Prognostic factors in patients with hepatitis C virus infection and systemic vasculitis.код для вставкиСкачать
ARTHRITIS & RHEUMATISM Vol. 63, No. 6, June 2011, pp 1748–1757 DOI 10.1002/art.30319 © 2011, American College of Rheumatology Prognostic Factors in Patients With Hepatitis C Virus Infection and Systemic Vasculitis Benjamin Terrier,1 Oren Semoun,2 David Saadoun,1 Damien Sène,1 Matthieu Resche-Rigon,2 and Patrice Cacoub1 associated with outcome. In multivariate analysis, severe fibrosis (HR 10.8) and the FFS (HR 2.49) were significantly associated with a poor prognosis. Treatment with the combination of PEGylated interferon plus ribavirin was associated with a good prognosis (HR 0.34), whereas treatment with immunosuppressive agents was associated with a poor outcome, after adjustment for the severity of vasculitis (HR 4.05). Among patients without severe fibrosis, the FFS was a good predictor of outcome, while among those with severe fibrosis, the severity of vasculitis had no prognostic value. Conclusion. At the time of the diagnosis of HCVrelated systemic vasculitis, severe liver fibrosis and the severity of vasculitis were the main prognostic factors. Use of antiviral agents was associated with a good prognosis, whereas treatment with immunosuppressant agents had a negative impact. Objective. Hepatitis C virus (HCV)–related systemic vasculitis can cause significant morbidity and mortality. Most studies of the prognosis of patients with HCV-related systemic vasculitis are based on heterogeneous studies performed before the era of antiviral therapy. The aim of this study was to analyze the clinical, biologic, and therapeutic factors associated with prognosis in a homogeneous series of patients with HCV-related systemic vasculitis who were followed up during the era of antiviral therapy. Methods. One hundred fifty-one consecutive HCV RNA–positive patients with vasculitis were prospectively followed up between 1993 and 2009 and were analyzed for clinical, biologic, and therapeutic factors associated with survival. Results. After a median followup period of 54 months, 32 patients (21%) had died, mainly of infection and end-stage liver disease. The 1-year, 3-year, 5-year, and 10-year survival rates were 96%, 86%, 75%, and 63%, respectively. Baseline factors associated with a poor prognosis were the presence of severe liver fibrosis (hazard ratio [HR] 5.31), central nervous system involvement (HR 2.74), kidney involvement (HR 1.91), and heart involvement (HR 4.2). The Five-Factors Score (FFS), a vasculitis scoring system, was significantly Mixed cryoglobulinemia (MC) is a systemic vasculitis (1–3) caused by monoclonal B cells producing pathogenic IgM with rheumatoid factor (RF) activity (4). The disease expression is variable, ranging from mild clinical symptoms (purpura, arthralgia) to lifethreatening complications (glomerulonephritis, widespread vasculitis) (1–3). Shortly after the discovery of hepatitis C virus (HCV) in 1989, evidence emerged that ⬎80% of cases of MC were associated with HCV infection (5–7). In addition, among patients with HCV-associated MC (HCV-MC), 5–10% will experience the development of symptomatic MC vasculitis (8,9). Less frequently, systemic vasculitis during HCV infection may occur in the absence of detectable cryoglobulin, although the presentation and therapeutic management of such vasculitis should be similar to that of HCV-MC vasculitis (10). Cryoglobulinemic vasculitis can cause significant morbidity and mortality. According to the literature, the worst prognostic factors are age ⬎60 years and renal 1 Benjamin Terrier, MD, David Saadoun, MD, PhD, Damien Sène, MD, PhD, Patrice Cacoub, MD, PhD: CNRS, UMR 7211, Groupe Hospitalier Pitié-Salpétrière, Assistance Publique Hôpitaux de Paris, and Université Pierre et Marie Curie Paris 6, Paris, France; 2 Oren Semoun, MD, Matthieu Resche-Rigon, MD, PhD: INSERM, U717, Hôpital Saint-Louis, Assistance Publique Hôpitaux de Paris, and Université Diderot Paris 7, Paris, France. Dr. Cacoub has received consulting fees, speaking fees, and/or honoraria from Bristol-Myers Squibb, Pfizer, and SanofiAventis (less than $10,000 each) and from Roche, Servier, Vifer Pharma, and Schering-Plough (more than $10,000 each). Address correspondence to Patrice Cacoub, MD, PhD, Department of Internal Medicine, Groupe Hospitalier Pitié-Salpétrière, 47 Boulevard de l’Hôpital, 75013 Paris, France. E-mail: email@example.com. Submitted for publication July 12, 2010; accepted in revised form February 17, 2011. 