RSS Feeds
Systemic Lupus Erythematosus: Emerging Concepts: Part 1: Renal, Neuropsychiatric, Cardiovascular, Pulmonary, and Hematologic Disease
- Dimitrios T. Boumpas, MD;
- Howard A. Austin, MD;
- Barri J. Fessler, MD;
- James E. Balow, MD;
- John H. Klippel, MD; and
- Michael D. Lockshin, MD
- From the National Institutes of Health, Bethesda, Maryland. Requests for Reprints: Dimitrios T. Boumpas, MD, National Institutes of Health, Building 10, Room 3N-112, Bethesda, MD 20892. Acknowledgments: The authors thank Drs. John L. Decker, Alfred D. Steinberg, and Paul H. Plotz for their leadership in planning and implementing studies of systemic lupus erythematosus at the National Institutes of Health; Dr. Nicholas J. Patronas for invaluable assistance and useful comments in the discussion of neuroimaging studies in patients with systemic lupus erythematosus; Dr. Catherine K. Chow for useful discussions on the role of computed tomography in the diagnosis of pulmonary disease; Drs. W. Travis and C.L. Phillips for their assistance with the interpretation of the pathology of Figure 3; and Ms. Lisa A. Miller and Mr. Andrew S. Lerner for manuscript preparation.
Abstract
Purpose: To review advances and controversies in the diagnosis and management of systemic lupus erythematosus with visceral involvement (renal, neuropsychiatric, cardiopulmonary, and hematologic disease).
Data Sources and Study Selection: Review of the English-language medical literature with emphasis on articles published in the last 5 years. More than 400 articles were reviewed.
Data Synthesis: Recent debates pertaining to lupus nephritis have focused on the value of kidney biopsy data and the role of cytotoxic drug therapies. Many studies have shown that estimates of prognosis are enhanced by consideration of clinical, demographic, and histologic features. For patients with severe lupus nephritis, an extended course of pulse cyclophosphamide therapy is more effective than a 6-month course of pulse methylprednisolone therapy in preserving renal function. Adding a quarterly maintenance regimen to monthly pulse cyclophosphamide therapy reduces the rate of exacerbations. Plasmapheresis appears not to enhance the effectiveness of prednisone and daily oral cyclophosphamide. Small case series have shown pulses of cyclophosphamide to be beneficial in patients with lupus and neuropsychiatric disease refractory to glucocorticoid therapy, acute pulmonary disease (pneumonitis or hemorrhage), and thrombocytopenia. Patients with systemic lupus erythematosus have an increased prevalence of valvular and atherosclerotic heart disease, apparently because of factors related to the disease itself and to drug therapy.
Conclusions: Cytotoxic agents are superior to glucocorticoid therapy for the treatment of proliferative lupus nephritis, but the optimal duration and intensity of cytotoxic therapy remain undefined. Definitive studies of the treatment of autoimmune thrombocytopenia and acute pulmonary disease and of the diagnosis and treatment of neuropsychiatric disease are not available.
Systemic lupus erythematosus is an extraordinarily complex autoimmune disease that touches on nearly all medical subspecialties [1]. Evidence from a broad range of basic science studies indicates that the pathogenesis of this disease is equally complex and may vary from patient to patient. The diverse expression of the common lupus syndrome may result from variable abnormalities in intersecting genetic, immunologic, hormonal, and environmental pathways. Although many uncertainties about pathogenic mechanisms remain, recent advances in diagnosis and treatment have substantially improved the prognosis of patients with systemic lupus erythematosus. As mortality rates decrease, issues such as comorbidity, complications of therapy, and overall quality of life are receiving increased attention.
We discuss recent advances in systemic lupus erythematosus. By necessity, this review is not comprehensive; we focus on changing concepts and new information. In this, the first part, we review issues related to the diagnosis and management of systemic lupus erythematosus with visceral involvement. In the second part, to be published in the 1 July issue, we examine selected topics related to dermatologic and joint disease, as well as issues related to the antiphospholipid antibody syndrome, pregnancy, hormonal therapy, and morbidity and mortality. We conclude with an overview of recent advances in the pathogenesis of the disease.
Renal Disease
The kidney is the viscus most commonly affected by systemic lupus erythematosus. With the use of sensitive light, electron, and immunofluorescence microscopy, at least modest abnormalities are seen in kidney biopsy specimens from almost all patients with lupus. Approximately 75% of renal biopsy specimens reported in several series have been classified as focal proliferative, diffuse proliferative, or membranous glomerulonephritis [2].
