Cryoglobulinemia

Abstract

Introduction Cryoglobulins are immunoglobulins (Ig) that precipitate when serum is incubated at temperatures lower than 37 °C. The existence of circulating cryoglobulins (cryoglobulinemia) is not always related to the presence of symptomatology, and the term “cryoglobulinemic syndrome” is used when patients present clinical manifestations (44). In 1933, Wintrobe and Buell (55) described cryoprecipitation as a laboratory phenomenon. In 1966, Meltzer and Franklin (27) described the typical clinical symptoms associated with cryoglobulinemia, particularly the triad of purpura, arthralgia, and weakness. In 1974, Brouet et al (7) classified cryoglobulins into 3 types: Type I cryoglobulins were those composed of single monoclonal immunoglobulins, and types II and III were those formed by monoclonal (type II) or polyclonal (type III) IgM with rheumatoid factor activity plus the corresponding antigen (usually polyclonal IgG). For this reason, types II and III are classically referred to as “mixed cryoglobulinemia.” Recently, new types of cryoglobulins have been described (41). Cryoglobulins have been observed in a wide variety of diseases, including malignancies, infections, and systemic autoimmune diseases (17,40). When there is no demonstrable underlying disease, the condition is called “essential cryoglobulinemia.” Since the initial report (36) in 1990 of the association between cryoglobulinemia and hepatitis C virus (HCV) infection, it has become clear that most cases of so-called essential cryoglobulinemia are in fact associated with HCV infection. This has led to important changes in the etiology, classification, and treatment of cryoglobulinemia in the last 10 years. Several studies have analyzed cryoglobulinemia in patients with HCV infection, but studies performed in groups of patients covering all etiologies of cryoglobulinemia are scarce (7,14,17,19,30). In this study, we analyzed the etiology, clinical manifestations, and immunologic features of a large series of consecutive patients with cryoglobulinemia from a single center (Hospital Clinic, Barcelona, Spain). Patients and Methods Patient selection Sera from 7,043 patients were tested for circulating cryoglobulins at the immunology department of our hospital between 1991 and 1999. A cryocrit of 1% or more was detected in 443 (6.29%) patients. Clinical and serologic characteristics of these patients were retrospectively collected in a protocol form. Salient features included in the protocol were the following: 1) age at onset of the disease, defined as the initial manifestation clearly attributable to cryoglobulinemia; 2) age at diagnosis, defined as the age when circulating cryoglobulins in serum were detected; 3) age at study, defined as the age when the patient entered the protocol study; 4) associated diseases such as hematologic, infectious, and autoimmune processes; 5) clinical manifestations at the diagnosis of cryoglobulinemia; 6) cumulative clinical manifestations during disease evolution (from the onset until the protocol study); 7) laboratory findings at time of entry in study. Study of etiologic factors Hepatitis B virus (HBV) surface antigen (HBsAg) was analyzed in all patients by enzyme-linked immunoabsorbent assay (ELISA) and antibodies to HCV by a third generation ELISA (Ortho 3.0 Diagnostic Systems, Neckargemund, Germany). Anti-HIV-1-p24 antibodies were analyzed in 153 patients by ELISA coated with HIV-1-p24 (ETI p24 K Sorin antigen recombinant kit, Saluggia, Italy). For confirmation, Western blots (Epitope, Beaverton, Oregon, USA) with HIV-1gp160, gp120, gp41, p65, p51, p24, and p18 bands were applied. Systemic autoimmune diseases were diagnosed based on the following criteria: 1) primary Sjögren syndrome (SS) according to the preliminary diagnostic criteria for SS proposed by the European Community Study Group (51); 2) systemic lupus erythematosus (SLE) according to the 1982 revised criteria of the American Rheumatism Association (46); 3) rheumatoid arthritis (RA) by American Rheumatism Association criteria (5); 4) systemic sclerosis by American Rheumatism Association preliminary criteria (3); 5) polymyositis-dermatomyositis by Bohan and Peter criteria (6); 6) primary antiphospholipid syndrome by the preliminary classification criteria (53); 7) systemic vasculitis by the Chapel Hill criteria (20). The presence of hematologic malignancies was confirmed by the clinical findings and biopsies of lymph nodes and/or bone marrow. Patients with non-Hodgkin lymphoma were further classified by histology (Working Formulation). Finally, essential cryoglobulinemia was considered in those cases in which no infectious, autoimmune, or hematologic disease was found. Definition of clinical features The clinical manifestations evaluated in our protocol were defined as follows: Articular involvement included arthralgia and/or nonerosive arthritis characterized by