1748 PROGNOSTIC FACTORS IN HCV-RELATED VASCULITIS involvement (11,12), with renal involvement being the main cause of death (1,11,13,14), followed by liver involvement, cardiovascular disease, infection, and lymphoma (11). Life-threatening complications of MC vasculitis were observed less frequently (in up to 14% of patients) and were associated with 66% mortality (15). However, most of these findings were derived from heterogeneous studies performed prior to the era of antiviral therapy, which included mostly HCVpositive and HCV-negative patients with MC, who may differ in terms of therapeutic management (i.e., use of antiviral therapy) and outcome, in particular regarding liver complications. Recently, we reported that the most common causes of death in HCV-MC vasculitis were infection, end-stage liver disease, cardiovascular disease, and (in only a small proportion of patients) vasculitisrelated organ involvement (16), reflecting recent changes in the therapeutic management of such patients. In the present study, we analyzed the clinical, biologic, and therapeutic factors associated with prognosis in a homogeneous series of 151 consecutive patients with HCV-MC vasculitis who were followed up in a tertiary care reference center during the era of antiviral therapy. PATIENTS AND METHODS Patients. The study group comprised 151 consecutive patients with HCV-related systemic vasculitis (69 men and 82 women; mean ⫾ SD age 60 ⫾ 13 years [range 31–85 years]) who were participants in a prospective and observational cohort study and were referred between January 1, 1993 and July 1, 2009 to the Department of Internal Medicine, Groupe Hospitalier Pitié-Salpétrière. All patients were followed up by the same physician (PC). Inclusion criteria for the study were as follows: 1) chronic active HCV infection, 2) HCV-related vasculitis, and 3) a minimum of 6 months of followup after the diagnosis. Exclusion criteria were the presence of hepatitis B virus surface antigen or anti–human immunodeficiency virus antibodies. Patients were considered to have HCV-related vasculitis if they had, in association with clinical manifestations of vasculitis (purpura or cutaneous ulcers, arthralgia, myalgia, peripheral neuropathy, renal involvement, cerebral vasculitis, gastrointestinal involvement, cardiac involvement), the following: histologically proven vasculitis (n ⫽ 96) and/or detectable MC with at least purpura as a clinical manifestation (n ⫽ 55). Patients without histologically proven vasculitis but with purpura and detectable cryoglobulin were considered to have small-vessel vasculitis, based on previously defined clinical and biologic criteria (1,17). For each patient, the type of vasculitis was classified according to the Chapel Hill criteria (17). According to these criteria and those of the American College of Rheumatology (18), patients were classified as having MC-type small-vessel vasculitis (n ⫽ 130) or polyarteritis nodosa (PAN)–like medium-vessel vasculitis (n ⫽ 21). Both MC vasculitis and 1749 Table 1. Baseline characteristics of the 151 patients with HCVrelated systemic vasculitis* Demographic features No. of men/no. of women Age, mean ⫾ SD years HCV infection related Genotype 1 2 3 4 5 Not available HCV RNA level, mean ⫾ SD log10 IU ALT, mean ⫾ SD ⫻ upper limit of normal Liver Metavir score Activity score, mean ⫾ SD Activity score ⱖ3 Fibrosis score, mean ⫾ SD Fibrosis score ⱖ3 Mixed cryoglobulinemia related Cryoglobulin positive Cryoglobulin level, mean ⫾ SD gm/liter Type II Type III C4, mean ⫾ SD gm/liter Rheumatoid factor positive Vasculitis-related features Deterioration of general health status Peripheral neuropathy Purpura Arthralgia/arthritis Kidney involvement Myalgia Central nervous system involvement Gastrointestinal tract involvement Heart involvement B cell non-Hodgkin’s lymphoma 69/82 60 ⫾ 13 84/136 (62) 22/136 (16) 13/136 (10) 11/136 (8) 6/136 (4) 15 5.9 ⫾ 0.9 1.9 ⫾ 1.2 1.3 ⫾ 0.8 10/136 (7) 2.0 ⫾ 1.2 48/146 (33) 135/151 (89) 1.32 ⫾ 1.62 110 (82) 25 (18) 0.09 ⫾ 0.09 97/129 (75) 35/151 (23) 111/151 (74) 104/151 (69) 76/151 (50) 50/151 (33) 23/151 (15) 14/151 (9) 12/151 (8) 6/151 (4) 22/151 (14) * Except where indicated otherwise, values are the number of patients/ number of patients assessed (%). HCV ⫽ hepatitis C virus; ALT ⫽ alanine aminotransferase. PAN-type vasculitis were included in the same analysis of the whole cohort to assess the prognostic factors for survival, because of the low number of patients and events (deaths) in the PAN-type vasculitis group, and because a previous study demonstrated that the type of vasculitis did not have an impact on survival (19). For each patient, clinical and biologic data were recorded at the time of the initial evaluation, during followup, and at the end of followup. Clinical data included the following: age, sex, deterioration of general health status (asthenia and/or weight loss ⱖ10% of initial body weight), arthralgia, myalgia, and cutaneous, neurologic, gastrointestinal, and cardiac involvement. Severe gastrointestinal involvement was defined as intestinal bleeding and/or an acute abdomen requiring surgery. Laboratory evaluation included a complete blood cell count, serum chemistry profile, rheumatoid factor analysis, and determination of the level of the serum C4 fraction of complement and the cryoglobulin level. A urinalysis was also completed to screen for hematuria, and a 24-hour urine collection was performed to quantify daily excretion of protein. Cryoglobulin levels were measured and classified as previously described (20). The presence of MC was defined by 1750 TERRIER ET AL Table 2. Baseline epidemiologic, virologic, clinical, and biologic factors associated with outcome in the 151 patients with HCV-related systemic vasculitis* Variable Demographic features Age, years ⱕ65 ⬎65 Sex Female Male HCV infection-related features Metavir activity