0031-3998/01/5003-0345 PEDIATRIC RESEARCH Vol. 50, No. 3, 2001 Copyright © 2001 International Pediatric Research Foundation, Inc. Printed in U.S.A. The Varicella-Autoantibody Syndrome CASSANDRA JOSEPHSON, RACHELLE NUSS, LINDA JACOBSON, MICHELE R. HACKER, JAMES MURPHY, ADRIANA WEINBERG, AND MARILYN J. MANCO-JOHNSON Departments of Pediatrics [C.J., R.N., L.J., M.R.H., A.W., M.J.M.-J.] and Preventive Medicine and Biometrics [J.M.], University of Colorado Health Sciences Center, Denver, Colorado 80262, U.S.A., and The Children’s Hospital, Denver, Colorado 80218, U.S.A. [C.J., R.N., L.J., M.R.H., A.W., M.J.M.-J.] ABSTRACT This cross-sectional study was conducted to determine the increased PS IgG antibody (p Ͻ 0.001) compared with the incidence of autoantibodies to phospholipids and coagulation children without acute VZV. For all groups combined, elevated proteins in children with acute varicella zoster virus (VZV) PS IgG antibody showed negative correlation with free PS (p Ͻ infection. Study groups included children with VZV alone or 0.0001) and positive correlation with prothrombin fragment 1ϩ2 complicated by purpura fulminans and/or thromboembolism. (p ϭ 0.0002). Autoantibodies were transient. Transient antiphos- VZV naïve children and children who had VZV Ͼ1 y before pholipid and coagulation protein autoantibodies were common sample collection formed a control group. Blood was assayed for with VZV infection, but were not predictive of thrombotic the following: free protein S (PS), protein C, antithrombin, and complications. (Pediatr Res 50: 345–352, 2001) prothrombin; antibody binding to these proteins; lupus anticoag- ulant; anticardiolipin antibody; antiphospholipid antibodies; and Abbreviations prothrombin fragment 1ϩ2. Data regarding coinfections was ACA, anticardiolipin antibody collected. Forty-three VZV-infected children showed an in- APA, antiphospholipid antibody creased frequency of lupus anticoagulant, anticardiolipin anti- AT, antithrombin body, antiphospholipid antibodies, and autoantibodies to PS, F II, prothrombin, or factor II protein C, prothrombin, and antithrombin in comparison to 52 F1؉2, prothrombin fragment 1ϩ2 children without acute VZV (p Ͻ 0.0001). Seventeen children GABHS, group A ␤-hemolytic Streptococcus LA, lupus with VZV and purpura fulminans and/or thromboembolism anticoagulant showed a statistically significant decrease in free PS, signifi- PC, protein C cantly increased PS IgG antibody, and significantly increased PF, purpura fulminans prothrombin fragment 1ϩ2(p Ͻ 0.0001) compared with the PF؎TE, purpura fulminans and/or thromboembolism group without acute VZV and the group with uncomplicated PS, protein S VZV. Twenty-six children with uncomplicated VZV showed VZV, varicella zoster virus In a 1990 review, Francis (1) reported the association of children with VZV infection. The aim of this study was to VZV and/or streptococcal infection in 30% of children with compare results obtained on blood samples from children with idiopathic purpura fulminans. Subsequently, acquired free PS uncomplicated acute VZV infection with samples from chil- deficiency was described in children and adults with acute dren with acute VZV complicated by symptomatic PFϮTE. It VZV infection and PFϮTE (2–8). We reported the LA and was hypothesized that coagulation abnormalities in the chil- free PS deficiency in seven previously healthy children with dren with symptomatic PFϮTE would be more prevalent and acute VZV infection complicated by PFϮTE (6). We previ- more severe but not intrinsically different from those found in ously found evidence suggestive of an autoimmune cause of healthy children with uncomplicated VZV infection. acquired PS deficiency. We now report a cross-sectional study designed to determine METHODS the prevalence of autoantibodies directed against a variety of phospholipids and coagulation proteins in previously healthy Clinical Methods Received January 4, 2000; accepted February 26, 2001. This cross-sectional study was performed with the approval Correspondence and reprint requests: Marilyn J. Manco-Johnson, M.D., Mountain of the Colorado Multi-Institutional Review Board and the States Regional Hemophilia and Thrombosis Center, PO Box 6507, Mail Stop F416, Pediatric Clinical Research Center at The Children’s Hospital Aurora, CO 80045-0507; e-mail: [email protected] Supported by NIH grant M01 RR00069, from the General Clinical Research Centers of Denver. Informed consent and assent forms were signed Program, National Center for Research Resources. before enrollment in the study. 