tenderness, swelling, or effusion involving 2 or more peripheral joints. Arthralgia and/or arthritis was considered as a manifestation of cryoglobulinemia, when cases with coincidental systemic autoimmune diseases (SLE, SS, or RA) were excluded. Fever: axillary temperature >37.5 °C. Lymphadenopathy: enlarged nodes (>0.5 cm) in the cervical region, axilla, or inguinal area, in the absence of hematologic malignancies. Purpura: recurrent palpable lesions that usually involved the lower extremities. Skin biopsies of purpuric lesions showed leukocytoclastic vasculitis. Leg ulcers: usually painless, invariably associated with purpura. Acral cyanosis, ischemia, and/or skin necrosis: affecting the tip of the nose or the ears, fingers, toes, or legs. Raynaud phenomenon: blanching of the fingers, toes, ears, nose, tongue, induced by exposure to cold, stress, or both. Livedo reticularis: reddish or cyanotic discoloration of the skin with a reticular pattern. Peripheral neurologic manifestations included paresthesia, numbness, and/or motor defects of the lower extremities, compatible with multiple mononeuritis or polyneuritis. Neuropathy was classified as sensory, motor, or sensory-motor neuropathy according to the symptoms and was confirmed by electromyography. Sural nerve biopsy, when performed, showed vasculitis involving vasa nervorum. Central neurologic involvement included cerebral ischemia (in the absence of hypercoagulability or vascular risk factors), spinal cord, or cranial nerve involvement. Nephropathy: a) persistent proteinuria >0.5 g/day, or b) altered urinalysis (hematuria, pyuria, red blood cell casts), or c) raised serum creatinine >1.5 mg/dL. Glomerular injury diagnosed by renal biopsy included: membranoproliferative glomerulonephritis, mesangial proliferative glomerulonephritis, and segmentary and focal glomerulonephritis. Patients with lupus nephropathy were excluded. Gastrointestinal involvement: intestinal vasculitis was considered in those patients with severe abdominal pain and intestinal bleeding, and confirmed by biopsy or postmortem. Lung involvement: a) pulmonary fibrosis (chronic diffuse interstitial infiltrates on X-ray with a restrictive pattern on pulmonary function studies), or b) acute pulmonary infiltrates, fever and cough in the absence of pulmonary edema, adult respiratory distress syndrome, infectious pneumonia, lung malignancy, and granulomatous disease, or c) hemoptysis. Laboratory studies Serum cryoglobulins were measured after centrifugation. Blood samples were obtained and kept at 37 °C for 30 min before separation. Serum was prepared by centrifuging at 37 °C for 10 min at 2,500 rpm. The serum obtained (10 mL) was transferred to a 15-mL glass graduated conical tube, and incubated at 4 °C for 7 days. If a precipitate or gel was detected, the tube was centrifuged at 2,500 rpm for 30 min at 4 °C. The cryoprecipitate was visually measured according to the graduated level of the glass tube and referred as a cryocrit (%) to the total volume sample. The sera supernatant was discarded, and the cryoprecipitate was washed twice with cold saline at 2,000 rpm for 5 min at 4 °C. The cryoprecipitate was resuspended with saline in a 37 °C water bath. If the cryoprecipitate became soluble, immunofixation with specific antisera to IgA, IgG, IgM, κ, and λ chains and anti total serum proteins was performed. The immunofixation was performed with a Helena “Immunofixation Agarose Kit” following the manufacturer’s procedure. After immunochemical analysis, cryoglobulins were classified according to Brouet et al (7). Other immunologic tests included antinuclear antibodies (ANA) (indirect immunofluorescence using mouse liver and Hep-2 cells as substrate) and precipitating antibodies to extractable nuclear antigens, including Ro/SS-A, La/SS-B, U1-snRNP, and Sm (counterimmunoelectrophoresis using calf and rabbit thymus and human spleen extracts). Antimitochondrial antibodies, antismooth muscle antibodies, type 1 anti-liver/kidney microsome antibodies, antiparietal cell antibodies, and antithyroidal antibodies (antithyroglobulin and antimicrosomal antibodies) were investigated by indirect immunofluorescence. Rheumatoid factor was detected by latex fixation and Waaler-Rose tests. Complement factors (C3 and C4) were estimated by nephelometry (Behring BNA nephelometer) and CH50 by Lachmann’s hemolytic technique (21). Antineutrophil cytoplasm antibodies were tested by an indirect immunofluorescence assay using ethanol-fixed normal human neutrophils. To establish the antigenic specificities of antineutrophil cytoplasm antibodies, serum samples were further tested with 2 ELISAs using as substrates myeloperoxidase (Bio-Carb, Lund, Sweden) and proteinase 3 (IBL, Hamburg, Germany). Statistical analysis We used conventional chi-square and the Fisher exact test to analyze qualitative differences. For comparison of quantitative parameters, the Student t-test was