score 0–2 3–4 Metavir fibrosis score 0–2 3–4 Genotype 1 2 3 4 5 Vasculitis-related features Deterioration of general health Absent Present Arthralgia/arthritis Absent Present Myalgias Absent Present Purpura Absent Present Peripheral neuropathy Absent Present Type of peripheral neuropathy Multiple mononeuropathy Polyneuropathy CNS involvement Absent Present Kidney involvement Absent Present GI involvement Absent Present Severe GI involvement Absent Present Heart involvement Absent Present B cell NHL Absent Present Type of vasculitis Mixed cryoglobulinemia Polyarteritis nodosa Biologic features Creatininemia ⬎140 moles/liter Absent Present No. (% of death)† HR (95% CI)‡ P 96 (10.42) 55 (40) 1 4.55 (2.51–11.06) ⬍0.0001 82 (17.07) 69 (26.09) 1 1.5 (0.75–3.02) 126 (23.81) 10 (20) 1 1.4 (0.33–5.96) 98 (11.22) 48 (43.75) 1 5.31 (2.53–11.14) 84 (22.62) 22 (18.18) 13 (15.38) 11 (9.09) 6 (16.67) 1 0.65 (0.22–1.9) 0.78 (0.18–3.35) 0.4 (0.05–3) 0.7 (0.09–5.23) 116 (18.97) 35 (28.57) 1 1.79 (0.84–3.78) 75 (21.33) 76 (21.05) 1 0.62 (0.31–1.25) 128 (21.09) 23 (21.74) 1 0.93 (0.36–2.42) 47 (25.53) 104 (19.23) 1 0.61 (0.3–1.26) 40 (15) 111 (23.42) 1 1.52 (0.63–3.7) 28 (21.43) 82 (24.39) 1 1.42 (0.57–3.56) 137 (18.98) 14 (42.86) 1 2.74 (1.12–6.7) 101 (18.81) 50 (26) 1 1.91 (0.94–3.87) 139 (21.58) 12 (16.66) 1 0.58 (0.14–2.45) 144 (21.53) 7 (14.28) 1 0.45 (0.06–3.28) 0.25 0.64 ⬍0.0001 0.84 0.13 0.18 0.88 0.18 0.35 0.6 0.027 0.07 0.46 0.43 0.038 145 (20.69) 6 (33.33) 1 4.2 (0.96–18.35) 0.59 129 (20.16) 22 (27.27) 1 1.27 (0.52–3.1) 121 (21.49) 30 (20) 1 0.85 (0.35–2.07) 120 (17.5) 31 (35.48) 1 2.56 (1.23–5.32) 0.72 0.009 PROGNOSTIC FACTORS IN HCV-RELATED VASCULITIS Table 2. 1751 Cont’d Variable Hematuria Absent Present Proteinuria ⬎1 gm/day Absent Present Type II MC Absent Present Cryoglobulin level ⬎1 gm/liter Absent Present Low C4 level Absent Present Rheumatoid factor positivity Absent Present No. (% of death)† HR (95% CI)‡ P 0.036 115 (19.13) 36 (27.78) 1 2.19 (1.03–4.66) 113 (17.7) 38 (31.58) 1 2.34 (1.14–4.81) 37 (29.73) 110 (18.18) 1 0.63 (0.3–1.32) 80 (26.25) 47 (14.89) 1 0.56 (0.24–1.33) 26 (23.08) 93 (18.28) 1 0.79 (0.29–2.09) 32 (37.5) 97 (16.49) 1 0.49 (0.23–1.03) 0.017 0.22 0.19 0.61 0.055 * Data for some patients were missing. HCV ⫽ hepatitis C virus; HR ⫽ hazard ratio; 95% CI ⫽ 95% confidence interval; CNS ⫽ central nervous system; GI ⫽ gastrointestinal; NHL ⫽ non-Hodgkin’s lymphoma. † Values are the number of patients according to the presence or absence of each variable (proportion of patients who died, according to the presence or absence of each variable). ‡ By univariate analysis. a serum cryoglobulin level ⬎0.05 gm/liter on at least 2 occasions. Virologic and immunologic serum markers, HCV antibodies, and RNA were detected as previously described (21). HCV genotyping was performed using a second-generation line probe assay (LiPA; Innogenetics). Liver damage was evaluated according to the previously validated Metavir scoring system (22), either to interpret a liver biopsy (n ⫽ 124) or using noninvasive biochemical markers (n ⫽ 27) (23). Both histologic scoring and noninvasive scoring of liver fibrosis were included in the same analysis to assess the prognostic impact of severe fibrosis on survival. The diagnosis of B cell nonHodgkin’s lymphoma (NHL) was based on World Health Organization criteria (24). The Five-Factors Score (FFS), a prognosis scoring system described by Guillevin et al (25) and used for the management of patients with PAN and Churg-Strauss syndrome, was retrospectively evaluated in the present cohort of patients with HCV-related systemic vasculitis. However, the FFS has never been previously assessed for patients with HCV-related vasculitis and was not intended for use in MC vasculitis. The FFS is based on the following 5 clinical items (the presence of each is accorded 1 point, for a maximum possible score of 5): proteinuria ⬎1 gm/day, renal insufficiency (serum creatinine level ⬎140 moles/liter), cardiomyopathy, severe gastrointestinal involvement, and central nervous system involvement. The FFS was calculated only at the time of the diagnosis of vasculitis, and the baseline clinical and biologic features required for the FFS were available for all patients. The time of therapeutic management and treatment-related data were also recorded. Response to antiviral therapy. The clinical response of vasculitis to antiviral treatment was defined by analyzing the course of the following main clinical signs: skin involvement (absence of purpura), peripheral neuropathy (clinical and/or electrophysiologic improvement at 2 successive examinations), renal involvement (normalization of serum creatinine level and disappearance of proteinuria), and absence of arthralgia. A complete clinical response was defined as an improvement in all baseline clinical manifestations. A partial response was defined as improvement in at least half of the baseline clinical manifestations. All patients who were not classified as having a partial or complete response were classified as nonresponders. A complete immunologic response was defined as the absence of serum cryoglobulin, and a partial immunologic response was defined as a ⬎50% decrease