Pathogenesis
Localization of immune complexes in the kidney appears to be the inciting event for the development of lupus nephritis. Autoantibodies that react with DNA and other cellular components are characteristic of human and murine systemic lupus erythematosus, but only a subset of the resulting immune complexes seems to be nephritogenic. Studies correlating the immunochemical properties of autoantibodies with the type and severity of nephritis have detected several features that may promote pathogenicity, including quantity, charge, class, isotype, idiotype, avidity for DNA, and efficiency of complement fixation [3]. Furthermore, cross-reactivity of anti-DNA autoantibodies with glomerular cell surface antigens, as well as with normal components of basement membrane and mesangial matrix, probably promotes glomerular immune complex formation and influences the location of these deposits within the glomerulus [4]. Thus, factors that lead to the deposition of many proinflammatory immune complexes in the subendothelial region of the glomerular capillary wall, adjacent to the circulation, are likely to induce (through release of complement components, cytokines, and other factors) cellular proliferation, an inflammatory response, necrosis, and eventually fibrosis [5]. Furthermore, a subset of autoantibodies may penetrate glomerular cells, bind to nuclei, and contribute to glomerular proliferation and proteinuria [6].
The subepithelial immune deposits characteristic of lupus membranous nephropathy probably evolve through the in situ interaction of autoantibodies with antigens, such as DNA or histones, that bind to glomerular basement membrane because of their affinity for basement membrane components such as fibronectin, collagen, laminin, and heparan sulfate [7]. Subepithelial immune complexes induce relatively little cellular proliferation or inflammatory response. Glomerular capillary wall injury appears to be induced by complement activation and formation of the membrane attack complex, C5b-9, that has been associated with this type of active immune complex-mediated injury [8].
Diagnostic Studies
Laboratory Evaluation
Serologic variables have been extensively evaluated as indicators of the activity of lupus nephritis. Serum complement abnormalities have correlated with the degree of renal histologic activity in several studies [9, 10]. Persistent C3 or CH50 complement depression has been associated with progression of kidney disease in some, but not all, groups of patients [10-12]. Antinuclear and anti-DNA antibody levels have been less consistently related to features of active glomerulonephritis [13]. Serologic abnormalities may develop many months before evidence of clinical renal involvement and should prompt close observation to detect changes in urinary sediment and protein excretion rate, which are frequently considered stronger indications for modifications of therapy.
Standard kidney function variables (such as serum creatinine level and creatinine clearance) are insensitive indicators of change in glomerular filtration rate and are likely to underestimate the severity of glomerulonephritis [14]. More accurate assessments of glomerular filtration rate are obtained by using inulin or iothalamate clearances or by using creatinine clearance after blocking tubular secretion of creatinine by cimetidine [15]. Nonetheless, even these measures of kidney function may fail to detect the extent of renal parenchymal injury because of intrarenal hemodynamic compensatory mechanisms that have been shown in animal models to augment filtration in perfused glomeruli [16].
Renal Biopsy
A classic clinical syndrome (for example, rapidly progressive glomerulonephritis) may, in some cases, obviate the need for a kidney biopsy to establish the type of lupus nephritis. Many patients, however, present with clinical features compatible with several of the classes of lupus nephritis for which different treatment strategies are usually recommended. In these patients, renal biopsy data may clarify an ambiguous situation and help to justify various therapeutic options. In the absence of significant proteinuria or urinary sediment abnormalities, we are usually reluctant to recommend renal biopsy outside the context of a research protocol.
Deliberations about treatment usually include an assessment of prognosis. Hypertension has been associated with renal disease progression and death [17]. The contribution of kidney morphologic evaluation to estimates of prognosis has been debated vigorously. Detailed records of the duration of nephritis [18] or the rate of change of renal function [19] provide an indication of the balance of reversible and irreversible injuries that may have occurred. Kidney biopsy data provide a more direct appraisal of the type of renal disease and have enhanced outcome predictions based on clinical data in several [20-23], but not all [18, 24], studies. Variations in conclusions may relate to the salutary effects of contemporary treatments as well as to differences in patient selection criteria, prognostic factors, and outcome measures studied. Several investigators have observed the prognostic effect of markedly active histologic features (such as cellular crescents, fibrinoid necrosis, and subendothelial immune deposits) combined with chronic, irreversible morphologic attributes (such as interstitial fibrosis, tubular atrophy, and glomerular sclerosis) [22, 25, 26].
