acute prognostic factors for post-thrombotic syndrome in children with limb dvt: a bi-institutional...
TRANSCRIPT
Thrombosis Research 131 (2013) 37–41
Contents lists available at SciVerse ScienceDirect
Thrombosis Research
j ourna l homepage: www.e lsev ie r .com/ locate / thromres
Regular Article
Acute Prognostic Factors for Post-Thrombotic Syndrome in Children with Limb DVT:A Bi-Institutional Cohort Study
Courtney A. Lyle a,⁎, Elizabeth Gibson b, Amy E. Lovejoy a, Neil A. Goldenberg b,c,1
a Department of Pediatrics, Division of Hematology/Oncology, University of California San Diego and Rady Children's Hospital, San Diego, CA, USAb Department of Pediatrics, Sections of Hematology/Oncology/Bone Marrow Transplantation and the Mountain States Regional Hemophilia and Thrombosis Center, University of Colorado,Aurora, CO, USAc Department of Medicine, Sections of Hematology/Oncology/Bone Marrow Transplantation and the Mountain States Regional Hemophilia and Thrombosis Center, University of Colorado,Aurora, CO, USA
Abbreviations: ACA, anticardiolipin antibodies; BMI,for Disease Control; CTV, computed tomogram with vHospital Colorado; dRVVT, dilute Russell viper venom testELISA, enzyme-linked immunosorbant assay; IVC, inferioronance venogram; PTS, post-thrombotic syndrome; RCHSDiego; SVC, superior vena cava; VTE, venous thromboeOrganization.⁎ Corresponding author at: Department of Pediatrics, Di
Billings Clinic, 2800 10th Ave North, Billings, MT 59101, UE-mail address: [email protected] (C.A. Lyle).
1 Current affiliation is: Departments of Pediatrics and Mgy, Johns Hopkins University School of Medicine, BaltimoResearch Institute, All Children's Hospital—Johns HopkinUSA.
0049-3848/$ – see front matter © 2012 Elsevier Ltd. Allhttp://dx.doi.org/10.1016/j.thromres.2012.10.012
a b s t r a c t
a r t i c l e i n f oArticle history:
Received 10 August 2012Received in revised form 6 October 2012Accepted 24 October 2012Available online 26 November 2012Keywords:Deep venous thrombosisPost-thrombotic syndromeDilute Russell Viper Venom TestChildrenVenous thromboembolism
Background: Early identification of children with deep venous thrombosis (DVT) of the limb who are at height-ened risk for post-thrombotic syndrome (PTS) is important in order to evaluate therapeutic interventions aimedat decreasing the risk and severity of PTS.Objective: We sought to evaluate acute prognostic factors for PTS in children following DVT of the limbs.Materials and Methods: In this bi-institutional mixed cohort study with prospective ascertainment of PTS using avalidated pediatric instrument, we collected data on patient/thrombus characteristics, thrombophilia testingresults, and outcomes in children (b21 years at event) diagnosed with acute limb DVT at Rady Children'sHospital of San Diego and Children's Hospital Colorado.Results:Median age at presentationwas 13 years (range, 0–18 years). Cumulative incidence (i.e. risk) of PTSwas23%, at a median follow-up duration of 33 months (range, 13.2-65 months). The presence of a lupus anticoagu-lant by dilute Russell Viper venom time (dRVVT) testing within twoweeks of DVT diagnosiswas associatedwith
markedly increased odds of developing clinically-significant PTS (OR: 16.8, 95%CI 1.60-176.2; P=0.02). Thepresence of an infectious or inflammatory condition at DVT presentation was neither associated with PTS risknor dRVVT positivity. CONCLUSION: An acutely positive dRVVT following diagnosis of limb DVT appears to bea significant prognostic factor for development of clinically significant PTS in children. Larger collaborative cohortstudies are required to substantiate these findings, evaluate other prognostic factors, and determinewhether thepresent association is modulated by persistent dRVVT positivity or beta-2-glycoprotein-I dependence.© 2012 Elsevier Ltd. All rights reserved.