in the baseline cryoglobulin level. Relapse was defined as the reappearance of clinical signs of vasculitis. A sustained virologic response was defined as the absence of detectable serum HCV RNA (⬍12 IU/ml) 6 months after discontinuation of antiviral treatment; the remaining patients were classified as virologic nonresponders. Assessment of the standardized mortality ratio (SMR). To compare the observed mortality with the expected mortality, an SMR was used. The SMR is the ratio of the observed patient mortality and the mortality in the total French population of individuals with the corresponding sex and year of birth. Expected mortality rates were obtained from the latest data from the French national Statistical Institute (INSEE; http://www.ined.fr/fr/pop_chiffres/france/mortalite_ causes_deces/deces_sexe_age), and only observations made before 2007 were considered. Statistical analysis. For the current study, the data for categorical variables are summarized as frequencies and percentages, and quantitative variables are presented as the mean ⫾ SD or medians, 25th and 75th percentiles, and minimum and maximum, as appropriate. Survival curves and their 95% confidence intervals (95% CIs) were estimated using the Kaplan-Meier method. The association of the variables with the occurrence of death was assessed using a univariate 1752 TERRIER ET AL Figure 1. Kaplan-Meier estimates of the survival rates for the 151 patients with hepatitis C virus–related systemic vasculitis. Dashed lines show the 95% confidence interval (95% CI). Cox proportional hazards model. The proportional hazards assumption for the Cox regression model fit were tested for all of the variables. The log-linearity of the quantitative variables was tested using generalized additive models. Age was split into 2 categories according to the estimated median. Multivariate Cox proportional hazards models were performed afterward and adjusted on the dichotomized Metavir fibrosis score (cutoff ⱖ3) that had the highest hazard ratio (HR). The followup length (median, 25th and 75th percentiles) was calculated using an inverse Kaplan-Meier method. A multivariate Cox proportional hazards model was performed, including the dichotomized Metavir F score (cutoff ⱖ3), the FFS, and their interaction. Patients lost to followup were censored on the date of their last clinical evaluation. All tests were 2-sided at the 0.05 significance level. Analyses were performed using the R.2.10.1 statistical package (online at http://www.Rproject.org). RESULTS Characteristics of the patients. One hundred fifty-one patients with HCV-related vasculitis were included. The characteristics of the patients are shown in Table 1. The main HCV genotype was type 1 (62%), and 48 of 146 patients (33%) had severe liver fibrosis, as defined by a Metavir score of ⱖ3 (including grade F3 in 23 patients and grade F4 in 25 patients). The main clinical features of HCV-related vasculitis included pe- ripheral neuropathy (74%), purpura (69%), arthralgia/ arthritis (50%), kidney involvement (33%), central nervous system involvement (9%), gastrointestinal involvement (8%), and heart involvement (4%). Twenty-two patients (14%) had B cell NHL. MC was diagnosed in 121 patients, while 30 patients had PAN-type vasculitis (Table 2). All but 16 patients had detectable cryoglobulin, with a mean ⫾ SD level of 1.32 ⫾ 1.62 gm/liter (range 0.05–7). The mean ⫾ SD serum C4 level was 0.09 ⫾ 0.09 gm/liter (range 0.01–0.41). Outcome. After a median followup of 54 months (interquartile range 26–89 months), 32 patients (21%) had died, including 26 patients with MC vasculitis and 6 with PAN-type vasculitis. Causes of death included infection (n ⫽ 10), end-stage liver disease (n ⫽ 10 [including hepatocellular carcinoma in 4 patients]), nonliver malignancy (n ⫽ 4 [including lung cancer in 2 patients, B cell NHL in 1 patient, and T cell lymphoma in 1 patient), cardiovascular disease (n ⫽ 4), renal failure (n ⫽ 2), central nervous system vasculitis (n ⫽ 2), and unknown cause (n ⫽ 2). Two patients died of both infection and end-stage liver disease. The 1-year, 3-year, 5-year, and 10-year survival rates were 96% (95% CI 93–99), 86% (95% CI 80–92), 75% (95% CI 67–84), and 63% (95% CI 52–76), respectively (Figure 1). Mortality PROGNOSTIC FACTORS IN HCV-RELATED VASCULITIS 1753 Figure 2. Kaplan-Meier estimates of the survival rates for the 151 patients with hepatitis C virus–related systemic vasculitis according to the presence or absence at baseline of severe liver fibrosis (A), heart involvement (B), central nervous system (CNS) involvement (C), and proteinuria (D), the serum creatinine level (E), and the Five-Factors Score (FFS) (F). Survival rates were compared using the log rank test in univariate analysis. Metavir fibrosis scores (F) of 3–4 indicate severe liver fibrosis; scores of 0–2 indicate the absence of liver fibrosis. Figure 3. Kaplan-Meier estimates of the survival rates for the 151 patients with hepatitis C virus–related systemic vasculitis, according to the Metavir fibrosis score (F) and the Five-Factors Score (FFS). Values in brackets are the 95% confidence intervals. NA ⫽ not applicable. 