used in large samples of similar variance, and the nonparametric Mann-Whitney U test was used for small samples. A value of p <0.05 indicated statistical significance. When several independent variables appeared to have statistical significance in the univariate analysis, a logistic regression test was performed for multivariate analysis to rule out possible confounding variables. The odds ratio was calculated for assessing the risk of appearance of each variable, with a confidence interval of 95%. This statistical analysis was performed using the SPSS program. Results General characteristics Of the 443 patients with cryoglobulinemia, 258 (58%) were women and 185 (42%) were men (female:male ratio, 3:2). The mean age at the time of cryoglobulinemia detection was 54 years (range, 14–91 yr), and, at the time of the study, was 56 years (range, 14–91 yr). The mean disease duration from onset until time of the study was 2 years (range, 0–20 yr). Cryocrit levels were <5% in 374 (84%) patients, 5%–10% in 44 (10%), and 10%–20% in 16 (4%) patients; only 9 (2%) patients had cryocrit levels >20%. Cryoglobulins were further characterized when sufficient cryoprecipitate was available, about 4%–5% of cryocrit (Table 1). Thus, 90 cryoprecipitates were analyzed by immunofixation. Thirty-one were classified as type I cryoglobulins, 50 as type II, and only 2 cases as type III. Type I cryoglobulins consisted of monoclonal IgG κ (15 patients), IgM κ (15 patients) and IgG λ (1 patient). In most patients (47 patients), type II cryoglobulins were composed of monoclonal IgM κ + polyclonal IgG. Finally, 7 cases were unclassifiable according to the classification of Brouet et al (see Table 1).TABLE 1: Cryoglobulinemia in 90 patientsWe compared the main characteristics of patients with type I cryoglobulinemia and mixed cryoglobulinemia (Table 2). Patients with mixed cryoglobulinemia showed a higher prevalence of cutaneous involvement (53% versus 30%, p peripheral neuropathy versus p and rheumatoid factor versus p of patients with type I cryoglobulinemia and patients with mixed disease was associated with cryoglobulinemia in patients, autoimmune diseases in and hematologic disease in patients (Table patients had no disease associated with cryoglobulinemia and were classified as essential factors in 443 patients with was the main etiologic factor in our patients, present in patients. HCV antibodies were in patients, in and human virus antibodies in of 153 patients We compared the main characteristics of patients according to the presence or absence of HCV (Table with those HCV infection, patients with cryoglobulinemia related to HCV showed a lower prevalence of associated autoimmune versus p and hematologic versus p In HCV patients showed a lower prevalence of involvement versus p cutaneous involvement versus p versus p fever versus p Raynaud versus p versus p versus p versus p and versus p In HCV patients showed a higher prevalence of CH50 versus p and rheumatoid factor versus p of patients with and HCV patients had a systemic autoimmune disease associated with patients had primary SS, 30 7 (2%) (2%) systemic and 3 primary antiphospholipid syndrome (see Table The main hematologic diseases included 16 (4%) patients with non-Hodgkin 3 with 3 with multiple 2 with 2 with and 2 with Of the 16 patients with non-Hodgkin were classified as 4 as and 4 as Finally, we analyzed the of the etiologies 1). we observed a between HCV and autoimmune of SS, of or hematologic of non-Hodgkin We that most patients with or had HCV (see Table 1: of the etiologies in 443 patients with manifestations Clinical manifestations attributable to cryoglobulinemia were present in patients during the evolution of the In the diagnosis of cryoglobulinemia was by the detection of circulating The main clinical manifestations at the onset and during the evolution of the disease are in Table the most manifestations were cutaneous involvement in patients, involvement in renal involvement in and peripheral neuropathy in patients. Other clinical manifestations at onset included in patients, fever in in 16 involvement in and pulmonary involvement in Clinical manifestations at onset and during disease evolution in of 443 patients evolution of the disease, the most cumulative manifestations were renal involvement in patients, cutaneous involvement in involvement in and peripheral neuropathy in patients. Other clinical manifestations included in patients, in fever in involvement in pulmonary involvement in 9 and intestinal in 5 was the most cutaneous (Table present in patients, by Raynaud in skin in patients, in 16 ischemia or changes in 9 in 5 and in 3 patients. involvement is in Table 7 those cases with lupus >1.5 was present in patients, proteinuria in and in patients. patients had membranoproliferative glomerulonephritis, mesangial glomerulonephritis, and and focal glomerulonephritis. Finally, neurologic features are