345 346 JOSEPHSON ET AL. Three study groups of children were identified and recruited from Enzyme Research (South Bend, IN, U.S.A.). Fibrinogen concurrently. Group 1 was composed of children who had no was purchased from Kabi Vitrum (Stockholm, Sweden). history of VZV infection or had VZV Ͼ1 y before sample Antibody to VZV. An IgG antibody response to VZV was collection, no personal or family history of bleeding or throm- determined using an ELISA assay (VZV STAT, BioWhittaker). botic disorders, and no evidence of infection or antibiotic use Lupus anticoagulant. The LA was deemed present if there for at least 4 wk before blood sampling. Group 2 included was prolongation of the activated partial thromboplastin time children with acute uncomplicated VZV infection and no (APTT) or dilute Russell viper venom time that did not correct personal or family history of bleeding or thrombotic disorders. with the addition of 1:1 standard normal plasma and corrected Group 3 was composed of children with acute VZV infection with addition of platelet phospholipid in accordance with the complicated by PFϮTE. criteria of the Lupus Anticoagulant Subcommittee of the In- Children were recruited at Pediatric Resident Continuity ternational Society for Thrombosis and Haemostasis (9). Clinics at Denver Health Medical Center, The Children’s Anticardiolipin antibody. IgG and IgM antibodies to cardi- Hospital of Denver, from staff members at The University of olipin were determined in ELISA assays (Inova, San Diego, Colorado Health Sciences Center, or by referral to one of the CA, U.S.A.) (10). authors. The diagnosis of VZV was made on visual inspection Antiphospholipid antibody. IgG and IgM antibodies in a of typical skin lesions and confirmed with IgG serology. The mixture of cardiolipin, phosphatidic acid, and phosphatidylser- diagnosis of purpura fulminans was made clinically and in- ine were semiquantitatively measured in an ELISA assay ␤ cluded typical palpable, indurated skin lesions of the trunk and containing 2-glycoprotein I (Asserachrom APA, American extremities progressing from dark red to black and resulting in Bioproducts, Parsippany, NJ, U.S.A.). In this manuscript, APA skin necrosis. All diagnoses of purpura fulminans were made will represent this specific ELISA test and not the more general in the pediatric intensive care unit of The Children’s Hospital term, antiphospholipid antibody. of Denver. All diagnoses of thromboembolism were confirmed Plasma concentrations of coagulation proteins. Free PS with high-resolution real-time and Doppler ultrasonography, antigen concentrations were determined in an ELISA assay magnetic resonance angiography, or pulmonary ventilation/ (Asserachrom Free Protein S, Diagnostic Stago, France). For perfusion scanning as appropriate. The diagnosis of Strepto- this study we chose to assay free PS, or the compartment of PS coccus was made on the basis of the result of clinically ordered not bound to circulating C4b-binding protein, because the free blood or skin cultures. compartment of PS contributes anticoagulant function and Children between the ages of 3 mo and 18 y were eligible for because free PS deficiency was found to be most severe in the the study. Nine milliliters of blood drawn into 1 mL of 3.8% seven originally published patients with VZV-induced PS de- sodium citrate and 5 mL of whole blood were collected from ficiency. PC activity was quantified in a chromogenic assay each subject. Citrated samples were immediately placed on ice (Coamatic Protein C, Chromogenix, Milan, Italy). F II activity and centrifuged twice at 1800 ϫ g for 20 min at 4°C. Whole was determined in a standard one-stage clotting assay. AT blood samples were allowed to clot at 37°C for 1 h before activity was assessed in an amidolytic assay (Coamatic Anti- separating serum. Aliquots of plasma and serum were frozen at thrombin III, Chromogenix). Ϫ70°C until the time of assay. Prothrombin fragment 1؉2. Plasma concentrations of the Three children whose samples were included in group 1 prothrombin activation fragment, F 1ϩ2, were determined subsequently developed VZV infection. They and their parents using an ELISA assay (Ezygnost F 1ϩ2, Behring Diagnostics agreed to additional blood sampling at the time of acute Inc., Westwood, MA, U.S.A.). uncomplicated VZV infection and 3 mo after VZV infection. Plasma concentrations of IgG and IgM antibodies to co- Blood samples drawn at the time of acute uncomplicated VZV agulation proteins. Binding of patient plasma IgG or IgM to from these three children were included in group 2. coagulation proteins was determined in ELISA assays. One hundred microliters of purified capture antigen diluted in car- Laboratory Methods bonate buffer was applied to a 96-well ELISA plate (Nunc Maxisorb substrate kit (contains peroxidase solution and 3, 3Ј, Plasma concentration of the vitamin K-dependent anticoag- 5, 5Ј tetramethylbenzidine, Pierce Chemical Co., Rockford, IL, ulant PS was determined for the