Clinical Features and Treatment of Scleroderma

84 Clinical Features and Treatment of Scleroderma


Some consider that the first description of systemic sclerosis (scleroderma) was put forth in 1753 by Cario Curzio (Naples, Italy).1 However, a careful review of the reported case suggests that the diagnosis was really scleroedema because of the distribution of the skin changes and the fact that the 17-year-old female improved with bloodletting, warm milk, and small doses of quicksilver. In 1836, Fantonetti (1791-l877), a Milanese physician, became the first to use the word scleroderma to designate a skin disease in an adult. However, it is likely that this patient also had scleroedema. The first convincing case of scleroderma was reported in 1842; then several acceptable cases were published before 1847, when Gintrac used the term sclerodermie, establishing this condition as a specific clinical entity.2 By 1860, numerous cases had been reported, and articles reviewing the disease were published. Maurice Raynaud (1834-1881) described a patient with sclerodermie and cold-induced “asphyxie locale”—this was the first description of Raynaud’s phenomenon in scleroderma. Sir William Osler described scleroderma while at Johns Hopkins Hospital between 1891 and 1897.3 Osler recognized the systemic nature of the disease and was so impressed with the severity of scleroderma that he wrote:

Matsui (Japan, 1924) emphasized the importance of visceral involvement as part of scleroderma based on several autopsies. Goetz (Capetown, 1945) further confirmed the multisystem involvement and suggested that the disease be named progressive systemic sclerosis.4

The concept of subtypes of scleroderma began in 1964, when Winterbauer reported cases with the CRST (calcinosis, Raynaud’s phenomenon, sclerodactyly, and telangiectasia) syndrome.5 A similar group of patients was reported in 1920 and was named after the authors—the Thiberge-Weissenbach syndrome. Velayos and colleagues recognized that esophageal dysmotility was common in these patients, so it is now called the CREST syndrome.6 In 1969, 58 autopsy cases of scleroderma were compared with matched controls.7 The organs found to be frequently and significantly involved by this disease were the skin, gastrointestinal tract, lungs, kidneys, skeletal muscle, and pericardium. This report first described the systemic nature of scleroderma vascular disease with findings of both kidney and lung arterial changes. In 1979, Rodnan introduced a clinical method to evaluate the extent of skin disease.8 Steen and Medsger with others conducted extensive surveys of large populations of scleroderma patients, defining the clinical course and specific subtypes of disease. In the 1970s, an expert subcommittee established diagnostic criteria, and Leroy and colleagues suggested the classification of two major subsets of disease defined by skin involvement: limited and diffuse. Later, work by several investigators recognized that scleroderma has an autoimmune basis with occurrence of specific autoantibodies associated with subtypes of disease and useful in predicting disease course.

Scientific work in the modern era has revealed details regarding the pathogenesis of the disease and has led to the recognition that scleroderma is a complex polygenetic autoimmune disease associated with a unique disease process involving tissue fibrosis. Although no drug has yet been discovered with clear disease-modifying properties and ability to fully control the underlying disease process, major progress has been made in managing specific organ disease. The discovery that an angiotensin-converting enzyme inhibitor could reverse the scleroderma renal crisis in the 1970s has changed the course of kidney disease in scleroderma and has improved the survival of patients. Current therapies for gastrointestinal, cardiac, pulmonary, vascular, and interstitial lung disease have improved quality of life and survival.


Incidence and Prevalence

Scleroderma is a rare disease with an estimated incidence of approximately 18 to 20 cases per million population per year, and a prevalence of 100 to 300 cases per million population. Reported rates vary by method of ascertainment, population under study, and definition of disease. Scleroderma is found in all races and in various geographic areas, but the prevalence and severity of disease vary among different racial and ethnic groups. The prevalence of scleroderma appears to be greater in the United States, where it has been estimated at 24.2 per 100,000, than in Europe, where recent studies in Iceland, England, France, and Greece have estimated rates ranging from 7.1 to 15.8 per 100,000.9 In Australia, estimates are similar to those in the United States, and Japan has a lower reported prevalence—similar to that in Europe.10 The highest scleroderma prevalence is reported among Choctaw Native Americans, in whom the disease appears to be more severe.11 Several surveys in the United States show that African-Americans have a higher age-specific incidence rate and experience more severe disease than whites.12 Occurrence is not different from urban to rural areas. Occasional reports have described unusually large numbers of cases of scleroderma observed in restricted geographic regions, suggesting nonrandom distribution of the disease. For example, in a rural area in the province of Rome, a geographic cluster of scleroderma and disease with related features was reported, suggesting prevalence 1000 times greater than expected.13 Likewise, a study in the United Kingdom reported a higher prevalence of scleroderma in three geographic areas clustered near airports compared with other areas of the United Kingdom.14

The average age at onset is between 35 and 50 years, and the disease is more common among women (3 to 7 : 1 female-to-male ratio). Disease onset in the elderly is well described; it is uncommon for the disease to become manifest before the age of 25 years. Older age at onset is associated with increased risk for multisystem disease and, in particular, pulmonary arterial hypertension (PAH). A progressive increase in the incidence of scleroderma has been noted with increasing age. Over the period from 1963 to 1982, in Pittsburgh, Pennsylvania, the highest rates of scleroderma were observed between the ages of 45 and 54 years in black women, and between the ages of 55 and 64 years in white women.15 Younger age at disease diagnosis among black women compared with white women was reported from Detroit, Michigan.9


Mortality among scleroderma patients is high, and most deaths are attributed directly to the disease manifestations.16 In fact, the prognosis in scleroderma is highly variable, and survival is influenced by disease subtype, degree of internal organ involvement, and comorbid conditions. Factors associated with poor prognosis include diffuse skin disease, the presence of pulmonary disease (particularly PAH), renal involvement, multisystem disease, evidence of heart disease, older age at disease onset, and the presence of anemia. The standardized mortality rate (SMR) is the measure used to assess the relative mortality of a disease in comparison with the general population. Surveys of scleroderma patients report an SMR from 1.46 to 7.1. A survey found that among 284 scleroderma patients who died, the median disease duration as estimated from the onset of Raynaud’s phenomenon until death was 7.1 years for patients with diffuse skin disease, and 15.0 years for those with limited skin disease. Among non–scleroderma-related causes of death are, as expected, infection, malignancy, and cardiovascular events. Although premature atherosclerosis has been implicated as the cause of early death in other inflammatory autoimmune diseases, an increased prevalence of macroscopic coronary artery or cerebrovascular involvement beyond that expected in the general population has not yet been demonstrated in scleroderma.