Treatment
Glucocorticoids
A mainstay of the treatment of systemic lupus erythematosus, glucocorticoids are often used alone as initial therapy for patients with lupus nephritis. Prednisone at low to intermediate doses is usually sufficient for patients with mesangial and mild focal proliferative glomerulonephritis. Studies now in progress are evaluating the effectiveness and toxicity of prednisone therapy given on alternate days and of other treatment strategies for patients with membranous lupus nephropathy [27].
Patients with diffuse proliferative or severe focal proliferative glomerulonephritis are candidates for vigorous immunosuppressive treatments intended to control intrarenal inflammation. In some cases, this control can be achieved by using daily, high-dose prednisone (1 mg/kg of body weight daily) for approximately 2 months and then tapering the dose to reduce the risk for glucocorticoid-associated toxicities. The systemic effects of glucocorticoids are well recognized. High-dose glucocorticoids may promote glomerular scarring by augmenting glomerular capillary perfusion pressures [28] and by elevating low-density lipoprotein (LDL) cholesterol levels, leading both to enhanced mesangial cell uptake of oxidized LDL cholesterol and to cellular injury [29].
Pulse intravenous methylprednisolone has been used as an intensive initial therapy for patients with lupus nephritis and other immune-mediated disorders. Favorable short-term results have been observed [19, 30], but this therapy has been less extensively studied as a maintenance therapy for chronic disorders such as lupus nephritis [31]. The effectiveness and toxicity of daily high-dose oral prednisone and pulse intravenous methylprednisolone therapy have not been rigorously compared in patients with lupus nephritis.
Cytotoxic Drugs
Immunosuppressive drug regimens that include cytotoxic drugs are more efficacious than prednisone alone in controlling clinical signs of active nephritis [32, 33], in preventing renal scarring [34], and ultimately in reducing the risk for end-stage renal failure, but they have not been shown to be more effective in reducing the risk for death [35, 36]. Among cytotoxic drug regimens, intermittent pulse cyclophosphamide therapy appears to have one of the most favorable therapeutic indexes [35-38]. Nonetheless, it has been recognized that intravenous pulse cyclophosphamide treatments are complicated, costly, inconvenient, uncomfortable, and potentially toxic.
Given these concerns, additional studies have been done to address several questions. First, could an initial intensive immunosuppressive regimen shorten the duration of treatment? Second, is there an advantage to sustained immunosuppressive therapy for lupus nephritis? And third, what are the long-term toxicities of intermittent pulse cyclophosphamide therapy?
According to a recent study [39], exacerbations of active renal and extrarenal disease are significantly more likely to occur in patients receiving an intensive 6-month course of pulse cyclophosphamide than in those receiving a more sustained 30-month course (Figure 1). However, pulse cyclophosphamide therapy may be associated with substantial side effects. The risk for fatal opportunistic infections of the pulmonary and central nervous systems among patients with active systemic lupus erythematosus treated with high-dose corticosteroids, cytotoxic drugs, or both has been emphasized [40]. Bladder cancer has been seen in patients treated with daily oral cyclophosphamide but has not been seen in our patients with lupus receiving pulse cyclophosphamide therapy [35, 39], presumably because of the protective effects of hydration, a dilute diuresis, and mesna (2-mercaptoethanesulfonate). Hematologic malignancies have been reported infrequently after intravenous cyclophosphamide therapy for lupus nephritis [41, 42]. A large, multicenter study is needed to determine the effect of pulse cyclophosphamide therapy on the risk for malignancy in patients with systemic lupus erythematosus. Sustained amenorrhea was seen in 11 of 39 women younger than 40 years of age who were treated with pulse cyclophosphamide therapy for systemic lupus erythematosus [43]. The risk for sustained amenorrhea is significantly related to both the patient's age at the start of pulse therapy and the number of doses of cyclophosphamide received. These data, coupled with observations about the risk for relapse after a short course of pulse cyclophosphamide, pose a challenging problem. Recurring inflammation may result in sclerosing lesions that increase the risk for progressive renal failure during subsequent episodes of active nephritis. A marked improvement in renal status (resolution of initial elevations in serum creatinine levels, low-grade proteinuria, and resolution of a nephritic urinary sediment) favor a shortened course of cytotoxic drug therapy [24]. On the other hand, there is concern that discontinuing immunosuppressive therapy as soon as renal status is improved may increase the risk for relapse. Consequently, it has been recommended that patients should receive quarterly pulse cyclophosphamide as maintenance therapy for approximately 1 year after achieving clinical renal remission, but this approach has never been rigorously tested.