in Table Peripheral neuropathy was diagnosed in patients, and included mixed in patients, multiple mononeuritis in 10 and in 5 patients. patients showed involvement: cerebral ischemia in 7 spinal involvement in 3 and cranial nerve in 2 manifestations during disease involvement during disease evolution manifestations during disease analyzed the between cryocrit levels and features (Table Patients with a cryocrit higher than showed cryoglobulinemia more than those with a cryocrit lower than versus p those patients with a cryocrit showed a higher of cutaneous versus p peripheral neuropathy versus p and versus p between cryocrit level and main features The most immunologic features were CH50 in of patients, rheumatoid factor in of in of in of in of antismooth muscle antibodies in of and antiparietal cell antibodies in of (Table We compared the immunologic of patients according to the presence or absence of symptoms (Table Patients with cryoglobulinemia showed a higher prevalence of versus p rheumatoid factor versus p versus 30%, p and antismooth muscle antibodies versus p compared with patients clinical features in patients with immunologic features in patients according to the presence or absence of clinical manifestations attributable to the main between immunologic and clinical features of Patients with cutaneous involvement showed a higher of versus p rheumatoid factor versus p and antibodies versus p with involvement showed a higher of rheumatoid factor versus p versus p antismooth muscle antibodies versus p and antibodies versus p Patients with peripheral neuropathy more had rheumatoid factor versus p and those with had a higher prevalence of CH50 versus p and versus p of immunologic features according to the presence or absence of main In this study, we analyze the main characteristics of the series to of patients from a single center with This series all patients with circulating cryoglobulins tested in a hospital between 1991 and and as such from In this selection to a was and the study the of In the of etiologic factors and clinical or immunologic features was variable, on the of the For in an study by et al in the main etiologic factors were only was only investigated in of the of in the and out of the present study not to the levels of in the study of and performed a and analyzed cryoglobulins only in patients. In laboratory were not in all 443 patients, to the of performed the presence of clinical manifestations, and, the of the Since the of HCV in several have HCV in of cases of cryoglobulinemia In this study, HCV is the main etiologic factor present in of patients with In we a lower prevalence of patients. the of the that a in cryoglobulinemia was by the existence of and/or in a of patients with cryoglobulinemia This fact a HCV infection, and studies have confirmed that than of patients have In we in of 153 patients tested for most showed an associated HCV infection. et al a prevalence of cryoglobulins in patients with a higher prevalence in those with HCV versus in has been in infection, the of cryoglobulins is associated with serum of and a antigenic of B In patients with and HCV infection, we a higher prevalence of hematologic and autoimmune diseases in patients, a of the hematologic and autoimmune in those patients HCV infection. we have described a clearly immunologic pattern in patients with and HCV infection. Patients with cryoglobulinemia related to HCV showed a higher of and rheumatoid that the of the by antigenic exposure to HCV a based on to the rheumatoid factor and the In patients HCV showed a higher prevalence of clinical autoimmune features and the presence of a systemic autoimmune disease in these patients, in the immunologic more than in HCV patients. Of the systemic autoimmune diseases, the etiologic factor in our study, primary SS and were the most diseases associated with et al patients with an associated systemic autoimmune disease The prevalence of cryoglobulinemia in primary SS has from to in but 2 studies that analyzed large series have a prevalence of We a higher prevalence of HCV in SS and cryoglobulinemia that the cryoglobulinemia observed in cases of SS to HCV infection. The autoimmune disease associated with cryoglobulinemia was The clinical significance of cryoglobulinemia in has been and has been related to the activity and of the disease and serum levels The presence of cryoglobulins in autoimmune diseases than or primary SS has been In our we patients with cryoglobulinemia and patients), systemic sclerosis patients) and primary antiphospholipid syndrome The third etiologic in our study patients with hematologic which were detected in patients. The most hematologic disease was non-Hodgkin present in 16 patients. has been that patients with cryoglobulinemia usually after a a possible of HCV in the of hematologic malignancies has been in non-Hodgkin lymphoma HCV has been in of patients. In a cryoglobulinemia, and