Several studies suggest that the main cause of death has changed over time. Improved survival in recent years is thought to be secondary to more effective treatments for specific organ-based complications. It seems clear that the natural history of scleroderma renal complications has changed secondary to the use of angiotensin-converting enzyme (ACE) inhibitors for renal crisis. Historically, patients with scleroderma renal crisis had 1-year survival less than 15%.17 Recent case-control studies conducted after the introduction of ACE inhibitors suggest greater than 85% 1-year survival. Pharmacologic prevention and treatment of interstitial lung disease and PAH are thought to have an impact. Cohort studies have demonstrated improved overall survival among scleroderma patients. At one scleroderma center, 10-year survival from disease diagnosis improved from 54% in the 1972-1981 group to 66% in the 1982-1991 group.18 Another survey found that 5-year survival among diffuse cutaneous scleroderma patients improved from 69% in 1990-1993 to 84% in 2000-2003 (P = .018), whereas 5-year survival among limited cutaneous scleroderma patients remained unchanged at 93% and 91%, respectively.19

Environmental Exposures

Geographic clustering of scleroderma cases suggests that environmental exposure may be responsible for the disease, but definitive proof is still lacking. Epidemics of scleroderma-like illness following exposure to dietary (l-tryptophan or adulterated rapeseed oil), occupational (silica), or pharmacologic toxins (e.g., bleomycin, gadolinium-based contrast media) support the idea that environmental exposure can trigger a fibrotic disease. Typical scleroderma is found among coal miners and workers exposed to silica, particularly male workers. A meta-analysis of the risk of silica exposure found that the combined estimator of relative risk (CERR) was 1.03 (95% confidence interval [CI], 0.74 to 1.44) in females and 3.02 (95% CI, 1.24 to 7.35) in males.20 Although occupational exposure to solvents (paint thinner or removers, mineral spirits, trichloroethylene, trichloroethane, perchloroethane, gasoline, aliphatic hydrocarbons, halogenated hydrocarbons, and BTX solvents containing benzene, toluene, or xylene) or polyvinyl chloride is reported to be associated with scleroderma, the role of these agents in causing disease is controversial and remains unproven. Likewise, case reports have suggested that silicone present in breast implants has triggered scleroderma, but large epidemiologic surveys have not found a greater incidence of scleroderma than expected in women with breast implants or exposure to silicone. Drugs implicated as potentially causative for systemic sclerosis (SSc)-like illnesses include bleomycin, pentazocine, and cocaine.

Genetic Factors

Although the absolute risk for family clustering remains low (1%), studies from Australia and the United States suggest that a family history of scleroderma increases the risk of disease, with a relative risk of first-degree relatives around 14 and of siblings in the range of 15 to 19.21 Genetic factors were investigated by a survey of 42 twin pairs among which at least one twin had scleroderma.22 The overall concordance rate was low at 4.7%, and was similar between monozygotic and dizygotic twin pairs. Although an association between major histocompatibility complex (MHC) alleles and scleroderma has been reported, this has not been confirmed consistently.23 This finding argues against strong genetic susceptibility and favors a role for environmental and/or epigenetic events.

Overview of Clinical Features

Diagnostic Criteria

A subcommittee of the American College of Rheumatology (ACR) established diagnostic criteria based on a consensus of experts who evaluated a multicenter survey of scleroderma patients compared with other patient groups (Table 84-1).24 The purpose of these criteria was to provide diagnostic certainty and consistency for research and other documentary exercises. The single finding of thickening of the skin typical of scleroderma proximal to the metacarpophalangeal (MCP) joints of the hands is considered a major criterion confirming the diagnosis. This includes changes on the face, arms, legs, or trunk. If skin thickening is found only distal to the MCP joints, then two of three minor criteria (digital pits, sclerodactyly, and pulmonary fibrosis on chest radiograph) must be present to confirm the diagnosis. These criteria are very specific (98%) for making the diagnosis of scleroderma but are considered too restricted in that they exclude patients with early or mild expression of the disease. Many patients with skin thickening limited to the fingers and some with no skin changes are recognized to have systemic disease and yet do not meet the ACR criteria. Large longitudinal surveys confirm that a high percentage of patients with definitive Raynaud’s phenomenon and typical scleroderma nail-fold capillary changes, along with scleroderma-specific autoantibodies, develop scleroderma within a 2- to 4-year period of follow-up.25 Therefore, although the ACR criteria have a definite role in defining cases for research purposes, the clinician needs to take into account subtle features of the disease to make an early diagnosis for management purposes. Most experts would accept the finding of three of the five features of the CREST syndrome (calcinosis, Raynaud’s phenomenon, esophageal dysfunction, sclerodactyly, telangiectasias) to make a clinical diagnosis of scleroderma. In clinical practice, one needs to recognize that some patients will not develop skin changes and yet they do have systemic disease (systemic sclerosis sine scleroderma), and many patients will present with partial expression of the disease (e.g., Raynaud’s with nail-fold changes only) and later will develop well-defined disease. Experts argue that patients with partial expression, especially when a specific scleroderma-related autoantibody is present, should be considered to have a diagnosis of scleroderma and should be treated accordingly.

Table 84-1 Establishing Diagnosis of Scleroderma

ACR Criteria
Must have (a) or two of (b), (c), or (d)
CREST Criteria
Must have three of the five features
Minor Criteria*
Must have all three

ACR, American College of Rheumatology; CREST, calcinosis, Raynaud’s phenomenon, esophageal dysfunction, sclerodactyly, telangiectasias; CXR, chest x-ray; HRCT, high-resolution computed tomography; MCPs, metacarpophalangeal joints; MTPs, metatarsophalangeal joints; SSc, systemic sclerosis.

* Some experts continue to classify patients with minor criteria only as undifferentiated connective disease with scleroderma features.

Proximal areas include the face. Autoantibody includes anticentromere, antitopoisomerase (Scl-70), and anti-RNA polymerase III.

Classification and Clinical Subsets

The scleroderma disease process is complex, and its clinical expression is very heterogeneous such that the disease is expressed in several distinct clinical phenotypes. Classifying patients into subtypes is useful both for investigative purposes and for clinical practice. A specific subtype can define increased risk for internal organ involvement and overall prognosis. The traditional way to classify patients is by the degree of skin thickening as determined by physical examination. The skin is scored by pinching a fold and deciding whether it is abnormally thickened because of the scleroderma process. A committee of experts decided by consensus that two major groups of patients can be identified based on the distribution of skin changes and associated clinical and laboratory outcomes as observed (Figure 84-1).26 Patients are considered to have diffuse skin disease if skin changes are found proximal to the elbows and/or knees or on the trunk, excluding the face. These patients tend to have higher risk of multisystem disease and poor survival. Patients are considered to have limited disease if skin changes occur distal to the elbows and/or knees and not on the trunk. Facial skin thickening can be present in the limited group. Some argue that the term CREST syndrome should be eliminated, and that these patients should be classified in the limited skin group. Others feel that the CREST syndrome is a unique subtype within the broader group of limited skin diseases with a distinct disease course.

