Earl D Silverman

SickKids, Toronto, Ontario, Canada

Are you Earl D Silverman?

Claim your profile

Publications (293)1374.24 Total impact

  • [Show abstract] [Hide abstract]
    ABSTRACT: Systemic lupus erythematosus (SLE) is characterized by increased cardiovascular risk in adult-onset and childhood-onset SLE (cSLE). Type I interferons (IFNs) appear to play a prominent role in premature vascular damage in adult SLE, at least in part, by inducing impairments in the phenotype and function of endothelial progenitor cells (EPCs), thereby hampering vascular repair. It is not clear if EPC dysfunction is present in cSLE in association with a type I IFN signature. Phenotype and numbers of EPCs were quantified in patients with cSLE, juvenile idiopathic arthritis (JIA), and matched healthy controls (HC). In a separate cohort of cSLE subjects, markers of subclinical atherosclerosis and endothelial dysfunction were quantified using standardized protocols, and analyzed for associations with type I IFN serum activity. EPC numbers and function were significantly decreased in cSLE as compared to JIA and HC. cSLE serum impaired HC EPC differentiation into mature endothelial cells, an effect blocked by type I IFN pathway inhibition. Type I IFN serum activity was not significantly associated with subclinical atherosclerosis and endothelial function in cSLE. As in adults, cSLE is characterized by phenotypic and functional EPC abnormalities, likely triggered by type I IFNs. While cross-sectional analysis detected no global association between type I IFN signatures and vascular measures of subclinical atherosclerosis, longitudinal assessments are needed to evaluate if progression of vascular damage in cSLE is associated with type I IFNs as in the adult population. This article is protected by copyright. All rights reserved. © 2015 American College of Rheumatology.
    04/2015; DOI:10.1002/art.39149
  • [Show abstract] [Hide abstract]
    ABSTRACT: Cytomegalovirus (CMV) is a beta-herpesvirus and antibodies to this virus are common in patients with systemic lupus erythematosus (SLE). However, few studies have examined the relationship between CMV infection and SLE. Our objectives were: 1) to determine the prevalence of CMV infection at the time of SLE diagnosis, and 2) to determine the risk factors for CMV infection. A database review of 670 patients with pediatric SLE (pSLE) seen over a 20-year period identified seven patients with a CMV infection detected at the time of diagnosis of SLE. CMV was diagnosed by serology, urine and bronchoalveolar lavage. Clinical manifestations, laboratory findings, virology studies and treatments were reviewed. CMV infection was detected in seven patients at the time of SLE diagnosis (1.04% of total cohort): six were female: mean age was 13 years. Predominant features included non-Caucasian ethnicity (p < 0.01 as compared to total SLE cohort), persistent fevers on prednisone in seven and nephrotic syndrome in four. Laboratory findings included: anemia in seven, lymphopenia in five, elevated liver enzymes in four, with anti-dsDNA and anti-RNP antibodies present in six and five, respectively. Six patients received ganciclovir and CMV hyperimmune globulin (Cytogam®) with the continuation of prednisone during CMV treatment. Six of seven fully recovered without sequelae (one without treatment) but one patient died with active CMV infection. There were 1.04% of patients with pSLE who developed CMV infection. All were of non-Caucasian ethnicity. Persistent fever despite prednisone, with concomitant anemia, may be additional clues to CMV infection in pSLE. We suggest all patients have routine testing for CMV immunity at initial presentation of pSLE. © The Author(s) 2015 Reprints and permissions: sagepub.co.uk/journalsPermissions.nav.