Introduction
While venous thromboembolism (VTE) in children is a seeminglyrare condition, the incidence is increasing dramatically and is associatedwith significant long-term morbidity. The estimated annual incidenceof VTE in children is 0.07-0.14/10,000; however, among hospitalized
body mass index; CDC, Centersenography; CHCO, Children's; DVT, deep venous thrombosis;vena cava; MRV, magnetic res-D, Rady Children's Hospital Sanmbolism; WHO, World Health
vision ofHematology/Oncology,SA. Tel.: +1 406 238 2500.
edicine, Divisions of Hematolo-re, MD, USA and All Children'ss Medicine, St. Petersburg, FL,
rights reserved.
patients this is estimated at 58/10,000 admissions and is continuingto rise, likely as a result of improvements in intensive interventionsfor critically ill children [1–3]. Nearly 10% of children experience arecurrent event within one year and approximately 25% with deep ve-nous thrombosis (DVT) affecting the limbs develop post-thromboticsyndrome (PTS), a manifestation of chronic venous insufficiency char-acterized by pain, swelling, ulceration, and/or functional impairment[4,5].
Although an elevated D-dimer and factor VIII level at diagnosis havebeen associated with a poor outcome (lack of thrombus resolution,recurrent thrombosis, or PTS) following VTE in children, better delin-eation of prognostic factors specifically for PTS in children has beenlimited [4]. Kuhle and colleagues reported that children with lack ofresolution of DVT were nearly four times more likely to develop PTSthan children with complete thrombus resolution [6]. However, thisimportant studydid not employ a validated pediatric PTS outcomemea-sure.More recently, in a small prospective cohort employing a validatedpediatric PTS outcomemeasure, Goldenberg et al. demonstrated a highrisk of PTS in children with completely veno-occlusive DVT in whomfactor VIII and D-dimer were both elevated [7].
Fig. 1. Flow diagram of disposition of events in the study population. RCHSD: RadyChildren's Hospital San Diego, CHCO: Children's Hospital Colorado.
38 C.A. Lyle et al. / Thrombosis Research 131 (2013) 37–41
Early identification of children with DVT who are at greatest riskfor PTS is important in order to evaluate therapeutic interventionsaimed at decreasing the risk and severity of PTS. Accordingly, in thisbi-institutional mixed cohort study with prospective ascertainment ofPTS, we sought to evaluate acute prognostic factors for PTS in childrenfollowingDVT of the limbs.We hypothesized that acute hypercoagulablestates frequentlymediated by inflammation (e.g., antiphospholipid anti-bodies and elevated factor VIII) would be prognostic of pediatric PTS.
Materials and Methods
Subjects
With institutional review board approvals, we conducted a mixedcohort study of children presenting with acute limb DVT at RadyChildren's Hospital San Diego (RCHSD) and Children's Hospital Colorado(CHCO). Patients presenting between January 2005 and March 2010 atRCHSD were identified retrospectively from the hematology database,enrolled after informed consent discussion and signature, and followedprospectively through September 2011. At CHCO study subjects wereenrolled with written informed consent in a prospective cohort studybetween March 2006 and December 2010; data for patients diagnosedwith acute VTE prior to March 2006 were collected retrospectivelyprior to that date and prospectively (including PTS) thereafter. Inclusioncriteria for the present sub-cohort across the two institutions consistedof age b21 years at diagnosis, acute limbDVT (all deep venous extremityveins and isolated superior vena cava (SVC), inferior vena cava (IVC),and azygous veins) confirmed by radiological imaging (compression ul-trasound with Doppler, computed tomogram with venography [CTV],and/ormagnetic resonance venogram [MRV]). Exclusion criteria includ-ed the following: initial treatment with thrombolysis or thrombectomyand lack of PTS evaluation 12 or more months following DVT diagnosis.
Data Collection
The following clinical data were collected: age, body mass index(BMI), thrombus location, presence of a central venous catheter atthe affected site (or within the preceding 30 days in the same vesselas VTE), co-morbid conditions, therapeutic intervention, recurrentthrombotic events, length of medical follow-up, laboratory analysisof inherited and acquired thrombotic risk factors, and occlusion statusof vessel. We determined BMI percentiles using Centers for DiseaseControl (CDC) growth charts for children≥2 years of age and WorldHealth Organization (WHO) growth charts for children less than2 years of age. A BMI was not calculated for preterm infants less than1 year of age at the time of diagnosis.