Another less popular classification system divides patients into three groups based on skin changes: limited (fingers only), intermediate (skin to the elbows or knees), and diffuse (skin above the elbows and/or knees and/or trunk). Studies using this classification have found that the intermediate group had distinct clinical outcomes, including an intermediate survival statistic between limited (best) and diffuse (worse).27 A subtype of disease with absence of skin fibrosis has been recognized and is referred to as scleroderma sine scleroderma. Serologic studies have shown that the presence of specific autoantibodies predicts clinical features of the disease, suggesting that the classification might be based on the autoantibody type (see “Natural History”). Patients with early or partial expression of the disease (e.g., Raynaud’s phenomenon alone with nail-fold capillary loop abnormalities) are often diagnosed as having undifferentiated connective tissue disease (UCTD). About 20% of these patients will develop clear features of scleroderma, usually within 2 to 4 years of follow-up.25

Scleroderma patients often have features of another rheumatic or autoimmune disease. These patients are considered to have an overlap syndrome. The most common overlap syndromes are polyarthritis, myositis, sicca complex, and hypothyroidism. Less common are primary biliary cirrhosis, autoimmune hepatitis, vasculitis, rheumatoid arthritis, and antineutrophil cytoplasm antibody (ANCA)-associated pauci-immune glomerulonephritis. Mixed connective tissue disease (MCTD) is an overlap syndrome, with features of scleroderma, polymyositis, lupus-like rashes, and rheumatoid-like polyarthritis.

Natural History of Disease

The natural history of scleroderma varies a great deal depending on the specific subtype of disease established in a patient. Scleroderma tends to be a chronic monophasic illness, unlike systemic lupus erythematosus, which is characterized by quiescence interrupted by sudden disease flares. The pattern of the disease and the degree or type of internal organ involvement are variable but generally can be predicted by the extent of skin affected. Internal organ disease is more severe in patients with the diffuse skin type than in those with limited scleroderma. The disease course is usually indolent in patients with sclerodactyly alone. In these patients, years pass from onset of Raynaud’s phenomenon before obvious organ dysfunction manifests, although explosive multisystem disease can emerge rapidly over several months in patients with widespread skin changes. Likewise, when patients with diffuse skin disease experience skin improvement, new internal organ disease is generally uncommon. The degree of skin thickening varies in extent and progression over time, with some cases stopping after several months with relatively mild disease, and others progressing over years to involve most of the body surface (see “Skin Involvement”).

Patients classified in the limited scleroderma group may have intermediate skin involvement, with skin extending over the hands and forearms or lower legs but not over the trunk. These patients appear to have an intermediate disease course that is not as severe as that of patients with widespread skin disease. The intermediate group has decreased survival compared with patients with only sclerodactyly. Although the degree of skin disease provides a rough guide by which to predict disease course, many exceptions have been noted. Variability is associated with other clinically defined factors, including age, gender, and racial/ethnic background. Late age at onset increases the risks for PAH and poor survival. Males may have a more aggressive disease course than females. African-American or Native American race predicts a worse disease course and worse multisystem disease.9

General Principles of Disease Evaluation

Measuring Disease Activity and Severity

One of the challenges in studying or caring for patients with scleroderma consists of determining and measuring disease activity, severity, and damage. These measures are not independent of each other and thus cannot be considered in isolation. Activity is a measure of ongoing disease that is dynamic and reversible; it is a direct reflection of ongoing biologic processes such as autoimmune-mediated inflammation or progressive tissue fibrosis. Severity is a measure of the impact of damage on the whole person or a specific organ at any given time. In the extreme, severity is a measure of irreversible end-stage damage. It is important to distinguish reversible disease activity from irreversible damage in that both can cause clinical distress, yet intervention differs for the two. Patients with scleroderma accumulate irreversible damage over time, thus the measure of activity is best done in early disease, when disease has the potential of being reversed.

Measurement of changes in skin thickness (modified Rodnan skin score) is used as a surrogate measure of disease severity and a predictor of extent of organ involvement and overall prognosis (see skin manifestation section).28 It is also used as a quantifiable, easily obtainable, and valid score, changing over time in response to therapy. Thus, the skin score is now used as a measure of disease activity and is employed in clinical trials as a primary outcome measure. In patients with diffuse skin disease, improvement in skin score is associated with better clinical outcome. A core set of items covering 11 domains (skin, musculoskeletal, cardiac, pulmonary, renal, gastrointestinal, health-related quality of life measures, global health, Raynaud’s phenomenon, digital ulcers, and the biomarkers erythrocyte sedimentation rate and C-reactive protein) is now considered to provide the minimum clinically relevant treatment effect values, as defined by consensus of experts in the field.29 More detailed scales for specific organs are also in use. For example, a skin assessment tool combines the modified Rodnan skin score with a visual analogue score (VAS) of the patient’s assessment of skin activity, a VAS score of the physician’s assessment of skin activity, and measures obtained by a durometer.30 A gastrointestinal scale has been developed to define mild to severe gastrointestinal disease.31 Experts who reviewed current outcome measures in assessing PAH as related to scleroderma decided that lung vascular and pulmonary arterial pressures and cardiac function as measured by right heart catheterization and echocardiography; exercise testing as measured by 6-minute walk and oxygen saturation with exercise, New York Heart Association functional class, severity of dyspnea on a VAS, and measures of quality of life and function obtained by Short Form 36 and the Health Assessment Questionnaire and Disability Index (HAQ-DI); and physician global scale as measured by survival are all important measures. Other outcome measures that are considered valid include Raynaud’s condition score for Raynaud’s severity, forced vital capacity, and Mahler’s dyspnea index to follow interstitial lung disease, serum creatinine, blood pressure, and complete blood count during a scleroderma renal crisis, as well as serum creatine phosphokinase to follow muscle disease.

A scleroderma severity scale has been developed by Medsger to assess disease severity status in individual patients both at a given time and longitudinally (Table 84-2).32 This scale is based on scoring that ranges from 0 (normal) to 4 (end-stage) for each organ system involved in scleroderma, including a general measure, along with measures of the peripheral vascular system, skin, joints and tendons, gastrointestinal tract, lung, heart, and kidney. This severity scale is now used by many experts to define the status of a patient in clinical trials and in research investigating risk for clinical outcomes.

The Health Assessment Questionnaire and Disability Index (HAQ-DI) is a self-administered tool initially developed for rheumatoid arthritis to measure functional impairment; it has been validated and used in scleroderma as well, where it correlates with objective physical signs and reflects the disease course. Higher degrees of impairment are found by HAQ scores in scleroderma patients with diffuse skin disease, higher skin scores, abnormal hand function, presence of myopathy or friction rubs, and joint pain.33 Change in the HAQ score occurs with changes in skin involvement and progressive organ disease.34 The HAQ is useful in evaluating patients with Raynaud’s and correlates with the severity and presence of digital ulcers. VAS instruments that assess disease in various domains specific for scleroderma are added to the HAQ to focus on scleroderma-specific issues, including measures of Raynaud’s and digital ulcers. In addition, the Short Form 36 survey, a simple self-report tool composed of 36 items in 8 domains, is used to assess health-related quality of life.

Developing an activity scale is much more challenging than establishing severity measures. Attempts include evaluating changes in skin score and evidence of active lung disease and documenting congestive heart failure or a scleroderma renal crisis. A composite score generated as a physician global assessment has been used to measure disease activity. This simple global score has merit but is influenced, besides activity, by organ severity and irreversible damage.35 A European study group developed the Valentini disease activity index, which currently is used in clinical studies.36 This index provides more of an organ-specific assessment than a measure of global activity. The Medsger disease severity scale has the potential to measure activity if used in studies with serial observations, when individual organ scales can be combined into a composite score. Intermediate clinical and serologic biomarkers are under investigation as candidates to define and predict disease activity and prognosis, but none has proved reliable or valid. Measures of acute phase factors (e.g., erythrocyte sedimentation rate, C-reactive protein) are nonspecific but are included by expert consensus in some systems designed to measure disease activity. Currently, no “gold standard” is available to measure disease activity in scleroderma, and defining biomarkers or other measures of disease activity remains a major challenge.


