritis, which further complicates comparison of various studies.
The National Institutes of Health (NIH) pulse cyclophosphamide regimen, established in a series of controlled trials over several decades,19,20,25,29 has been the standard against which other treatments were compared, either directly or indirectly. Early studies showed comparable efficacy of daily oral and monthly intravenous pulse cyclophosphamide therapy. However, the greater risks for cumulative toxicity with daily administration, particularly hemorrhagic cystitis and bladder cancer, has led to abandonment of essentially all but short courses (2 to 6 months) of daily cyclophosphamide therapy in lupus nephritis. The most recent NIH study showed that renal remission was achieved somewhat more rapidly with the combination of pulse methylprednisolone and pulse cyclophosphamide therapy.1 Extended follow-up (median 11 years) of this cohort demonstrated persistent benefit of cyclophosphamide-containing regimens compared to methylprednisolone alone,1 without added toxicity. Pulse cyclophosphamide is effective in most patients but seems to be less so in blacks.13,14
A 6-month course of monthly bolus cyclophosphamide is effective1 in inducing renal response in many patients, but maintenance therapy with prolonged courses of quarterly pulse cyclophosphamide is needed to maintain response.1 In an attempt to establish the optimal maintenance regimen, cyclophosphamide, azathioprine, and mycophenolate were compared as maintenance therapies after cyclophosphamide induction in a randomized controlled study in a mainly Hispanic and African-American population.1 Maintenance therapy with azathioprine or mycophenolate mofetil was as effective as quarterly cyclophosphamide in preserving renal function, but they appeared to be superior only when renal outcomes were combined with mortality (event-free survival).1 However, the median length of follow-up was less than 3 years. This is important because in previous studies most of the difference between various treatment regimens became apparent only after 5 years,1 emphasizing the importance of long-term follow-up to assess the real impact of any treatment in patients with lupus nephritis.
Protracted cyclophosphamide therapy decreases the flare rate and improves long-term outcomes, but is associated with significant treatment-related morbidities,1 most notably infertility, which is of major concern for the patients since the majority are women of child-bearing age. The risk of infertility increases with the cumulative dose and age of the patient.1 Therefore, several alternative regimens have been tested to replace or reduce the dose of cyclophosphamide, and there is a growing body of evidence that various immunosuppressive combinations are effective in the short and medium term in proliferative lupus nephritis.
The Euro-Lupus nephritis study has recently demonstrated the utility of a short-course, low-cumulative-dose cyclophosphamide regimen (cyclophosphamide 500 mg intravenously every 2 weeks for 3 months) followed by azathioprine maintenance in white patients.1 Long-term follow-up of the cohort revealed no difference between the low-dose and high-dose cyclophosphamide groups in the rate of renal impairment after 68 months of follow-up, with about 20% of the original cohort having some degree of renal impairment.1
Alternatives to pulse cyclophosphamide induction therapy commonly used at various centers around the world include azathioprine and mycophenolate mofetil. Mycophenolate mofetil (MMF) was claimed to be equivalent to daily oral cyclophosphamide in a small study of 42 patients with diffuse proliferative glomerulonephritis.1 In a more recent large randomized controlled study of 140 patients with lupus nephritis, MMF was at least as effective as pulse cyclophosphamide in inducing renal remission at 6 months and with fewer side effects in the MMF group. Short-term follow-up of the two groups did not show any significant differences in renal flares or end-stage renal disease.1
Most studies indicate that azathioprine adds marginally to the efficacy of prednisone alone.1 Thus, at the present time, azathioprine is used as primary therapy mainly in milder forms of lupus nephritis, in patients strongly opposed to use of cyclophosphamide, and as maintenance therapy after induction of remission with cyclophosphamide.
Given the methodologic differences in various studies, it is impossible to give an unequivocal interpretation of the available data to determine a “one-fits-all” strategy to treat proliferative lupus nephritis. Several effective treatment options are available for the short and medium term and physicians can make their decisions based on the demographic and clinical characteristics of their patients, their personal experience with various treatments, and patient preferences.
Various treatment options and practical recommendations for management of lupus nephritis are summarized in Table 30.3. Corticosteroids are effective in the acute control of nephritis and are included in all treatment regimens. It is important not to withhold corticosteroids for fear of complications, but rather to test regularly the feasibility of reducing doses (preferably to alternate day), and to be willing to substitute alternative immunosuppressive strategies. The goal of treatment is to induce sustained remission that can be defined as normal renal function (less than 30% worsening of serum creatinine from baseline), no proteinuria (or at least <1 g/day) and inactive urine sediment. The time to reach remission varies from patient to patient but can be prolonged in those with severe disease. Stabilization at 3 months and significant improvement with the ability to reduce the dose of corticosteroids at 6 months are good indications of effective therapy. Practical guidelines for administration of pulse cyclophosphamide therapy, ovarian protection and monitoring for the most common side effects of immunosuppressives can be found elsewhere in this book (Chapter 46).