Results of thrombophilia assessment were collected for analysis asputative prognostic factors if obtainedwithin twoweeks of presentation,not including genetic testing (which could be obtained at any timepoint). Thrombophilia testing performed in the local and referral clinicallaboratories consisted of the following: protein C and antithrombin activ-ities by chromogenic assay; free protein S antigen by enzyme-linkedimmunosorbant assay (ELISA); factor V Leiden and prothrombinG20210A polymorphisms by polymerase chain reaction; lupus anticoag-ulant testing by dilute Russell viper venom test (dRVVT); factor VIIIactivity by one-stage clotting assay; and D-dimer by either quantitativeor semi-quantitative immunoturbidimetric assay or fluorescence immu-noassay. Anticardiolipin antibodies, ß2-gylcoprotein-I antibodies, andhomocysteine were infrequently tested at RCHSD. Severe protein C defi-ciency was defined in children greater than three months of age with alevel b20% and a mild deficiency if the level was b40%. A severe proteinS deficiencywas defined if the levelwas b20% and amild deficiency if thelevel was b40%, irrespective of age. We characterized a severe anti-thrombin deficiency in children greater than three months of age witha level b30% and a mild deficiency if the level was b60%. The dRVVT
was positive if the screen:confirm ratio was >1.20 at RCHSD or >1.22at CHCO.
In concordance with present guidelines for pediatric PTS outcomereporting we evaluated the principal outcome, clinically-significantPTS, using a standardized outcome assessment tool; in this case, theManco-Johnson instrument was employed [7,8]. This outcome assess-ment tool combines a scoring system for the presence of physical exam-ination abnormalities including edema, superficial collateral veins,venous stasis dermatitis, and venous stasis ulcers with a scoring systemfor a history of functional limitations of activities of daily living orchronic lower extremity pain either with physical activity or at rest.Clinically-significant PTS was defined as a score of at least 1 in bothcategories. Evaluation for PTS was undertaken at ≥1 year. For serialmeasurements beyond one year, the latest measurement was used foranalysis.
Statistical Analysis
Fisher's exact testing or Mann Whitney U testing was employed tocompare differences in frequencies or distributions (respectively) ofvariables of interest between subjects with and without PTS. Univariatelogistic regression was performed to evaluate risk factors for PTS.Variables included age, BMI, infectious or inflammatory condition atdiagnosis, and thrombophilia findings. Thrombophilia results wereevaluated categorically, with the exception of factor VIII levels, whichwere analyzed as a continuous variable. We determined a priori thatantiphospholipid antibody as a category would be highly collinearwith dRVVT as a specific test for the lupus anticoagulant; therefore,only one of the two variables would be included in a multivariatemodel. We further determined a priori that variables for which Pb0.1
Table 2Univariate Logistic Regression Analysis of Prognostic Factors for Clinically-SignificantPTS⁎ Following Diagnosis of VTE.
Prognostic Factor Univariate Odds Ratio(95% CI)
P value
Age 1.15 (0.96-1.38) 0.12Body mass index 0.99 (0.96-1.02) 0.38Male Gender 0.50 (0.10-2.48) 0.40Central Venous Line 0.42 (0.07-2.45) 0.33Veno-Occlusion at Diagnosis 4.15 (0.42-40.6) 0.22Proximal DVT 0.88 (0.08-9.79) 0.91Inflammatory or Infectious condition 0.87 (0.17-4.43) 0.87Factor VIII 1.00 (0.99-1.02) 0.78dRVVT 16.8 (1.60-176.2) 0.02D-dimer>500 ng/mL or equivalent‡ 2.14 (0.2-22.5) 0.53Inherited thrombophilia# 1.72 (0.32-9.10) 0.53
⁎ Defined as a score of 1 or greater in both physical examination and functional limita-tion categories, according to the Manco-Johnson instrument, a standardized PTS assess-ment tool.
‡ Values of 500 ng/mL at RCHSD and CHCO from 2006–2008 were determined to beequivalent to laboratory values of 1.0 ug/mL obtained after 2008 at CHCO.
# Defined as presence of one or more of the following: factor V Leiden polymor-phism, prothrombin G20210A polymorphism, or plasma protein C activityb40% after6 months of age; plasmas free protein S antigen level b40%, or plasma antithrombinactivityb60% after 6 months of age.
39C.A. Lyle et al. / Thrombosis Research 131 (2013) 37–41
in univariate logistic regression would be included in a multiple logisticregression model; however, given that only one variable met thiscriterion, no adjustment via multivariate analysis was necessary (seeResults). All hypothesis tests were two-tailed and performed usingSAS 9.2 statistical software (SAS Institute, Cary, NC, USA).