Alternatives to this regimen of pulse cyclophosphamide therapy are currently under investigation at the National Institutes of Health and other centers [44]. That more consistent and rapid responses could be achieved by combining pulse cyclophosphamide and pulse methylprednisolone therapies or by synchronizing pulse cyclophosphamide therapy and plasmapheresis has been suggested [45]. Plasmapheresis appears not to enhance the effectiveness of prednisone and daily oral cyclophosphamide [46]. Other experimental immunosuppressive regimens, including total lymphoid irradiation [47] and cyclosporine [48, 49], have been described but have not been prospectively compared with other options. There is particular interest in the development of more targeted treatments involving monoclonal antibodies and biologic-response modifiers; these are emerging as products of basic research [50, 51].
Neuropsychiatric Disease
Neuropsychiatric symptoms are common in patients with systemic lupus erythematosus and can be separated into primary events, which result directly from immune-mediated injury to the central nervous system, and secondary events, which result from disease in other organs, complications of therapy, or both (Table 1) [52-55]. Primary neuropsychiatric events (neuropsychiatric lupus) typically occur in a setting of clinical or serologic evidence of active systemic lupus erythematosus. Although relatively uncommon, systemic lupus erythematosus may first present with neurologic disease and should be included in the differential diagnosis of neuropsychiatric symptoms, especially when these occur in young patients [54].
Pathogenesis
Multifocal cerebral cortical microinfarctions associated with microvascular injury are the predominant histopathologic abnormalities attributed to neuropsychiatric lupus; vasculitis is rare [56]. The leading hypothesis about the pathogenesis of neuropsychiatric lupus invokes vascular occlusion because of vasculopathy, vasculitis (in rare cases), leukoagglutination or thrombosis, and antibody-mediated neuronal cell injury or dysfunction. Antibodies may be produced intrathecally or may gain access to the central nervous system from a blood-brain barrier disturbed by the vascular injury (Table 1).
Clinical Presentation
All parts of the nervous system (central, peripheral, or autonomic) may be involved (Table 2) [53]. There are many neuropsychiatric manifestations of systemic lupus erythematosus; the organic brain syndrome has received increased attention in recent years and is discussed here briefly.
Organic Brain Syndrome
This syndrome usually manifests with various degrees of memory impairment, apathy, and loss of orientation, intellect or judgment, but agitation, delirium, stupor, or coma may occur in severe cases. Symptomatic organic brain syndrome has been reported in as many as 20% of patients with systemic lupus erythematosus [55]. Limited cognitive impairment has been reported in 20% to 70% of these patients when examined using formal neuropsychologic testing [55, 57] and may occur independently of other manifestations of neuropsychiatric lupus or systemic disease activity [58, 59]. Slow but progressive deterioration may occur in a few patients and results in severe, debilitating dementia. Dementia may be caused by active ongoing neuropsychiatric lupus or may result from previous insult to the central nervous system or from multiple infarctions caused by antiphospholipid antibodies [55, 60].
Diagnostic Studies
Analysis of the cerebrospinal fluid is one of the most helpful diagnostic studies. Although elevated cell counts, protein levels, or both are found in only about one third of patients, analysis of the cerebrospinal fluid helps exclude acute or chronic infectious meningitis. Elevated intrathecal IgG, IgM, or IgA indices (or a combination of these) and oligoclonal bands have been seen in patients with diffuse neuropsychiatric disease but are not specific to neuropsychiatric lupus [55]. More specialized tests, such as measurements of interleukin-6 and interferon-α, are not readily available in most places [54].
Autoantibodies
Antineuronal antibodies are present in the serum of as many as 75% of patients with systemic lupus erythematosus and neuropsychiatric lupus [53]. Similarly, antiribosomal P protein antibodies have been found in 45% to 90% of patients with systemic lupus erythematosus and psychosis or major depression in most [61], but not all [62], studies. These autoantibodies may be found in as many as 25% of unselected patients with systemic lupus erythematosus, and thus their presence must be interpreted with caution [55]. Antineuronal antibodies and antiribosomal P protein antibodies are more likely to be seen in patients with diffuse rather than focal central nervous system involvement. This suggests that the former may be autoantibody mediated [53, 55].
The specific antigens that induce the antineuronal antibodies are beginning to be identified. Hanson and colleagues [63] described a 50-kd antigen in the plasma membrane of brain synaptic terminals (neuronal compartments vital for normal brain function) that bound antibodies present in the sera of 19 of 20 patients with neuropsychiatric lupus. Antiribosomal P protein antibodies were initially found to be directed to cytoplasmic (ribosomal) proteins. More recently, however, reactive P peptides have also been found on the cell surface; this suggests that these autoantibodies may directly affect the function and viability of cells that express this antigenic target [64]. In case–control studies, anticardiolipin antibodies have been associated with specific neuropsychiatric features, including cerebrovascular accidents, multi-infarction dementia, seizures, intracranial arterial and venous thrombosis, chorea, and acute transverse myelitis [60, 65]. With the exception of cerebrovascular accidents, the association between these syndromes and antiphospholipid antibodies has not been firmly established.