1754 TERRIER ET AL Table 3. Treatments and response to therapy–related factors associated with outcome in the 151 patients with HCV-related systemic vasculitis* Variable Antiviral therapy PEG-IFN␣ plus ribavirin Not used Used IFN plus ribavirin Not used Used IFN monotherapy Not used Used Cotreatment Corticosteroids Not used Used Plasmapheresis Not used Used Immunosuppressive agents Not used Used Rituximab Not used Used Response to therapy Clinical response None Partial Complete Immunologic response None Partial Complete Virologic response Nonsustained Sustained No. (% of death)† HR (95% CI)‡ 51 (41.18) 100 (11) 1 0.34 (0.16–0.67) 126 (18.25) 25 (36) 1 1.23 (0.55–2.86) 137 (17.52) 14 (57.14) 1 3.08 (1.79–18.81) 92 (14.13) 59 (32.2) 1 2.25 (1.06–4.76) 124 (17.74) 27 (37.04) 1 1.90 (0.90–5.22) 138 (17.39) 13 (61.54) 1 4.63 (2.02–10.58) 110 (25.45) 41 (9.76) 1 0.78 (0.28–2.15) 20 (65) 31 (19.35) 83 (10.84) 1 0.25 (0.09–0.66) 0.1 (0.04–0.23) 24 (54.17) 33 (9.09) 61 (11.48) 1 0.17 (0.05–0.58) 0.15 (0.06–0.38) 67 (31.34) 59 (8.47) 1 0.19 (0.07–0.52) P 0.002 0.3 0.003 0.03 0.09 ⬍0.0001 0.63 ⬍0.0001 ⬍0.0001 0.0004 * HCV ⫽ hepatitis C virus; HR ⫽ hazard ratio; 95% CI ⫽ 95% confidence interval; PEG-IFN␣ ⫽ PEGylated interferon ␣. † Values are the number of patients according to the presence or absence of each variable (proportion of patients who died, according to the presence or absence of each variable). ‡ By univariate analysis. was increased among patients with HCV-related vasculitis compared with that among individuals in the total French population with the corresponding sex and year of birth (SMR 2.7 [95% CI 1.8–3.9]). Prognostic factors at baseline. Baseline factors associated with a poor prognosis were determined by univariate analysis (Table 2 and Figure 2). Age ⬎65 years was associated with a poor prognosis (HR 4.55, 95% CI 2.51–11.06 [P ⬍ 0.0001]). The only HCV-related feature associated with a poor prognosis was the presence of severe liver fibrosis as defined by a Metavir fibrosis score of ⱖ3 (HR 5.31, 95% CI 2.53–11.14 [P ⬍ 0.0001]). Clinical manifestations significantly associated with a poor prognosis were central nervous system involvement (HR 2.74, 95% CI 1.12–6.7 [P ⫽ 0.027]), kidney involvement (HR 1.91, 95% CI 0.94–3.87 [P ⫽ 0.07]), and heart involvement (HR 4.2, 95% CI 0.96– 18.35 [P ⫽ 0.038]). Neither gastrointestinal involvement nor the type of vasculitis (MC versus PAN) was associated with a poor outcome. Biologic features associated with a poor prognosis were a serum creatinine level ⬎140 moles/liter (HR 2.56, 95% CI 1.23–5.32 [P ⫽ 0.009]), hematuria (HR 2.19, 95% CI 1.03–4.66 [P ⫽ 0.036]), and proteinuria ⬎1 gm/day (HR 2.34, 95% CI 1.14–4.81 [P ⫽ 0.017]). The FFS was significantly associated with outcome in patients with HCV-related systemic vasculitis (HR 1.85 [95% CI 0.69–6.47] and HR 2.62 [95% CI 1.30–9.22] in patients with scores of 0–1 and ⬎1, respectively; P ⫽ 0.038). In multivariate analysis, the Metavir fibrosis score (HR 10.8, 95% CI 3.63–32.14 [P ⬍ 0.0001]) and the FFS (HR 2.49, 95% CI 1.29–4.8 [P ⫽ 0.007]) PROGNOSTIC FACTORS IN HCV-RELATED VASCULITIS were significantly associated with a poor prognosis. Management of patients after 2001, i.e., after the introduction of combination treatment with PEGylated interferon ␣ (PEG-IFN␣) plus ribavirin in our cohort, was associated with a good prognosis (HR 0.4, 95% CI 0.19–0.81 [P ⫽ 0.009]), whereas no significant difference in clinical and biologic presentation, liver fibrosis, and severity of vasculitis was noted between patients treated before 2001 and those treated after 2001 (data not shown). The interactions between significant prognostic factors were evaluated, i.e., liver fibrosis was assessed using the Metavir fibrosis score, and the severity of vasculitis was assessed using the FFS. Survival curves and rates according to the severity of liver fibrosis, the severity of vasculitis, and their interactions are shown in Figure 3. In patients without severe liver fibrosis (Metavir score ⱕ2), the FFS was a good predictor of outcome (HR 1, 2.49, and 6.2 in patients with scores of 0, 1, and ⬎1, respectively). In contrast, in patients with severe liver fibrosis (Metavir score ⱖ3), the only factor associated with prognosis was the Metavir fibrosis score, while the severity of vasculitis assessed using the FFS no longer had prognostic value (HR 10.8, HR 10.25, and HR 9.74 in patients with scores of 0, 1, and ⬎1, respectively). Prognostic factors during followup. The factors associated with outcome during followup, i.e., the treatments used and the response to therapy, are shown in Table 3. Treatment with PEG-IFN␣ plus ribavirin was associated with a good prognosis (HR 0.34, 95% CI 0.16–0.67 [P ⫽ 0.002]), whereas IFN monotherapy was associated with a poor prognosis (HR 3.08, 95% CI 1.79–18.81 [P ⫽ 0.003]). Corticosteroids (HR 2.25, 95% CI 1.06–4.76 [P ⫽ 0.03]) and immunosuppressive agents (HR 4.63, 95% CI 2.02–10.58 [P ⬍ 0.0001]) were both associated with a poor outcome in univariate analysis. Treatment with plasmapheresis (HR 1.90, 95% CI 0.90– 5.22 [P ⫽ 0.09]) and rituximab (HR 0.78, 95% CI 0.28–2.15 [P ⫽ 0.63]) were not associated with a poor outcome. After adjustment for the severity of vasculitis, treatment with immunosuppressive agents was still associated with a poor prognosis (HR 4.05, 95% CI 1.75–9.36 [P ⫽ 0.001]), while treatment with corticosteroids was not (HR 1.79, 95% CI 0.77–4.16 [P ⫽ 0.17]). Regarding response to therapy, a complete clinical response (HR 0.1, 95% CI 0.04–0.23 [P ⬍ 0.0001]), a complete immunologic response (HR 0.15, 95% CI 0.06–0.38 [P ⬍ 0.0001]), and a sustained virologic response (HR 0.19, 95% CI 0.07–0.52 [P ⫽ 0.0004]) were associated with a good prognosis. 