Measurement of the autoantibodies present in scleroderma can be helpful in determining the clinical features and prognosis of the disease, in that specific scleroderma-related autoantibodies have been established as strong predictors of disease outcome and the pattern of organ complications (Table 84-3).37 The three most frequently observed types of scleroderma-specific autoantibodies are anticentromere, antitopoisomerase I, and anti-RNA polymerase III antibodies.

Table 84-3 Autoantibodies and Associated Phenotypes in Scleroderma

Antigen Subtype Clinical Phenotype
Topoisomerase 1 (Scl70) Diffuse Pulmonary fibrosis, cardiac involvement
Centromere (protein B, C) Limited Severe digital ischemia, PAH, sicca syndrome, calcinosis
RNA polymerase III Diffuse Severe skin disease, tendon rubs, renal crisis (±sine scleroderma)
U3-RNP (fibrillarin) Diffuse or limited Primary PAH; esophageal, cardiac, and renal involvement; muscular disease
Th/To Limited Pulmonary fibrosis, rare renal crisis, lower GI dysfunction
B23 Diffuse or limited PAH, lung disease
Cardiolipin, β2GPI Limited PAH, digital loss
PM/Scl Overlap Myositis, pulmonary fibrosis, acro-osteolysis
U1-RNP Overlap SLE, inflammatory arthritis, pulmonary fibrosis

GI, gastrointestinal; GPI, glycoprotein I; PAH, pulmonary arterial hypertension; RNA, ribonucleic acid; RNP, ribonucleoprotein particle; SLE, systemic lupus erythematosus.

Adapted from Boin F, Rosen A: Autoimmunity in systemic sclerosis: current concepts, Curr Rheumatol Rep 9:165–172, 2007.

Anticentromere patients tend to be older white women. Skin disease usually is limited and most often involves just the fingers. The disease course is indolent, and often the diagnosis is delayed until anticentromere antibodies are discovered or late organ disease becomes evident. Patients frequently develop features of the CREST syndrome. Subcutaneous calcinosis is a late manifestation, appearing in areas such as small clusters on the fingers and along the forearm and anterior lower leg. Raynaud’s phenomenon can become severe and is associated with increased risk for macrovascular disease. Higher risk for digital gangrene and amputation is noted in the anticentromere–positive group. Interstitial lung disease is less common and is isolated and progressive; pulmonary vascular disease with PAH and right heart failure occur in a significant proportion of patients. Overall, however, anticentromere antibodies generally are predictors of a favorable prognosis. The presence of anticentromere antibodies can also be seen in patients with primary Sjögren’s syndrome who do not have scleroderma.

Antitopoisomerase I antibodies are seen in 30% of African-American patients and are correlated with a poor prognosis and higher scleroderma-related mortality. Interstitial lung disease (ILD) is highly associated with anti-topoisomerase, independent of the degree of skin disease. Patients usually have diffuse skin disease and are at risk for rapid skin changes and renal crisis, usually in the first few years from disease onset. Raynaud’s phenomenon may be the first symptom, followed in the first years of disease by skin changes and joint-tendon involvement leading to contractures, particularly in the fingers and elbows. The degree of skin thickening varies a great deal in these patients, but usually the level of severity of skin involvement is established in the first 1 to 3 years of disease.

Anti-RNA polymerase III antibodies are associated with rapid and aggressive diffuse skin disease and renal involvement. These patients have the worst cutaneous involvement with rapid widespread skin disease associated with signs and symptoms of deep tissue fibrosis involving joints, tendons, and muscles. Flexion contractures of fingers, wrist, elbows, shoulders, hips, knees, and ankles are complications that can occur within a few months of onset of disease activity. Friction rubs are common. A “fibrosing” myopathy without inflammation, along with skin and joint disease, leads to loss of strength and flexibility. Significant disability occurs in early disease. Anti-RNA polymerase III–positive patients are not likely to have severe gastrointestinal disease and are relatively protected from developing ILD or pulmonary vascular disease. However, they are at high risk (approximately 25% to 40% of patents) for developing scleroderma renal disease with hypertensive crisis. Risk is greatest in the early years, when skin is rapidly changing. Heart disease can be a late complication, but major disability from skin and deep tissue disease is the most worrisome long-term problem in this subtype of scleroderma.

Other antinucleolar antibodies are found among subgroups of patients with scleroderma and are associated with specific clinical phenotypes and clinical outcomes. Of these, anti-Th/To antibodies and anti-PM/Scl antibodies are associated with limited skin disease, whereas anti-U3-RNP (fibrillarin) antibodies are associated with diffuse disease. Anti-Th/To is at increased risk for development of severe ILD and pulmonary hypertension. Anti-U3-RNP is another predictor of a less favorable prognosis with a higher frequency of internal organ involvement, including interstitial lung disease, PAH, and renal crisis. It is frequently present in African-Americans and is associated with diffuse skin disease and poor prognosis. Hyperpigmented and hypopigmented skin changes are common, and contractures of large joints associated with inflammatory or fibrotic muscle disease lead to significant morbidity. Cardiac disease is often subclinical until late-stage disease, when both right and left heart failure can occur.

Anti-PM/Scl, anti-Ku, and anti-U1-RNP antibodies are seen mainly in patients with overlap syndromes. The presence of anti-Pm/Scl is associated with acute onset of weakness due to inflammatory muscle disease and higher risk for interstitial lung disease. Anti-Ku positivity has been demonstrated to be strongly associated with muscle and joint involvement. Anti-U1-RNP is seen in patients with mixed connective tissue disease and is common among African-Americans with limited scleroderma. Polyarthritis, myositis, and lupus with skin or renal involvement are common complications. A subset can develop diffuse skin disease, and risk for late onset of PAH is increased overall. Although an uncommon anti-B23 autoantibody is associated with pulmonary arterial hypertension, up to 11% of patients with scleroderma can test negative for antinuclear antibodies.

Clinical Manifestations and Treatment

General Principles of Management

Several basic principles can be of help in managing patients with scleroderma. The first is that scleroderma encompasses subsets of disease, each with a unique clinical phenotype; thus carefully determining the specific subtype of the patient sets the scene for deciding therapy. Clinical problems and disease activity in scleroderma are highly variable in expression, even within a given subtype. This variability is thought to be secondary to different susceptibility to, or risk for, systemic complications. It is also due to the presence of a dynamic biologic process unique to scleroderma. Therefore, patients move through stages of the disease process with disease activity that varies with time and can shift in nature. Signs of classic inflammatory events are often present during the initial phase of the disease, but later, an indolent subclinical fibrotic process is dominant and gradually causes organ damage. Although the skin may be the site of the most dramatic physical findings, scleroderma is a multisystem disease, and the physician must search for early organ-based complications, including cardiopulmonary, gastrointestinal, renal, or musculoskeletal involvement. Understanding the shortcomings of current investigations, not getting locked into “traditional” therapy, focusing on an organ-specific approach, and defining the disease subtype and level of disease activity are most important in establishing optimal management. Although no good studies have been conducted to prove the effectiveness of this approach, it seems reasonable to use combination therapy with rapid control of the inflammatory immune process early on, followed by maintenance immunosuppression for an extended time. The anti-inflammatory program is then coupled with organ-specific therapy (see later sections). The missing component in this comprehensive approach is the availability of a direct antifibrotic agent able to prevent the progression of fibrosis once the inflammatory phase is under control. Obviously, early intervention is most important for success; once irreversible tissue or organ damage is noted, supportive care becomes the main and often the only option.