Neuroimaging Studies
Magnetic resonance imaging shows greater contrast and detail than computed tomography, but it also shows more clinically silent abnormalities and incidental findings [66]. Computed tomography is sufficient for the initial diagnosis of most mass lesions and of intracranial hemorrhage demanding immediate intervention, and it requires less patient cooperation than magnetic resonance imaging [67].
The findings of magnetic resonance imaging in neuropsychiatric lupus reflect the underlying histopathologic findings of vascular injury and may involve the white or gray matter of the brain parenchyma (Figure 2). In general, patients with focal neurologic findings or focal seizures are more likely than patients with diffuse involvement to have abnormalities detected by magnetic resonance imaging. In most cases, this imaging cannot distinguish between primary and secondary neuropsychiatric events. It is also impossible to establish a diagnosis of neuropsychiatric lupus from neuroimaging studies in the absence of clinical history. Furthermore, in many cases, the correlation between magnetic resonance imaging findings and clinical presentation is poor. For example, interpretation of white matter abnormalities (which are often clinically silent) is problematic because their prevalence in patients with nonsystemic lupus erythematosus increases from 20% in persons younger than 50 years of age to 90% in persons older than 70 years of age [66].
Single-photon emission computed tomography has been used by several investigators to show a higher prevalence of abnormal cortical perfusion in patients with neuropsychiatric lupus than in patients with systemic lupus erythematosus and no neuropsychiatric symptoms [68]. Unfortunately, quantitative, rigorous assessments of the clinical utilities of magnetic resonance imaging and single-photon emission computed tomography have not been done in large case series or well-controlled comparison trials. Furthermore, data on the diagnostic accuracy of, the diagnostic and therapeutic effect of, and the change in patient outcomes attributable to these tests are not available [66, 69].
Management
Therapy for neuropsychiatric lupus differs according to the type of presentation, its severity, and the nature of the underlying process (for example, inflammatory or thrombotic). Unfortunately, decisions about immunosuppressive therapy must currently be made in the absence of data from randomized, controlled studies [54]. Glucocorticoids remain the first line of therapy for the major manifestations of neuropsychiatric lupus. In patients with severe disease or in those who do not respond to standard prednisone therapy, pulse methylprednisolone therapy may be helpful. Small case series [45, 70-72] have shown intravenous pulses of cyclophosphamide to be useful in patients with very severe disease (for example, those with strokes, cerebritis, acute transverse myelitis, and coma), in those who fail to respond to glucocorticoids, and in those who relapse while glucocorticoid therapy is being tapered. Plasmapheresis and intravenous immunoglobulin have also been proposed as adjunct therapies for life-threatening neuropsychiatric lupus, but only on the basis of case reports. In patients with thrombosis of intracerebral vessels (for example, cerebral vein thrombosis), and in the absence of generalized lupus activity or high-grade cerebrospinal fluid abnormalities, therapy should be based predominantly on anticoagulation [54]. The role of immunosuppressive therapy in these cases is not clear.
Cardiac Disease
Cardiovascular involvement has been receiving increased attention in patients with systemic lupus erythematosus. Recent prospective studies using advanced diagnostic methods have allowed the delineation of the prevalence and significance of discrete cardiac manifestations such as valvular disease, myocardial dysfunction, and pericardial disease [73-77]. In other studies [78, 79], coronary artery disease has been found to have a substantial effect on mortality and morbidity in patients with systemic lupus erythematosus; it accounts for as many as one third of all deaths seen in this population [78, 79].
Valvular Heart Disease
The spectrum of lupus-related valvulopathies has been expanded to include valve leaflet thickening with or without valve dysfunction (regurgitation or stenosis) in addition to the more characteristic valve lesions of Libman-Sacks endocarditis (nonbacterial verrucous endocarditis) [73]. The prevalence of lupus-associated valvulopathy ranges from 18% to 74% depending on the cohort of patients studied, the duration of the disease, and the mode of diagnosis (for example, autopsy studies or transthoracic or transesophageal echocardiography) [73, 76, 80]. Valvular abnormalities may progress to hemodynamically significant lesions that require valve replacement [73, 81]. Furthermore, an increased risk for infectious endocarditis has been reported in retrospective reviews of patients with systemic lupus erythematosus [82, 83]. Because valvular lesions (such as Libman-Sacks endocarditis) may be overlooked by echocardiography and do not necessarily correlate with physical examination findings, some have suggested that antibiotic prophylaxis be considered for all patients with systemic lupus erythematosus having dental or surgical procedures [84, 85], but this recommendation has not been widely accepted.