1755 DISCUSSION In the present study, we analyzed prognostic factors in patients with HCV-related systemic vasculitis, during the era of antiviral therapy. The most striking observations were the poor long-term prognosis of patients with HCV-related systemic vasculitis, the major prognostic impact of liver fibrosis at the time of the diagnosis of vasculitis, the negative impact of the severity of vasculitis efficiently assessed by the FFS, and the interaction between these 2 prognostic factors. Patients with HCV-related systemic vasculitis have a poor long-term prognosis, with 5-year and 10year survival rates of 75% and 63%, respectively. Their prognosis is comparable with that of patients with hepatitis B virus–related vasculitis, for whom the 5-year and 10-year survival rates are 73% and 60%, respectively (26). In contrast, the long-term prognosis of patients with chronic HCV infection in the absence of vasculitis is better, with a 5-year survival rate of 95% (27); in a recent study in Norway, the 14-year survival rate was found to be 88% (28). The poor prognosis of patients with HCV-related vasculitis compared with that for patients with chronic HCV infection without vasculitis could be related, in addition to vasculitis-related manifestations, to a higher rate of advanced liver fibrosis in the former group of patients. In our cohort, one-third of patients had severe liver fibrosis at the baseline evaluation. This point is also supported by previous studies showing that patients with HCV-related MC have a higher rate of cirrhosis and mortality than their counterparts without MC (29,30). In addition, patients with HCV-related vasculitis had increased mortality compared with individuals in the total French population with the corresponding sex and year of birth. The use of SMRs is of interest to give a measure of the marginal death rate linked to HCV-related vasculitis. However, SMRs do not account for potential differences in characteristics between the study and reference populations other than the variables used for stratifying the mortality rates. In previous studies focusing on MC-related vasculitis, advanced liver disease was rarely shown to be associated with outcome, and liver complications accounted for ⬃10–15% of the causes of death, far behind MC-related complications and renal involvement (1,11–14). These discrepancies may be explained by the heterogeneity regarding HCV positivity status and the time of inclusion of the patients in previous studies, when the management of vasculitis-related complications was less efficient. In addition, better combinations of antiviral agents have emerged in the past decade, 1756 leading to a change in the management and prognosis of HCV-infected patients. The positive impact of efficient antiviral combination treatment is supported by the fact that patients managed after 2001 (i.e., after the introduction of combination treatment with PEGIFN␣ plus ribavirin in our cohort) had a better prognosis than the others, whereas these patients did not differ regarding clinical and biologic presentation, liver fibrosis, and severity of vasculitis. In contrast, the prognosis of patients treated with previous antiviral regimens providing lower sustained virologic response rate (i.e., standard IFN␣ monotherapy or standard treatment with IFN␣ plus ribavirin) was poorer compared with the prognosis of those treated with PEG–IFN␣ plus ribavirin. Besides liver fibrosis severity, severe vasculitisrelated manifestations were significantly associated with a poor outcome. These findings are consistent with studies focusing on patients with primary systemic vasculitides. Interestingly, the FFS, validated in patients with PAN and Churg-Strauss syndrome (25), was shown to be a good predictor of outcome in patients with HCV-related systemic vasculitis. In primary systemic vasculitides, patients with an FFS ⱖ1 are usually treated with the combination of corticosteroids and immunosuppressive agents, such as cyclophosphamide, in order to improve outcome (25,31). Currently, the standard of care for patients with HCV-related vasculitis includes an optimal antiviral therapy (PEG-IFN␣ plus ribavirin) but no systematic use of immunosuppressive agents (21,32). The analysis of factors associated with outcome during followup showed that treatment with PEG-IFN␣ plus ribavirin was significantly associated with a good prognosis, while the use of immunosuppressive therapy, most often in combination with high-dose corticosteroids, was associated with a poor outcome, even after adjustment for the severity of vasculitis. We used a multivariable model, adjusted notably for the severity of vasculitis, to assess the prognostic impact