Raynaud’s Phenomenon

Raynaud’s phenomenon (RP) is an exaggerated vascular response of the digital arterial circulation triggered by cold ambient temperature and emotional stress. The diagnosis of RP is based on a history of excessive cold sensitivity and recurrent events of sharply demarcated pallor and/or cyanosis of the skin of the digits (Figure 84-2). Blanching reflects digital arterial vasospasm, and cyanosis occurs secondary to the deoxygenation of sluggish venous blood flow. Some skin blushing (redness) may follow owing to reactive hyperemia after regular blood flow has been restored. Raynaud’s phenomenon occurs in 3% to 15% of the general population. It is more common among females (3 to 4 : 1) and is likely to begin before 20 years of age. During cold exposure (particularly during shifting temperatures and winter months), Raynaud’s attacks increase in frequency and intensity. Raynaud’s phenomenon can be categorized clinically into primary and secondary forms (Figure 84-3). Primary RP occurs when no disease process is associated with recurrent vasospastic events. Distinguishing primary RP from that associated with an underlying disorder is frequently challenging. Young age at onset (<20 years), symmetric manifestations of symptoms, mild to moderate severity, no association with digital ulceration or tissue gangrene, normal nail-fold capillary examination, and a negative antinuclear antibody (ANA) titer are all indicative of primary RP.38 Secondary RP occurs in a variety of settings, including connective tissues disorders and other rheumatic conditions, occupational trauma (e.g., hypothenar hammer syndrome), the use of certain drugs (e.g., antimigraine agents, ergotamine derivatives, bleomycin), increased blood viscosity, and compressive or obstructive vascular disease (e.g., thoracic outlet syndrome, atherosclerosis, thromboangiitis obliterans).

Nail-fold capillaroscopy is the tool most commonly used at the bedside to distinguish patients with primary RP from those with scleroderma or another rheumatic disease. Maricq and associates first described the abnormal pattern of nail-fold capillary vessels seen in scleroderma.39 The simplest method of detection is to coat the skin of the nail fold with immersion oil and then view the area using a bifocal dissecting microscope or a hand-held device such as an ophthalmoscope set at 20 or 40 diopters. A computerized nail-fold videocapillaroscopy technique is used in specialty centers. This technique provides enhanced digital images that can assess local blood flow and follow the disease course.40 Although patients with primary RP may have normal, thin, palisading nail-fold capillary loops (Figure 84-4), in secondary RP, capillary loop dilation/enlargements and dropouts represent the salient features. Patterns of capillary abnormalities appear to correlate with the course of systemic disease manifestations. Capillary dilation (giant capillaries), microhemorrhages, and some disorganization of the capillary network are typical of early disease; dropouts, avascular areas, and signs of neoangiogenesis with bizarre architectural distortion manifest at later stages of scleroderma.41,42 Approximately 20% to 30% of patients with RP and abnormal nail-fold capillary changes will develop clinical features of scleroderma, usually within a 2- to 3-year period.25 Patients presenting with RP, nail-fold capillary changes, and the presence of a scleroderma-related autoantibody have a 70% to 80% chance of developing scleroderma within 2 to 3 years from presentation. Therefore, capillaroscopy represents an important standard tool that can be used to examine patients presenting with RP.

In scleroderma, RP and digital ischemia are the clinical manifestations of both fixed structural vascular disease and abnormal regulation of local vasomotor control. Cold-induced vasoconstriction of peripheral blood vessels normally occurs via sympathetic stimulation. Abnormal thermoregulation is associated with a nonvasculitic vasculopathy characterized by endothelial dysfunction and a fibrotic proliferation, which produces an increase in collagen content of the intima of small and medium vessels and causes loss of vessel flexibility and obliteration of their lumina. Digital pitting with loss of fingertip tissue and small, painful, superficial ulcerations are very common and usually are noted secondary to disease affecting the small arteries and arterioles of the skin (Figure 84-5). Large, deep ulceration of the distal portion of the finger is a consequence of larger-vessel (e.g., digital artery) occlusion associated with severe vasospasm. The latter event usually presents as a sharp demarcation of the distal digit with intense, localized pain secondary to ischemia. Failure to reverse these events may lead to loss of the whole digit or limb secondary to deep tissue infarction.

Although a number of methods may be used to objectively score attacks of RP, no test is considered practical or reproducible enough to replace the clinical criteria for diagnosis and management. Patients with an active RP attack will present with coolness of the involved distal fingertips and/or toes, which can be associated with a line of demarcation of skin pallor or cyanosis. In later stages of severe RP, if left untreated, one may find signs of critical digital ischemia resulting in painful and unremitting digital ulcerations. Such findings warrant immediate medical treatment. Several methods may be used to assess the severity of RP. Traditionally, the patient is asked to use a diary to record the frequency and duration of attacks during days of usual activity. Raynaud’s condition score (RCS) is a patient-based measure that takes into account the impact of RP on the patient, including pain, discomfort, and effect on daily function. Other laboratory-based measures, including laser Doppler, thermography, and plethysmography methods, are used in an attempt to obtain objective data.

Treatment of Raynaud’s Phenomenon and Digital Ischemia

The primary goal in the management of RP is prevention of digital ischemia through the use of nonpharmacologic and pharmacologic measures (Figure 84-6). In the setting of acute digital ischemia, rapid intervention using both treatment modalities is required. The primary and most important nonpharmacologic therapy for prevention is avoidance of cold ambient temperatures, particularly transitioning from a warm or hot environment to a cold one. A warm ambient temperature reduces the frequency and severity of RP. All patients with RP should understand that clothing should be layered and loose-fitting, with the goal of maintaining a warm core body temperature, not just warmth of the affected extremities. Patients who have an acute ischemic event are best treated with rest in a warm environment (home or hospital), insulated from cold temperatures. Other potential therapies include minimizing emotional distress (reducing sympathetic tone) and avoiding aggravating factors such as smoking, sympathomimetic drugs (e.g., preparations for the common cold), migraine headache medications (e.g., serotonin agonists), and nonselective β-blockers (e.g., propranolol). Although behavioral therapies (biofeedback, autogenic training, and classical conditioning) are reported to be helpful, their benefit is controversial, and they play no role in the management of acute ischemia related to scleroderma.43 In fact, biofeedback alone is not helpful for primary RP.