The pathogenesis of valvular heart disease in systemic lupus erythematosus is unknown, but contributing factors may include verrucous vegetations, fibrinoid degeneration of valve cusps, valvulitis, fibrotic scarring (possibly worsened by glucocorticoids), vasculitis, or rupture of chorda tendineae [81]. The relation between antiphospholipid antibodies and the development of cardiac valvular abnormalities is controversial. Valvular lesions have been seen with increased frequency in patients with the primary antiphospholipid syndrome and in patients with systemic lupus erythematosus and antiphospholipid antibodies [86-88]; this has led to speculation about a possible causal relation [80]. However, valvular abnormalities are also seen in patients with systemic lupus erythematosus who lack antiphospholipid antibodies; this suggests that additional pathogenetic factors may be operative in the development of endocardial lesions in patients with systemic lupus erythematosus [76, 89].
Accelerated Atherosclerosis
Atherosclerosis is emerging as a significant cause of death and illness in patients with systemic lupus erythematosus. The mortality rate from coronary artery disease in patients with systemic lupus erythematosus is estimated to be ninefold greater than predicted population-based rates [90]. Severe atherosclerotic narrowing of coronary arteries has been well documented on autopsy studies, even in patients younger than 35 years of age.
Although the pathogenesis of accelerated atherosclerosis is unknown, it is believed to be multifactorial. Traditional cardiac risk factors, such as hypertension, obesity, and hyperlipidemia, are observed with high frequency in patients with systemic lupus erythematosus. Fifty-three percent of these patients have three or more risk factors; this prevalence greatly exceeds the prevalence seen in a matched population [78]. Glucocorticoid-induced dyslipoproteinemia [91] and complications that result from disease involvement in other organ systems (for example, renal disease leading to hypertension and hyperlipidemia) may also potentiate the atherosclerotic process. Circulating immune complexes may promote intracellular cholesterol accumulation and therefore may be an additional compounding factor [92].
Pulmonary Disease
The involvement of the respiratory system in systemic lupus erythematosus is relatively common, and its clinical manifestations are diverse (Table 3). Primary pulmonary involvement caused by systemic lupus erythematosus may follow a variable course from incidental abnormalities noted on chest radiography or pulmonary function testing to acute or chronic disease. Acute pulmonary disease tends to develop in association with generalized lupus activity, whereas chronic pulmonary involvement may progress independently of disease activity in other organs.
Acute Pulmonary Disease
Acute Lupus Pneumonitis and Alveolar Hemorrhage
Alveolar hemorrhage and acute lupus pneumonitis are uncommon, life-threatening syndromes associated with systemic lupus erythematosus; they result from acute injury to the alveolar-capillary unit [99]. Acute lupus pneumonitis is characterized by the abrupt onset of fevers, dyspnea with hypoxemia, and patchy alveolar infiltrates on chest radiography, without evidence of an underlying infection. The alveolar hemorrhage syndrome, which is less common than acute lupus pneumonitis, has a similar presentation, except that it is associated with an acute decline in hemoglobin levels caused by bleeding within the lung [99].
Although they have not been the subject of prospective controlled studies, glucocorticoids are generally accepted as the first line of therapy for acute immune-mediated lung injury in patients with systemic lupus erythematosus. Pulse methylprednisolone therapy seems to be effective in treating various immune-mediated pulmonary hemorrhage syndromes [100]. The addition of azathioprine or cyclophosphamide is generally advocated in patients who are critically ill or unresponsive to glucocorticoids [101]. Plasmapheresis as an adjunct to immunosuppressive therapy has been used in rapidly deteriorating patients [45]. Therapy must be initiated expeditiously because these syndromes can accelerate rapidly, leading to respiratory failure and death. Overall mortality rates for both syndromes are high, ranging from 50% to 90% despite treatment [93, 102]. However, a recent case series of patients with systemic lupus erythematosus and pulmonary hemorrhage suggested that the prognosis for these patients may not be as grim as was previously thought and that early diagnosis combined with aggressive immunosuppression may favorably affect survival rates [102].