of treatments, which is the conventional modeling approach used to handle and reduce the impact on such confounders in estimations (33). The use of other approaches such as propensity score methods should be avoided in our case because of the small number of patients and the fact that propensity score approaches are developed to evaluate only one rather than several treatments. The analysis of causes of death in our cohort of patients revealed that the main causes were HCV-related hepatopathy and severe infections in patients treated with immunosuppressive agents, whereas vasculitis-related deaths were less frequent. These find- TERRIER ET AL ings provide evidence on the morbidity and mortality related to the use of immunosuppressive agents in such patients. In contrast, short-term treatment with corticosteroids, rituximab, and/or plasmapheresis, which is not associated with a poor outcome, could be used according to initial presentation in order to control severe and life-threatening manifestations. The absence of a negative impact of rituximab and/or plasmapheresis in patients with HCV-related vasculitis may be explained by the fact that such treatments are frequently introduced in combination with antiviral agents in order to shorten the period of time to clinical remission and improve renal response and cryoglobulin clearance rates, but rarely in association with high-dose corticosteroids. Last, using a multivariate model, we analyzed the interaction between liver fibrosis and severity of vasculitis as assessed by the FFS. We showed that in patients without severe liver fibrosis, the FFS was a good predictor of outcome. However, in patients with severe liver fibrosis, the severity of vasculitis at baseline was not associated with outcome, with the Metavir fibrosis score being the only remaining prognostic factor. These findings may help clinicians in the management of patients with HCV-related systemic vasculitis, in particular those with advanced liver fibrosis. It is noteworthy that in our experience, the use of IFN-based combinations and/or rituximab remains safe in HCV RNA– positive patients with systemic vasculitis and advanced liver fibrosis. Limitations of the study include its open-label and observational design. However, the baseline clinical and biologic findings and information regarding the efficacy and tolerance of treatments were available for almost all patients, except for a very low proportion (5%) who were lost to followup within the first 12 months. In addition, all patients were treated at the same center by the same physician, using homogeneous management. We cannot exclude the possibility that statistically nonsignificant factors may be clinically relevant, and that the observed statistical power may be lower than expected. In conclusion, HCV RNA–positive patients with systemic vasculitis have a poor long-term outcome. At the time of the diagnosis of vasculitis, the main prognostic factors include liver fibrosis and vasculitis-related organ involvement, but in patients with severe liver fibrosis, the severity of vasculitis is not associated with outcome. In contrast with treatment using immunosuppressive agents in primary systemic vasculitides, treatment with immunosuppressive agents seems to have a PROGNOSTIC FACTORS IN HCV-RELATED VASCULITIS negative impact on survival, regardless of the severity of vasculitis. AUTHOR CONTRIBUTIONS All authors were involved in drafting the article or revising it critically for important intellectual content, and all authors approved the final version to be published. Dr. Cacoub had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis. Study conception and design. Terrier, Cacoub. Acquisition of data. Terrier, Saadoun, Sène, Cacoub. Analysis and interpretation of data. Terrier, Semoun, Resche-Rigon, Cacoub. REFERENCES 1. Gorevic PD, Kassab HJ, Levo Y, Kohn R, Meltzer M, Prose P, et al. Mixed cryoglobulinemia: clinical aspects and long-term follow-up of 40 patients. Am J Med 1980;69:287–308. 2. Cacoub P, Costedoat-Chalumeau N, Lidove O, Alric L. Cryoglobulinemia vasculitis. Curr Opin Rheumatol 2002;14:29–35. 3. Sansonno D, Dammacco F. Hepatitis C virus, cryoglobulinaemia, and vasculitis: immune complex relations. Lancet Infect Dis 2005;5:227–36. 4. Mazzaro C, Franzin F, Tulissi P, Pussini E, Crovatto M, Carniello GS, et al. Regression of monoclonal B-cell expansion in patients affected by mixed cryoglobulinemia responsive to ␣-interferon therapy. Cancer 1996;77:2604–13. 5. Ferri C, Greco F, Longombardo G, Palla P, Moretti A, Marzo E, et al. Antibodies to hepatitis C virus in patients with mixed cryoglobulinemia. Arthritis Rheum 1991;34:1606–10. 6. Agnello V, Chung RT, Kaplan LM. A role for hepatitis C virus infection in type II cryoglobulinemia. N Engl J Med 1992;327: 1490–5. 7. Cacoub P, Fabiani FL, Musset L, Perrin M, Frangeul L, Leger JM, et al. Mixed cryoglobulinemia and hepatitis C virus. Am J Med 1994;96:124–32. 8. Cacoub P, Poynard T, Ghillani P, Charlotte F, Olivi M, Piette JC, et al, for the MULTIVIRC Group. Extrahepatic manifestations of chronic hepatitis C. Arthritis Rheum 1999;42:2204–12. 