Drug therapy for RP in scleroderma includes the use of a variety of oral or systemic vasodilators. Calcium channel blockers are considered to be first-line therapeutic agents in the treatment of RP. This class of medication works primarily by inducing arterial vasodilation through direct inhibition of contraction of vessel smooth muscle cells; however, these agents provide additional benefits by reducing oxidative and inhibiting platelet activation.44,45 Most published studies evaluating the efficacy of calcium channel blockers in RP have employed dihydropyridines (e.g., nifedipine, amlodipine, nisoldipine, isradipine, felodipine), and few have looked at nondihydropyridines, which appear to be more efficacious in primary than in secondary RP.46,47 A meta-analysis of 17 studies evaluating the efficacy of short- and long-acting formulations of calcium channel blockers reported a 33% reduction in attack severity and a near 50% reduction in the number of attacks per week.48 The current recommendation is to use an extended-release formulation of a calcium channel blocker for treatment of nonurgent RP. The dose should be titrated to clinical efficacy, but common side effects (hypotension, headache, pedal edema) should be monitored. For urgent cases of RP, a short-acting formulation of the medication is preferred, but titration should be carefully monitored. Although calcium channel blockers are the agents most likely to be effective, a host of other vasodilators are used, including nitrates, phosphodiesterase-5 inhibitors (e.g., sildenafil), intravenous prostaglandins, and sympatholytic agents (e.g., prazosin).

Among other agents being tested in the treatment of RP are drugs that enhance nitric oxide availability—serotonin uptake blockers (e.g., fluoxetine), Rho-kinase inhibitors, local injection of botulinum toxin (Botox), antioxidants (acetylcysteine), angiotensin receptor blockers, and selective α2-adrenergic receptor antagonists. A combination of these agents, if tolerated, can be used in refractory cases. Various formulations of topical and oral nitroglycerin are employed.49,50 Although some degree of efficacy is obtained, use of nitroglycerin is limited by substantial side effects, including headache, dizziness, and local skin irritation. The cyclic guanine monophosphate phosphodiesterase-5 inhibitors are thought to be effective in the treatment of RP by sustaining the vasodilatory effect of nitric oxide.51,52 Studies using phosphodiesterase inhibitors have reported variable effects, and the ideal drug and dosing have not yet been defined. Despite the theoretical advantage of angiotensin-converting enzyme (ACE) inhibitors, a study of quinapril showed that these medications did not affect the occurrence of digital ulcers or the frequency or severity of RP episodes.53 Thus, although ACEs are critical for the treatment of a scleroderma renal crisis (see renal section), they do not have a major role in the management of RP.

New emphasis is being placed on prevention of scleroderma vascular disease through the use of immunosuppressive and vasoprotective drugs. HMG-CoA (3-hydroxy-3-methylglutaryl-coenzyme A) reductase inhibitors (statins) can modify progression of vascular injury and can prevent vascular ischemia through different pleiotropic functions, as by improvement of endothelial dysfunction, reduction of clotting, and introduction of some anti-inflammatory effects.54,55 Data suggest that statins may increase the number of endothelial progenitor cells and may improve vascular remodeling after injury.56,57 In a randomized, placebo-controlled study of scleroderma patients with secondary RP, statin use significantly decreased the number of digital ulcers formed.58

Intravenous infusions of vasodilating prostaglandins (alprostadil, epoprostenol, iloprost, treprostinil) reduce the severity and frequency of Raynaud’s attacks and are most helpful during periods of sustained critical ischemia. These prostaglandins are efficacious in the treatment of RP because of their strong vasodilating effect, inhibition of platelet aggregation, and enhancement of vascular function through other mechanisms. Intravenous prostacyclins can be given intermittently to patients with scleroderma during winter months or throughout the year, or they can be used acutely during a vascular crisis. Oral prostaglandins have variable absorption and are not yet available.

Endothelin-1 is strongly implicated in the pathogenesis of RP and the vascular disease of scleroderma.59 Bosentan, an endothelin-1 receptor blocker, was studied in a randomized, placebo-controlled, double-blind multicenter study that showed efficacy in preventing new digital ulcers, along with improved hand function, but it did not improve healing of existing ulcers compared with placebo.60

Sympathectomy is a viable option for patients with RP who are unresponsive to medical therapy; it should be considered as acute intervention during a critical ischemic event. Localized digital sympathectomy with lysis of fibrosis around the vessel is effective for acute ischemia and has mostly replaced cervical sympathectomy.61,62 Responses to sympathectomies suggest that the long-term benefit derived from these interventions is limited. In fact, Raynaud’s attacks eventually recur, and medical therapy is needed to prevent new events.

Careful assessment for correctable macrovascular disease should be done when confronted with acute digital ischemia, particularly when the whole finger is demarcated, or when the event involves a lower extremity. In this setting, appropriate studies such as arterial Doppler ultrasound or angiographic imaging are warranted. If macrovascular disease is present, vascular surgery may help to alleviate the occlusive process. Preferential involvement of the ulnar artery has been reported in scleroderma patients.63

Patients with critical digital ischemia should be hospitalized to reduce vasospastic activity, maintain warmth, and permit rapid initiation of vasodilator therapy. Intravenous vasodilating prostaglandins can be infused to maximize vasodilation. Antiplatelet therapy with low-dose aspirin may be useful, but its benefit is unproven in the acute setting. Heparinization may be considered during acute ischemic crises of the digits, but chronic anticoagulation in scleroderma is not recommended. Chemical sympathectomy of the affected digit, performed by local infiltration with lidocaine or bupivacaine, may provide immediate relief. For refractory cases, a surgical approach to digital sympathectomy is used.

Ischemic digital lesions should be treated with topical antibiotics and daily cleansing with soap and water. Débridement procedures should be performed very cautiously because tissue trauma may extend the injury owing to the avascular nature of the digital tissue. Digital lesions that progress to dry gangrene should be permitted to undergo autoamputation. Surgical amputation is best offered only for intractable pain or deep tissue infection.

Skin Involvement

The most overt clinical manifestation of scleroderma is skin disease. The degree of involvement can vary among patients, and involvement can change in severity and distribution over time in the same individual. Almost every scleroderma patient presents with skin thickening and hardening due to increased collagen and extracellular matrix deposition in the dermis. The distribution of skin changes is characteristic, with more frequent and intense involvement of fingers, hands, forearms, distal legs, feet, and face, and, to a lesser degree, the proximal limbs and anterior trunk. Sparing of the midback is typical. Scleroderma is classically subdivided into limited and diffuse cutaneous forms (Figure 84-7). Limited scleroderma is defined by skin thickening restricted to the face and limbs distally to the elbows and knees. Commonly, in this form of the disease, only the fingers and the face are involved. In contrast, diffuse cutaneous involvement is characterized by widespread skin thickening, including proximal limbs and truncal areas. Proximal skin involvement defines the diffuse cutaneous subset but may be absent in early stages of disease. Signs of skin thickening are not apparent in the setting of other disease manifestations (e.g., Raynaud’s phenomenon, nail-fold capillary abnormalities), nor is evidence of internal organ involvement. This clinical presentation is referred to as systemic sclerosis sine scleroderma. Traditionally, patients are clustered into limited and diffuse skin subsets, but evidence suggests the existence of an intermediate group of patients. Each subset of disease defined by the degree of skin thickening has a unique pattern of disease manifestation and risk for specific clinical outcomes. Therefore, expression of skin disease can be used as a predictor or a “clinical biomarker” of the disease course. The variable degree of skin involvement can be quantified by physical examination. The most widely accepted scoring system (modified Rodnan skin score) is applied by pinching the skin in 17 different body areas and subjectively averaging the thickness of each specific site from 0 = normal to 3 = very thick (Figure 84-8). The skin score (maximum, 51) is a useful clinical measurement tool that can be used to quantify the severity of skin disease.28 Although the overall extent of skin involvement as measured by the skin score can predict certain clinical outcomes, it does not define the quality of the skin process nor the level of disease activity at any single point in time. Therefore, it is important to follow the skin score over time and to measure changes sequentially to monitor disease progression.