Acute Reversible Hypoxemia
A newly described syndrome of acute reversible hypoxemia has been reported in a subpopulation of acutely ill patients with systemic lupus erythematosus [94]. The exact pathogenesis of this syndrome is unclear. It has been postulated that the elevated levels of complement split products detected in the plasma of these patients may activate circulating neutrophils, which aggregate within the pulmonary vasculature, and partly account for the observed diminished oxygenation capacity. In this series, hypoxemia resolved and the level of complement split products decreased after treatment with glucocorticoids for extrapulmonary disease manifestations.
Chronic Pulmonary Disease
Interstitial Lung Disease
Chronic interstitial lung disease may develop as a consequence of acute pneumonitis or as an independent manifestation of systemic lupus erythematosus [99, 103]. Radiographic findings of interstitial lung disease may be more common than symptoms. Discriminating between inflammatory alveolitis (which is treatable) and chronic fibrosis is the primary difficulty in determining the appropriate therapeutic approach to chronic lupus pneumonitis. Bronchoalveolar lavage and gallium scanning were used to evaluate interstitial lung disease in the past, but their correlation with disease activity or potential therapeutic response has been questioned. High-resolution computed tomography is emerging as an important noninvasive tool in the evaluation of idiopathic pulmonary fibrosis because high-resolution computed tomographic patterns (such as a ground-glass appearance that suggests active inflammation or a reticular pattern with honeycombing that suggests irreversible fibrosis) have been shown to correlate with histologic findings on open lung biopsy and with response to therapy [104-106] (Figure 3). Although this has not been critically examined in interstitial lung disease associated with systemic lupus erythematosus, high-resolution computed tomographic scans with a pattern that suggests chronic fibrosis may help to avoid unnecessary immunosuppression.
Pulmonary Hypertension
Increasingly recognized as a complication of systemic lupus erythematosus, pulmonary hypertension in this condition closely resembles primary idiopathic pulmonary hypertension. The Raynaud phenomenon has been seen in as many as 75% of patients with systemic lupus erythematosus and pulmonary hypertension and in only 25% of the general population with systemic lupus erythematosus [107]. Serologic testing shows a high incidence of antiribonucleoprotein antibodies, rheumatoid factor, and antiphospholipid antibodies [107]. The pathogenesis of pulmonary hypertension associated with systemic lupus erythematosus is unknown, but the condition is believed to result from vascular occlusion caused by vasoconstriction, vasculopathy or vasculitis, platelet aggregation or thrombosis, and parenchymal lung disease [93, 107, 108]. Currently, no therapy has been proved efficacious for the treatment of pulmonary hypertension associated with either idiopathic or systemic lupus erythematosus. Prognosis is poor, and a steady decline in pulmonary function, despite treatment, is usual [109]. Although it has been little studied, heart-lung transplantation may be an option in patients with systemic lupus erythematosus, advanced pulmonary hypertension, and minimal disease activity in other organs [109, 110].
Hematologic Disease: Autoimmune Thrombocytopenia
Autoimmune thrombocytopenia occurs in as many as 25% of patients with systemic lupus erythematosus, and it may be severe (platelet count less than 20 × 109/L) in approximately 5% of these patients. A positive test result for antinuclear antibodies has been reported to occur in as many as 30% of patients with chronic idiopathic thrombocytopenic purpura, and this may present diagnostic and therapeutic problems [111]. The presence of high-titer antinuclear antibodies in combination with antibodies to extractable nuclear antigens (for example, Ro/SS-A, La/SS-B, ribonucleoprotein, or Smith) or anti-double-stranded DNA antibodies increases the likelihood that additional manifestations of systemic lupus erythematosus will develop at some future date [112].
Pathogenesis and Clinical Presentation
Elevated platelet-associated IgG is found in patients with systemic lupus erythematosus with and without thrombocytopenia. It is not clear whether this indicates platelet-specific antibody or adsorbed immune complexes [111]. Antiplatelet autoantibodies bind to one or more surface glycoproteins (usually the glycoprotein IIb-IIIa complex). The antibody-coated platelets are ingested by macrophages of the spleen, liver, lymph nodes, and bone marrow, which carry receptors for the Fc region of immunoglobulin. Phagocytosis is more efficient when the antiplatelet antibodies belong to the IgG1 and IgG2 subtypes [113]. Antibodies to specific target antigens (other than the glycoprotein IIb-IIIa complex) on platelets of patients with systemic lupus erythematosus and thrombocytopenia have also been described [114]. Several studies have documented an association between the development of thrombocytopenia and the presence of antiphospholipid antibodies [115]. Other case reports [116] have suggested an association between systemic lupus erythematosus and a syndrome similar to thrombotic thrombocytopenic purpura. It is difficult to differentiate between thrombotic thrombocytopenic purpura and the catastrophic occlusion syndrome in patients with antiphospholipid antibody. Patients with thrombocytopenia are at increased risk for bleeding either spontaneously or after trauma. Spontaneous bleeding is rare unless the platelet count is less than 5 × 109/L, an associated congenital or acquired defect in platelet function is present, or a coagulopathy is present [117, 118]. Although thrombocytopenia rarely causes a fatal hemorrhage in patients with systemic lupus erythematosus, it may be a marker of more aggressive and severe disease and has a substantial negative effect on the prognosis of systemic lupus erythematosus [79].