Results
Sixty-two subjects with limb DVT across the RHCSD and CHCO sitesmet eligibility criteria, 35 ofwhom (56%) underwent a PTS evaluation at≥1 year post-diagnosis. Patient selection and disposition are shown inFig. 1, via a flow diagram. Table 1 summarizes subject characteristicsseparated by outcome group (presence versus absence of PTS atleast 12 months following diagnosis). Median age at presentation was13 years (range, 0–18 years). Cumulative incidence (i.e. risk) of clini-cally significant PTS was 23%, at a median follow-up duration (fromdiagnosis to PTS assessment) of 33 months (range, 13.2-65 months).Physical findings of PTS in most patients at both centers were com-prised of chronic intermittent or persistent swelling; none involvedvenous stasis ulceration. Functional symptoms consisted mainly ofpain that limited aerobic activities +/− pain with activities of dailyliving.
Table 2 presents the results of univariate logistic regression forputative prognostic factors for PTS.
While complete veno-occlusion was present acutely in all childrenwho developed PTS as compared to only 56% of children without PTS,acute veno-occlusion was not identified as a prognostic factor for PTSin logistic regression (OR: 4.15, 95% CI: 0.42-40.6; P=0.22). Follow-up duration was slightly longer among subjects who did, versus didnot, develop PTS (40 months vs. 30 months, respectively), this difference
Table 1Patient Characteristics by Outcome Group.
Variable All DVT eventsn=35
Cln=
Age (year)Median 13 15Range 0.2-18 0.
Gender (%)Male 22 (63) 4Female 13 (37) 4
BMI %Median 73.1 68Range 0.1-99.8 0.
Clinical Characteristics (%)Central Line Associated‡ 14 (40) 2Proximal Location 31 (89) 7Infection or Inflammatory ConditionΔ 14 (40) 3
Laboratory ParametersFV Leiden Heterozygote – (%) 7 (20) 2FV Leiden Homozygote – (%) 1 (3) 1FII G20210A heterozygote– no. tested (%) 0/34 (0) 0/Moderate/Severe protein C, protein S, or ATdeficiency – no. tested (%)
3/30 (10) 0/
FVIII Level (IU/dL) (n=19) (nMedian (n =19) 221 22Range 99.9-413.7 21
dRVVT – no. positive /tested (%) 11/26 (42) 6/D-dimer>500 (ng/mL) (%)⁎ 20/26 (77) 6/
Veno-Occlusion at Diagnosis (%) 23 (66) 8Veno-Occlusion 3 months after diagnosis (%) 7(20) 2Recurrent VTE (%) 7 (20) 2Length of Follow-Up (months)
Median 33 40Range 13.2-65 19
‡ Central line in the affected vessel at time of VTE or within the preceding 30 days.Δ Associated Inflammatory/Infectious Conditions included documented acute infection
osteomyelitis n=1, necrotizing enterocolitis n=1, central nervous system tuberculosis nketoacidosis n=1, systemic lupus erythematosus n=1, anti-NMDA receptor encephalitis n⁎ Values of 500 ng/mL at RCHSD and CHCO from 2006–2008 were determined to be equ
was also not clinically significant (P=0.27). A positive dRVVTwithin twoweeks of diagnosiswas associatedwith increased odds of developing PTS(OR: 16.8, 95% CI: 1.60-176.2; P=0.02). Additional antiphospholipidantibodies were not evaluated in the analysis, as anticardiolipin andanti-ß2-glycoprotein antibodies were not uniformly tested at bothcenters (see also Methods). The presence of an infectious (e.g. sepsis,
inically Significant PTS8
No Clinically Significant PTSn=27
P value
.5 10 0.066-17 0.2-18
(50) 18 (67) 0.41(50) 9 (33)
.75 76.45 0.481-96 18.9-99.8
(25) 12 (44) 0.43(88) 24 (89) 1.00(38) 11 (41) 1.00
(25) 5 (19) 1.00(13) 0 (0) 0.238 (0) 0/26 (0) 1.007 (0) 3/23 (13) 1.00
=4) (n=15)6.5 208 0.555.9-295.5 99.9-413.77 (86) 5/19 (26) 0.026 (100) 14/20 (70) 0.28(100) 15 (56) 0.03(25) 5 (19) 0.65(25) 5(19) 0.65
30 0.27-60 13-65
at diagnosis (sepsis n=2, influenza n=2, pneumonia n=2, viral encephalitis n=1,=1), inflammatory bowel disease n=1, hemolytic uremic syndrome n=2, diabetic=1, malignancy n=1. (Several subjects with>1 disorder.)ivalent to laboratory values of 1.0 ug/mL obtained after 2008 at CHCO.