non-Hodgkin lymphoma has been is that in our series we patients with these 3 We the the etiologies of cryoglobulinemia (see 1). We a etiologic factors such as HCV infection, SS, and non-Hodgkin et al that mixed cryoglobulinemia and monoclonal rheumatoid factor used as laboratory factors for lymphoma in is possible that the in the patient of cryoglobulinemia, HCV infection, and SS the of Thus, the association of SS with cryoglobulinemia and HCV as as the possible evolution of 1 of these 3 into a non-Hodgkin lymphoma a HCV infection, cryoglobulinemia, SS, and of cryoglobulins into 3 based on immunochemical was proposed by Brouet et al (7) in 1974, and is used it a with associated diseases and clinical The most type in our 90 characterized patients was type II mixed cryoglobulinemia et al detected type II mixed cryoglobulinemia in of we report 7 cases of cryoglobulins that not into the classification of Brouet et al a fact in have described the presence of IgM with of polyclonal immunoglobulins, and have proposed the existence of a new the type IgM and polyclonal In study, et al that cases of mixed cryoglobulinemia an monoclonal IgG on The existence of a has been described in a patient with SS In our study, we report 7 cases with cryoglobulins IgM + monoclonal or associated with polyclonal IgG. This a from to monoclonal that cryoglobulinemia is a the existence of these cryoglobulins to an In the present we a prevalence of symptoms in of patients. studies have a prevalence of symptoms in of patients with cryoglobulinemia associated with HCV lower The most clinical features in our study were and manifestations, in with those observed by et al We a of renal involvement in our patients similar to that by et al only a of patients with cryoglobulinemia is an We the of several factors in the of symptomatology, such as etiologic and immunologic The appearance of symptoms was related to a higher of cryocrit association with a systemic autoimmune disease, and presence of immunologic rheumatoid and we a association clinical and immunologic cutaneous involvement was with and rheumatoid involvement with rheumatoid factor and neuropathy with rheumatoid and with studies have analyzed the association between symptoms and etiologic or immunologic et al an association between peripheral neuropathy and the existence of higher rheumatoid factor and and study the existence of cryocrit levels and levels with a of glomerulonephritis. Finally, in the study by et al patients with cryoglobulinemia associated with autoimmune diseases showed Raynaud and syndrome more Patients with mixed cryoglobulinemia showed a higher of cutaneous involvement and peripheral neuropathy (see Table 2). This prevalence of symptoms in patients with mixed cryoglobulinemia a of cryoglobulinemia to systemic vasculitis of an of the induced by the of IgM rheumatoid factor with the in a than that observed in type I cryoglobulinemia to a of the (44). The of mixed cryoglobulins is related to the existence of an IgM with rheumatoid factor and we in a higher prevalence of rheumatoid factor activity in our patients with mixed cryoglobulinemia compared with those with type I In we analyzed the main and immunologic features in 443 patients with cryoglobulinemia from a single HCV was the main etiologic factor by primary SS and non-Hodgkin lymphoma after a analysis, it was not possible to in of patients to this as essential A was detected such as primary SS, and non-Hodgkin we in of patients; purpura, renal and neuropathy were the most The appearance of was associated with higher cryocrit the existence of an underlying autoimmune disease, and the presence of immunologic such as rheumatoid and We analyzed the clinical manifestations and immunologic features of 443 consecutive patients women and 185 mean 56 with cryoglobulinemia from a single diseases were detected in patients, autoimmune diseases in hematologic diseases in and essential cryoglobulinemia in Hepatitis C virus (HCV) was in patients, Hepatitis B virus surface antigen in and human virus antibodies in patients had primary Sjögren syndrome, 30 systemic lupus 7 (2%) and (2%) systemic Finally, 16 (4%) patients had non-Hodgkin 3 3 multiple and 2 We observed a between HCV and autoimmune in Sjögren syndrome, in or hematologic in non-Hodgkin Clinical manifestations attributable to cryoglobulinemia were present in patients. The most manifestations were cutaneous involvement in patients, renal in involvement in and peripheral neuropathy in patients. Other clinical manifestations included in patients, fever in and in Patients with showed cryoglobulinemia more than those with <5% versus p Patients with showed a higher of cutaneous features versus p peripheral neuropathy versus p and renal involvement versus 30%, p Finally, patients with cryoglobulinemia showed a higher prevalence of versus p rheumatoid factor versus p and antinuclear antibodies versus 30%, p