Cutaneous involvement in scleroderma begins with clinical signs of inflammation. This is called the edematous phase because it is characterized by nonpitting edema of affected body areas. In limited scleroderma, this event is mild and is restricted to the digits; however, in the diffuse form of the disease, cutaneous swelling and edema can be widespread and so impressive in the limbs as to mimic a fluid overload state such as congestive heart failure. Edema can also cause local tissue compression. For example, upon involvement of the wrist area, scleroderma patients are not infrequently diagnosed with carpal tunnel syndrome (especially at disease onset) to explain hand and wrist discomfort. In association with edema, signs and symptoms of inflammation are common. Erythema of the skin and intense pruritus and pain (see Figure 84-7C) are characteristic of advancing active diffuse skin disease. This pain has a neuropathic quality with a reported “pins and needles” sensation. The disease process leads to loss of skin appendages, as well as decreased hair growth and loss of sweat and exocrine glands; thus the skin surface becomes dry and uncomfortable. Small papules can be seen in areas of trauma as the result of scratching, giving the surface a cobblestone texture. The edematous phase continues for several weeks but eventually gives way to a fibrotic stage, with protracted activity that may last months or years.

During the fibrotic phase, acute inflammation is clinically less obvious, and deposition in the dermis of excessive collagen and other extracellular material thickens the skin, making it inflexible and causing further loss of skin appendages. Fibrosis extends beyond the dermis into the deeper layers with loss of subcutaneous adipose tissue (lipodystrophy). In late stages of the disease, skin actually thins with atrophy and has a noninflammatory bound-down appearance. Deeper tissue fibrosis causes permanent contractures around joints or may involve underlying muscle, causing a myopathy (see Figure 84-7D). Patients with diffuse cutaneous scleroderma develop the most dramatic widespread skin changes; those with limited skin disease may note only puffy fingers and digital thickening typical of sclerodactyly. A masked facies, small oral and orbital apertures, and vertical furrowing of the perioral skin are consequences of skin and soft tissue fibrosis. In some patients, gum atrophy and facial skin tightening make the front teeth appear more prominent. Flexion contractures of fingers, wrists, and elbows often appear secondary to dermal sclerosis and fibrosis with atrophy of underlying tissues. Skin ulcers can develop as a complication of avascular fibrotic or damaged thinned skin and are very common at sites of trauma, such as over the digital metacarpophalangeal or proximal interphalangeal joints or at the tip of the elbows, particularly when joint contractures are present at these sites (see Figure 84-5). Ulcerations may be noted secondary to underlying vascular disease and tissue ischemia (see vascular disease section). Ankle or lower extremity ulcers occur rarely secondary to macrovascular occlusive disease or comorbid conditions (venous disease). Hypopigmentation (vitiligo-like) and/or hyperpigmentation of the skin (“salt-and-pepper” appearance) can develop typically on the face, arms, and trunk (Figure 84-9A). General tanning of the skin may be seen, even in the absence of sun exposure.

Active skin involvement might persist for the first 12 to 18 months of the disease, with no further clinical signs of inflammation or progressive skin fibrosis seen after this interval. During this later stage, the skin begins to repair and can return to normal texture or, in areas most severely affected (e.g., fingers, hands), can become thin and atrophic. During this recovery phase, new robust hair growth is seen, particularly on the forearms, and itching and pain disappear, consistent with spontaneous resolution of disease activity. After years from disease onset, the skin rarely relapses again into an active phase and gradually can recover normal texture and color. Patients who established over time their phenotype as limited to the fingers (plus or minus facial changes) do not convert to the diffuse form of the disease.

Telangiectasias are erythematous matted skin lesions of vascular origin; for this reason, they blanch on local pressure. They are composed of vasodilated postcapillary venules without evidence of inflammation and resemble the type of lesions seen in Osler-Weber-Rendu disease (hereditary hemorrhagic telangiectasia). Telangiectasias develop primarily on the fingers, hands, face, and mucous membranes, but they may also be found on the limbs and trunk (Figure 84-9B). They tend to become more numerous over time in both limited and diffuse types of skin disease, and are more obvious in white patients with limited scleroderma. The biologic mechanism leading to the development of telangiectasias in scleroderma is thought to be related to the underlying chronic tissue hypoxia that stimulates abnormal secretion of vascular growth factors (e.g., vascular endothelial growth factor [VEGF]). Thus, the development of telangiectasias may indicate ongoing vascular injury and abnormal vascular repair or angiogenesis. The total burden of telangiectasias is associated with increased risk of pulmonary hypertension, suggesting that they are clinical biomarkers of systemic vascular disease. Nail-fold capillary abnormalities can be observed after immersion oil is applied to the skin surface at the base of the fingernails, and direct visualization is performed using an ophthalmoscope or by videocapillaroscopy. In early scleroderma, dilated capillary loops (giant capillaries) and microhemorrhages may be seen. At later stages, the nail-fold capillaries are attenuated and disorganized (see section on vascular changes).

Despite evidence of active inflammation during the early edematous phase of scleroderma skin involvement, systemic corticosteroids do not appear to be effective in stopping disease progression. A variety of other immunomodulatory drugs and newer agents with potential “disease-modifying properties” (i.e., antifibrotic) have been used to control the skin disease, but to date, none has proved uniformly successful (see treatment section). In the early active stage of diffuse scleroderma, pruritus can be one of the most distressing symptoms. Antihistamines, analgesics, or tricyclic antidepressants (e.g., doxepin) are often used but usually provide only partial benefit. Dryness of the skin surface results from damage to the exocrine structures caused by decreased or absent natural oil (sebum) production. This can worsen pruritus, leading to skin trauma due to repetitive scratching. Ulceration and secondary tissue infection may also result. The best approach to treatment is characterized by frequent topical application of an emollient preparation, periodic cleansing with soap and water, and use of topical antibiotics for any traumatic skin ulceration. Physical therapy is most important to prevent severe skin and joint contractures, and to help patients with activities of daily living.

Gastrointestinal Involvement

Almost every scleroderma patient has symptoms of gastrointestinal disease, ranging from mild gastroesophageal reflux disease (GERD) to severe bowel dysfunction, which can be life threatening. Virtually any segment of the gastrointestinal tract can be affected (Table 84-4).