40 C.A. Lyle et al. / Thrombosis Research 131 (2013) 37–41
osteomyelitis, or influenza) or inflammatory trigger (e.g. inflammatorybowel disease, systemic lupus erythematosus, or malignancy) at presen-tation was not significantly associated with PTS (OR: 0.87, 95% CI:0.17-4.43; P=0.87), suggesting that the association between dRVVTpositivity and increased PTS risk was not mediated by overt infection orinflammation, but could still be associated with sub-clinical inflamma-tion. This was further supported by a lack of association betweendRVVT positivity and presence of an infectious/inflammatory disorderat diagnosis (P=1.00).
Discussion
In this study, we evaluated the influence of clinical characteristicsand laboratory parameters at diagnosis of limb DVT on the risk ofdeveloping clinically significant PTS in children. The cumulative inci-dence of PTS defined in this cohort, at 23%, is concordantwith publishedfindings from a systematic review of the pediatric PTS literature [5]. Ourfindings suggest a nearly seventeen-fold increased risk of PTS in chil-dren with limb DVT who have positive lupus anticoagulant testing bydRVVT within two weeks of DVT diagnosis. This observation appearedto be independent of the presence of an underlying infectious or inflam-matory condition at DVT diagnosis.
The dRVVT is one of several laboratory assays to evaluate the pres-ence of a lupus anticoagulant, a common risk factor for the developmentof VTE in children [9–11]. In a prospective study of 110 adult patientswith DVT, Roumen-Klappe and colleagues demonstrated that elevatedinflammatory markers at diagnosis (IL-6 and CRP) were associatedwith increased venous outflow resistance (ameasure of residual throm-bus). Additionally, subjects with increased outflow resistance threemonths following diagnosis had an increased risk of developing PTS atleast one year after DVT diagnosis [12]. Shbaklo et al. also identifiedan association of elevated IL-6 and ICAM-1 four months following DVTdiagnosis with an increased risk of PTS during a two-year follow-upperiod [13]. While the mechanism has not been fully elucidated, it islikely that ongoing inflammatory processes contribute to a delay inthrombus resolution, resulting in persistent venous obstruction andan increased risk of PTS. Although a clinical diagnosis of a systemicinflammatory or autoimmune condition was not significantly associat-ed with PTS risk in our study, it is possible that a subclinical inflamma-tory response is linked to PTS. Possible mechanistic explanationsinclude: (1) the lupus anticoagulant (or a subclinical inflammatorystate that provokes the lupus anticoagulant) induces venous valvularinflammation that results in valvular fibrosis and insufficiency(an important contributor to PTS); and/or (2) the lupus anticoagulantconfers a hypofibrinolytic state that delays or prevents recanalizationin cases of veno-occlusive thrombosis, resulting in prolonged venoushypertension (also an important contributor to PTS). Future clinicalresearch, as well as animal studies, is warranted to further investigatethese proposed mechanistic hypotheses.
It is possible that the association between a lupus anticoagulant andPTS is dependent upon the presence of ß2-glycoprotein I-dependentantibodies, as described with initial thromboembolic events [14,15].However, ß2-glycoprotein antibody testing was performed in only aminority of subjects; thus an association between a lupus anticoagulantand ß2-glycoprotein antibodies could not be established in this study.The proposed mechanism for thrombogenesis involves the disruptionof the annexin “shield” above the phospholipid bi-layer through theactivity of anti-ß2-glycoprotein I antibodies. By compromising theannexin shield, procoagulant proteins are able to bind to the phospho-lipid surface and initiate coagulation [16,17]. de Laat and colleaguesdemonstrated that patients with ß2-gylcoprotein I-dependent lupusanticoagulant antibodies had evidence of increased coagulation activitythat correlated with a history of thrombosis that was not appreciated inpatients with ß2-glycoprotein I-independent lupus anticoagulants or inthe control population [18]. While these data support ß2-glycoproteinI-dependent lupus anticoagulants as a risk for initial thrombosis, the
results of the present study suggest these antibodies may also impairthrombus resolution, as the presence of a lupus anticoagulant(dRVVT) was associated with PTS.