9. Vassilopoulos D, Calabrese LH. Hepatitis C virus infection and vasculitis: implications of antiviral and immunosuppressive therapies [review]. Arthritis Rheum 2002;46:585–97. 10. Terrier B, Sene D, Dechartres A, Saadoun D, Ortonne N, Rouvier P, et al. Systemic vasculitis in patients with hepatitis C virus infection with and without detectable mixed cryoglobulinemia. J Rheumatol 2011;38:104–10. 11. Ferri C, Sebastiani M, Giuggioli D, Cazzato M, Longombardo G, Antonelli A, et al. Mixed cryoglobulinemia: demographic, clinical, and serologic features and survival in 231 patients. Semin Arthritis Rheum 2004;33:355–74. 12. Tarantino A, Campise M, Banfi G, Confalonieri R, Bucci A, Montoli A, et al. Long-term predictors of survival in essential mixed cryoglobulinemic glomerulonephritis. Kidney Int 1995;47: 618–23. 13. Meltzer M, Franklin EC, Elias K, McCluskey RT, Cooper N. Cryoglobulinemia: a clinical and laboratory study. II. Cryoglobulins with rheumatoid factor activity. Am J Med 1966;40: 837–56. 14. Della Rossa A, Tavoni A, D’Ascanio A, Catarsi E, Marchi F, Bencivelli W, et al. Mortality rate and outcome factors in mixed cryoglobulinaemia: the impact of hepatitis C virus. Scand J Rheumatol 2010;39:167–70. 15. Ramos-Casals M, Robles A, Brito-Zeron P, Nardi N, Nicolas JM, Forns X, et al. Life-threatening cryoglobulinemia: clinical and immunological characterization of 29 cases. Semin Arthritis Rheum 2006;36:189–96. 1757 16. Landau DA, Scerra S, Sene D, Resche-Rigon M, Saadoun D, Cacoub P. Causes and predictive factors of mortality in a cohort of patients with hepatitis C virus-related cryoglobulinemic vasculitis treated with antiviral therapy. J Rheumatol 2010;37:615–21. 17. Jennette JC, Falk RJ. Small-vessel vasculitis. N Engl J Med 1997;337:1512–23. 18. Lightfoot RW Jr, Michel BA, Bloch DA, Hunder GG, Zvaifler NJ, McShane DJ, et al. The American College of Rheumatology 1990 criteria for the classification of polyarteritis nodosa. Arthritis Rheum 1990;33:1088–93. 19. Saadoun D, Terrier B, Semoun O, Sene D, Maisonobe T, Musset L, et al. Hepatitis C virus-associated polyarteritis nodosa. Arthritis Care Res (Hoboken) 2011;63:427–35. 20. Musset L, Diemert MC, Taibi F, Thi Huong Du L, Cacoub P, Leger JM, et al. Characterization of cryoglobulins by immunoblotting. Clin Chem 1992;38:798–802. 21. Cacoub P, Lidove O, Maisonobe T, Duhaut P, Thibault V, Ghillani P, et al. Interferon-␣ and ribavirin treatment in patients with hepatitis C virus–related systemic vasculitis. Arthritis Rheum 2002;46:3317–26. 22. The French Metavir Cooperative Study Group. Intraobserver and interobserver variations in liver biopsy interpretation in patients with chronic hepatitis C. Hepatology 1994;20:15–20. 23. Imbert-Bismut F, Ratziu V, Pieroni L, Charlotte F, Benhamou Y, Poynard T. Biochemical markers of liver fibrosis in patients with hepatitis C virus infection: a prospective study. Lancet 2001;357: 1069–75. 24. Jaffe ES. The 2008 WHO classification of lymphomas: implications for clinical practice and translational research. Hematology Am Soc Hematol Educ Program 2009:523–31. 25. Guillevin L, Lhote F, Gayraud M, Cohen P, Jarrousse B, Lortholary O, et al. Prognostic factors in polyarteritis nodosa and Churg-Strauss syndrome: a prospective study in 342 patients. Medicine (Baltimore) 1996;75:17–28. 26. Pagnoux C, Seror R, Henegar C, Mahr A, Cohen P, Le Guern V, et al, for the French Vasculitis Study Group. Clinical features and outcomes in 348 patients with polyarteritis nodosa: a systematic retrospective study of patients diagnosed between 1963 and 2005 and entered into the French Vasculitis Study Group database. Arthritis Rheum 2010;62:616–26. 27. Ngo Y, Munteanu M, Messous D, Charlotte F, Imbert-Bismut F, Thabut D, et al. A prospective analysis of the prognostic value of biomarkers (FibroTest) in patients with chronic hepatitis C. Clin Chem 2006;52:1887–96. 28. Kristiansen MG, Lochen ML, Gutteberg TJ, Mortensen L, Eriksen BO, Florholmen J. Total and cause-specific mortality rates in a prospective study of community-acquired hepatitis C virus infection in northern Norway. J Viral Hepat 2011;18:237–44. 29. Kayali Z, Buckwold VE, Zimmerman B, Schmidt WN. Hepatitis C, cryoglobulinemia, and cirrhosis: a meta-analysis. Hepatology 2002;36:978–85. 30. Saadoun D, Asselah T, Resche-Rigon M, Charlotte F, Bedossa P, Valla D, et al. Cryoglobulinemia is associated with steatosis and fibrosis in chronic hepatitis C. Hepatology 2006;43:1337–45. 31. Gayraud M, Guillevin L, le Toumelin P, Cohen P, Lhote F, Casassus P, et al, and the French Vaculitis Study Group. Longterm followup of polyarteritis nodosa, microscopic polyangiitis, and Churg-Strauss syndrome: analysis of four prospective trials including 278 patients. Arthritis Rheum 2001;44:666–75. 32. Saadoun D, Resche-Rigon M, Thibault V, Piette JC, Cacoub P. Antiviral therapy for hepatitis C virus–associated mixed cryoglobulinemia vasculitis: a long-term followup study. Arthritis Rheum 2006;54:3696–706. 33. Stukel TA, Fisher ES, Wennberg DE, Alter DA, Gottlieb DJ, Vermeulen MJ. Analysis of observational studies in the presence of treatment selection bias: effects of invasive cardiac management on AMI survival using propensity score and instrumental variable methods. JAMA 2007;297:278–85.