Table 84-4 Gastrointestinal Manifestations of Scleroderma

Site Manifestation Management






Small and large intestine




Patients may report that chewing is difficult because of decreased facial flexibility caused by skin and deeper tissue fibrosis. Perioral skin tightening can result in a decreased oral aperture and inability to open the mouth wide enough for proper dental care or to bite into large solid foods such as an apple (Figure 84-10). Dry membranes resulting from decreased saliva production can lead to difficult mastication. In some patients, periodontal disease and gum regression cause loosening of teeth, which further affects chewing capacity.

Upper pharyngeal function is usually normal, but a subset of patients may develop a myopathy that causes weakness of pharyngeal muscles, mimicking a neuromuscular disease.64 This manifestation is characterized by pharyngeal dysfunction with problems initiating swallowing and frequent coughing due to laryngeal penetration and can be associated with increased risk for aspiration of foods or liquids. Although limited pathologic studies of the upper pharyngeal structures have examined scleroderma, both myositis and fibrosis are known to occur.


Dysphagia resulting from esophageal disease is the most common gastrointestinal symptom, occurring in approximately 90% of patients. Heartburn, regurgitation, and dysphagia for pills and solids (more than liquids) are the most common symptoms, but atypical retrosternal pain (particularly at night), periodic cough, a sense of food “sticking” in the esophagus, and nausea can result from esophageal dysfunction. Esophageal involvement in scleroderma is characterized by loss of normal smooth muscle function, especially in the lower two-thirds of the esophagus, and hypotonia of the lower esophageal sphincter (Figure 84-11A). Functional studies of esophageal motility suggest that neural dysfunction is common in patients with scleroderma, and that this may precede myopathic dysfunction and histologic changes in smooth muscle layers.65 Manometric evaluation has identified the presence of esophageal hypomotility in areas that later did not show any histologic abnormality of the smooth muscle. Other studies have shown esophageal smooth muscle activity in response to pharmacologic stimulation (methacholine challenge) early in the disease, but not in patients with more advanced scleroderma. The cause of these abnormalities is unknown, but clear evidence indicates that neural dysfunction precedes muscle disease.

Tissue fibrosis is often evoked as the cause of esophageal disease in scleroderma. However, pathologic studies demonstrate atrophy of the smooth muscle layers of the distal esophagus in the absence of significant fibrosis. Ischemia of the esophagus in scleroderma is suggested by functional studies that show impaired esophageal blood flow following cold exposure and rewarming protocols. Noninflammatory intimal layer hyperplasia in arterioles of the gastrointestinal tract has been described. Inflammatory infiltrates usually are not present in the smooth muscle unless severe transmural esophagitis is present. This suggests that inflammation is not a cardinal feature of the pathogenesis of esophageal smooth muscle lesions in scleroderma.

The severity of esophageal symptoms may not accurately reflect the seriousness of the underlying disease. Therefore, every patient with scleroderma should be fully evaluated for esophageal involvement. If untreated, gastrointestinal reflux may lead to esophagitis, bleeding, esophageal strictures, or precancerous lesions such as Barrett’s esophagus. The asymptomatic patient without abnormal laboratory testing (anemia, positive evidence of gastrointestinal bleeding) does not require specific interventions. Patients with mild GERD can be treated empirically with proton pump inhibitors and then can be followed clinically if they become symptom free with normal laboratory data. When more severe symptoms are present (e.g., GERD with dysphagia), or when treatment with proton pump inhibitors fails, an upper gastrointestinal endoscopy is necessary to fully assess the anatomy of the esophagus and the extent of related disease. Direct endoscopy is appropriate to rule out Barrett’s esophagus, stricture, or a site of uncontrolled esophagitis or bleeding. Barium swallow and cine-esophagography are sensitive tests for esophageal strictures. However, direct measurement of esophageal motility via esophageal manometry may be needed if the cause of symptoms such as atypical chest pain is unclear. Although complications from Barrett’s esophagus (e.g., esophageal cancer) are uncommon, periodic re-endoscopy to reassess status is recommended, even if symptoms are under therapeutic control.

It is critical that patients alter eating behavior to complement medication therapy. Patients often do better by eating more frequent smaller meals, avoiding food intake for several hours before bedtime, moving the main meal toward mid-day, taking a walk after meals to help gastroesophageal emptying, and eliminating foods that aggravate symptoms (e.g., spicy sauces, caffeinated or carbonated beverages). Bedtime is often the time of severe reflux. This can be improved by avoiding filling the stomach at least 2 to 3 hours before bedtime and by elevating the head and trunk during sleep. Treatment of reflux by suppression of gastric acid with antacids or histamine-2 (H2)- is helpful but overall disappointing in scleroderma. Conversely, proton pump inhibitors (e.g., omeprazole, lansoprazole) can be very effective. Usually, long-term daily use is required. Higher doses may be used periodically for breakthrough symptoms. If symptoms are not controlled with recommended medication dosages, then a formal 24-hour ambulatory pH study may be necessary to determine whether persistent symptoms are due to uncontrolled acid reflux. A prokinetic drug (e.g., metoclopramide, domperidone, erythromycin) should be added when symptoms of dysphagia or endoscopic findings of esophagitis are present despite the use of effective acid suppression. These medications tend to be effective during early disease but are less likely to help later on, with advanced esophageal dysfunction. Long-term use of prokinetics may be needed. Many patients respond to relatively low doses, for example, at bedtime alone. Given the increased risk of neurologic complications with the use of metoclopramide, many recommend domperidone as a relatively safer alternative when long-term prokinetic treatment is needed.


The degree of delayed gastric emptying (gastroparesis) is underappreciated in scleroderma, and gastroparesis often causes early satiety, aggravation of GERD symptoms, anorexia, abdominal pain, a sensation of bloating, or nausea. Frequently, weight loss in scleroderma patients is a consequence of poor caloric intake due to lack of appetite from poor gastric function. Prokinetics are used to improve gastric emptying and related symptoms. Gastritis or gastric ulcer can occur. Dilation of the microvasculature in the gastric mucosa is found in a subset of patients. This manifestation, also called gastric antral vascular ectasia (GAVE), is thought to be caused by an abnormal angiogenesis, leading to bizarre dilation of microvessels and arterial-venous (A-V) malformations similar to the telangiectasias seen in the skin, lips, and oral mucous membranes. In general, these lesions are asymptomatic, but they can be responsible for occult gastrointestinal bleeding. Extensive clusters of A-V malformations lead to the presence of longitudinal red stripes in the inner lining of the stomach, converging to the pylorus and described on endoscopy as “watermelon stomach,” based on their appearance (Figure 84-11B). Endoscopic therapy with laser photocoagulation or cryotherapy can be used to ablate bleeding vessels. Rarely, bleeding associated with GAVE is refractory and requires multiple transfusions, repeated laser or cryotherapy, or even gastric surgery (e.g., antral or gastric resection).

< div class='tao-gold-member'>

Only gold members can continue reading. Log In or Register to continue

Stay updated, free articles. Join our Telegram channel

Jul 3, 2016 | Posted by in RHEUMATOLOGY | Comments Off on Clinical Features and Treatment of Scleroderma

Full access? Get Clinical Tree

Get Clinical Tree app for offline access