Although lupus anticoagulant testing is a component of antiphos-pholipid antibody syndrome (APS) testing, this study sought to evaluatethe association of factors near the time of VTE diagnosis with the risk ofdeveloping PTS, not the association of APS and PTS. The ultimate goal isto identify very early in the course of VTE (i.e., acutely) those patientswho are at greatest risk of PTS, so that appropriate interventional strat-egies toward PTS prevention (e.g. thrombolysis, anti-inflammatoryagents, etc.) can be evaluated in the group at highest-risk. For thisreason, evaluation of APS (defined as persistence of the antibody for3 months) as a risk factor for PTS would have been a sub-optimalapproach toward the identification of acute factors that may be usedto inform future interventional strategies.
Strengths of the study include the bi-institutional composition of thecohort and the use of a standardized pediatric PTS instrument forwhichvalidation data have been published and training is available [7,19]. Atthe same time, several limitations are noteworthy. First, given the rath-er small sample size of the bi-institutional cohort, confidence intervalsaround the odds of clinically significant PTS for a positive dRVVT werewide, andwe could have failed to detect an association between certainfactors (e.g., FVIII activity) and development of PTS when such an asso-ciation in fact exists (i.e., type II error). The relatively small sample sizemay have also contributed to the lack of statistical significance betweenthe presence or absence of veno-occlusion at diagnosis and PTS, as wellas an association between duration of follow-up and identification ofPTS. Secondly, although comprehensive antiphospholipid antibodytesting at the time of diagnosis of VTE would have yielded additionalinformation regarding other antiphospholipid antibodies and the riskof PTS, the extent of such testing, and specific assays employed, arevariable across pediatric centers. Therefore from a data validity perspec-tive, we focused our analysis upon the lupus anticoagulant using thedRVVT method, which was systematically employed at both participat-ing sites. In addition, ESR, CRP, and IL-6 testing was not routinelyperformed, thus could not be evaluated. Future research should empha-size comprehensive evaluation of acute thrombophilia and inflammatorymarker testing. Third, ascertainment of PTS was limited by the presenceof staff trained in PTS assessment and attrition rates. At CHCO, the vastmajority of subjects adhered to long-term follow-up recommendationsvia the outpatient clinic, and a large group of clinical and research staffwere trained in standardized PTS assessment. As such, standardizedPTS assessment at≥1 year was widely available for cohort study analy-sis. At RCHSD, patient adherence to long-term follow-up was lower,and fewer staff were trained in standardized PTS assessment; hence,availability of standardized PTS assessment at≥1 year wasmore limitedfor cohort study analysis.
In conclusion, our study identifies an acute phase lupus anticoagu-lant as a prognostic factor for PTS in children diagnosed with limbDVT. This finding should be further evaluated in a larger collaborativecohort study in which thrombus characteristics are evaluated (occlu-sion status), comprehensive thrombophilia testing is performed,acquired factors (including antiphospholipid antibodies) are seriallyassessed, and global assay assessments of coagulative and fibrinolyticcapacity are explored. Through improved understanding of prognosticfactors for PTS following limb DVT in children, future risk-stratifiedtreatment strategies employing thrombolytic interventions and/oranti-inflammatory agents can be developed and tested in randomizedcontrolled pediatric clinical trials.
Conflict of Interest Statement
Dr. Goldenberg was supported in this work in part by a career de-velopment award from the National Institutes of Health, NationalHeart Lung and Blood Institute (K23HL084055). The authors declareno conflicts of interest.
41C.A. Lyle et al. / Thrombosis Research 131 (2013) 37–41
References
[1] Andrew M, David M, Adams M, Ali K, Anderson R, Barnard D, et al. Venous throm-boembolic complications (VTE) in children: first analyses of the Canadian Registryof VTE. Blood 1994;83:1251-7.
[2] van Ommen CH, Heijboer H, Buller HR, Hirasing RA, Heijmans HS, Peters M. Venousthromboembolism in childhood: a prospective two-year registry in TheNetherlands.J Pediatr 2001;139:676-81.
[3] Raffini L, Huang YS,Witmer C, Feudtner C. Dramatic increase in venous thromboem-bolism in children's hospitals in the United States from 2001 to 2007. Pediatrics2009;124:1001-8.