Treatment
Randomized controlled studies of therapy for autoimmune thrombocytopenia are not available, and treatment is based on extrapolation from studies of patients with idiopathic thrombocytopenic purpura and small case series in patients with systemic lupus erythematosus. Therapy is aimed at attaining a complete and lasting response. If this is unsuccessful, subsequent therapy should be directed toward maintaining “safe” platelet counts (20 × 109/L) [117]. Therapy may not be necessary in patients with no evidence of systemic lupus erythematosus activity in other organs, a platelet count of more than 20 to 30 × 109/L, and no evidence of coexisting coagulopathy or platelet dysfunction. For all other patients, glucocorticoids are usually recommended as initial treatment. A randomized clinical trial in children and adults with acute idiopathic thrombocytopenic purpura showed similar efficacy for low-dose (0.25 mg/kg daily) and high-dose (1.0 mg/kg daily) glucocorticoids (prednisone or prednisolone) [119]. Patients who fail glucocorticoid therapy or relapse during tapering may benefit from splenectomy, danazol, intravenous immunoglobulin, vinca alkaloids, or intermittent pulses of methylprednisolone, dexamethasone, or cyclophosphamide [111, 120-123]. Pulse cyclophosphamide therapy is particularly attractive in patients with attendant involvement of other organs (such as the kidneys, the central nervous system, and the lungs) [123]. For patients with severe refractory idiopathic thrombocytopenic purpura, combination chemotherapy with cyclophosphamide and prednisone combined with either vincristine, vincristine and procarbazine, or etoposide has been reported to be beneficial in a case series of 10 patients [124].
- Copyright ©2004 by the American College of Physicians
References
- 1.↵
- 2.↵
- 3.↵
- 4.↵
- 5.↵
- 6.↵
- 7.↵
- 8.↵
- 9.↵
- 10.↵
- 11.↵
- 12.↵
- 13.↵
- 14.↵
- 15.↵
- 16.↵
- 17.↵
- 18.↵
- 19.↵
- 20.↵
- 21.↵
- 22.↵
- 23.↵
- 24.↵
- 25.↵
- 26.↵
- 27.↵
- 28.↵
- 29.↵
- 30.↵
- 31.↵
- 32.↵
- 33.↵
- 34.↵
- 35.↵
- 36.↵
- 37.↵
- 38.↵
- 39.↵
- 40.↵
- 41.↵
- 42.↵
- 43.↵
- 44.↵
- 45.↵
- 46.↵
- 47.↵
- 48.↵
- 49.↵
- 50.↵
- 51.↵
- 52.↵
- 53.↵
- 54.↵
- 55.↵
- 56.↵
- 57.↵
- 58.↵
- 59.↵
- 60.↵
- 61.↵
- 62.↵
- 63.↵
- 64.↵
- 65.↵
- 66.↵
- 67.↵
- 68.↵
- 69.↵
- 70.↵
- 71.↵
- 72.↵
- 73.↵
- 74.↵
- 75.↵
- 76.↵
- 77.↵
- 78.↵
- 79.↵
- 80.↵
- 81.↵
- 82.↵
- 83.↵
- 84.↵
- 85.↵
- 86.↵
- 87.↵
- 88.↵
- 89.↵
- 90.↵
- 91.↵
- 92.↵
- 93.↵
- 94.↵
- 95.
- 96.
- 97.
- 98.
- 99.↵
- 100.↵
- 101.↵
- 102.↵
- 103.↵
- 104.↵
- 105.↵
- 106.↵
- 107.↵
- 108.↵
- 109.↵
- 110.↵
- 111.↵
- 112.↵
- 113.↵
- 114.↵
- 115.↵
- 116.↵
- 117.↵
- 118.↵
- 119.↵
- 120.↵
- 121.↵
- 122.↵
- 123.↵
- 124.↵
Related Clinical Content