[4] Goldenberg NA, Knapp-Clevenger R, Manco-Johnson MJ. Elevated plasma factorVIII and D-dimer levels as predictors of poor outcomes of thrombosis in children.N Engl J Med 2004;351:1081-8.
[5] Goldenberg NA, Donadini MP, Kahn SR, Crowther M, Kenet G, Nowak-Gottl U, et al.Post thrombotic syndrome in children: a systematic review of frequency of occur-rence, validity of outcome measures, and prognostic factors. Haematologica Nov2010;95(11):1952-9.
[6] Kuhle S, Koloshuk B, Marzinotto V, Bauman M, Massicotte P, Andrew M, et al. Across-sectional study evaluating post-thrombotic syndrome in children. ThrombRes 2003;111:227-33.
[7] Goldenberg NA, Durham JD, Knapp-Clevenger R, Manco-Johnson MJ. A thrombo-lytic regimen for high-risk deep venous thrombosis may substantially reduce therisk of postthrombotic syndrome in children. Blood 2007;110:45-53.
[8] Goldenberg NA, Brandao L, Journeycake J, Kahn S, Monagle P, Revel-vilk S, et al.Perinatal And Paediatric Haemostasis Subcommittee Of The Scientific AndStandardization Committee Of The International Society On Thrombosis And H:Definition of post-thrombotic syndrome following lower extremity deep venousthrombosis and standardization of outcome measurement in pediatric clinicalinvestigations. J Thromb Haemost 2012;10:477-80.
[9] Triplett DA. Use of the dilute Russell viper venom time (dRVVT): its importanceand pitfalls. J Autoimmun, vol. 15. England: Academic Press; 2000. p. 173-8.
[10] Male C, Lechner K, Eichinger S, Kyrle PA, Kapiotis S,WankH, et al. Clinical significance oflupus anticoagulants in children. J Pediatr, vol. 134; 1999. p. 199-205. United States.
[11] Manco-Johnson MJ, Nuss R. Lupus anticoagulant in children with thrombosis. AmJ Hematol 1995;48:240-3.
[12] Roumen-Klappe EM, Janssen MC, Van Rossum J, Holewijn S, Van Bokhoven MM,Kaasjager K, et al. Inflammation in deep vein thrombosis and the developmentof post-thrombotic syndrome: a prospective study. J Thromb Haemost, vol. 7;2009. p. 582-7England.
[13] Shbaklo H, Holcroft CA, Kahn SR. Levels of inflammatory markers and the devel-opment of the post-thrombotic syndrome. Thromb Haemost, vol. 101; 2009.p. 505-12Germany.
[14] Zoghlami-Rintelen C, Vormittag R, Sailer T, Lehr S, Quehenberger P, Rumpold H,et al. The presence of IgG antibodies against beta2-glycoprotein I predicts therisk of thrombosis in patients with the lupus anticoagulant. J Thromb Haemost,vol. 3; 2005. p. 1160-5England.
[15] de Laat HB, Derksen RH, Urbanus RT, Roest M, de Groot PG. beta2-glycoproteinI-dependent lupus anticoagulant highly correlates with thrombosis in theantiphospholipid syndrome. Blood, vol. 104; 2004. p. 3598-602United States.
[16] Andree HA, Stuart MC, Hermens WT, Reutelingsperger CP, Hemker HC, FrederikPM, et al. Clustering of lipid-bound annexin Vmay explain its anticoagulant effect.J Biol Chem 1992;267:17907-12.
[17] Rand JH, Wu XX, Andree HA, Ross JB, Rusinova E, Gascon-Lema MG, et al.Antiphospholipid antibodies accelerate plasma coagulation by inhibitingannexin-V binding to phospholipids: a "lupus procoagulant" phenomenon. Blood1998;92:1652-60.
[18] de Laat B,Wu XX, van LummelM, Derksen RH, de Groot PG, Rand JH. Correlation be-tween antiphospholipid antibodies that recognize domain I of beta2-glycoprotein Iand a reduction in the anticoagulant activity of annexin A5. Blood, vol 109; 2007.p. 1490-4United States.
[19] Goldenberg NA, Pounder E, Knapp-Clevenger R, Manco-Johnson MJ. Validation ofupper extremity post-thrombotic syndrome outcome measurement in children.J Pediatr 2010;157:852-5.