    Lupus 01/2015; DOI:10.1177/0961203314565443 · 2.48 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: Objective: To evaluate the influence of ethnicity on self-reported health related quality of life (HRQOL) in the Canadian childhood-onset systemic lupus Erythematosus (cSLE) population.Methods: Patients with cSLE at 4 pediatric centers were consecutively enrolled. Sociodemographics and multiple disease activity measures were collected. The Child Health Questionnaire (CHQ) was administered and analysed by ethnicity.Results: We enrolled 213 cSLE patients, and complete data from 196 patients with these ethnicities were analysed: White (33%), Asian (32%), South Asian (16%), Black (11%), Latino/Hispanic (5%), and Aboriginal (4%). Compared to healthy children, cSLE patients rated their HRQOL significantly more poorly in 9 of 10 individual domains, and in 4 of 10 domains when compared to a cohort of juvenile arthritis. Within the cSLE cohort, CHQ scores were lower in 5 of 10 domains in white patients versus nonwhite ethnicities (p<0.05 for each). Physical summary scores (PhS) were lower for white patients compared to the other ethnicities aggregated together (46.0 ± 11.9 vs. 50.4 ± 10.1, p = 0.009); however, psychosocial summary scores (PsS) were similar among the groups (40.5 ± 14.6 vs. 42.8 ± 12.7, p = 0.26). Disease activity measures, including Systemic Lupus Disease Activity Index, Systemic Lupus Activity Measure and physician global visual analogue scale were similar across ethnicities. However, patient reported Systemic Lupus Activity Questionnaire symptom scores were greater in patients of white ethnicity compared to those of Asian ethnicity (8.2 ± 5.8 vs. 4.5 ± 4.7, p=0.004).Conclusion: The self- and parent-reported health status of Canadian cSLE patients differed across ethnicities, with white patients reporting lower HRQOL, despite similar and overall low disease activity. © 2014 American College of Rheumatology.
    12/2014; 66(12). DOI:10.1002/acr.22363
  • [Show abstract] [Hide abstract]
    ABSTRACT: Objective: Prognosis studies examine outcomes and/or seek to identify predictors or factors associated with outcomes. Many prognostic factors have been identified in Systemic Lupus Erythematosus (SLE) but few have been consistently found across studies. We hypothesized that this is due to lack of rigor of study designs. This study aimed to systematically assess the methodological quality of prognosis studies in SLE.Methods: A search of prognosis studies in SLE was performed using MEDLINE and EMBASE, from January 1990 to June 2011. A representative sample of 150 articles was selected using a random number generator and assessed by 2 reviewers. Each study was assessed by a risk-of-bias tool according to 6 domains: study participation, study attrition, measurement of prognostic factors, measurement of outcomes, measurement/adjustment for confounders and appropriateness of statistical analysis. Information about missing data was also collected.Results: A cohort design was used in 71% of studies. High risk of bias was found in 65% for confounders, 57% of studies for study participation, 56% for attrition, 36% for statistical analyses, 20% for prognostic factors and 18% for outcome. Missing covariate or outcome information was present in half of the studies. Only 6 studies discussed reasons for missing data and 2 imputed missing data.Conclusions: Lack of rigorous study design— especially in addressing confounding, study participation and attrition, and inadequately handled missing data— has limited the quality of prognosis studies in SLE. Future prognosis studies should be designed with consideration of these factors to improve methodological rigor. © 2014 American College of Rheumatology.
    10/2014; 66(10). DOI:10.1002/acr.22322
  • [Show abstract] [Hide abstract]
    ABSTRACT: Objective. To evaluate and determine the most effective immunosuppressive therapy for the induction treatment of proliferative lupus nephritis (PLN) based on renal remission. Methods. A systematic review of randomized controlled trials was conducted. The outcomes were renal remission at 6 months: (1) normalization of serum creatinine [(sCr), or within 15% of the normal range, i.e., sCr < 132 mu mol/l - creatinine remission]; and (2) proteinuric remission (prU < 0.5 g/day/1.73m(2)). A Bayesian network metaanalysis was used. Results. The OR (95% credible interval) of inducing an sCr remission at 6 months was 1.70 (0.51, 6.87) for mycophenolate mofetil (MMF) versus cyclophosphamide (CYC); 2.16 (0.38, 13.36) for tacrolimus (Tac) versus CYC; and 1.25 (0.13, 10.51) for Tac versus MMF. For proteinuric remission the OR was 1.46 (0.81, 3.04) for MMF versus CYC; 1.96 (0.80, 5.11) for Tac versus CYC; and 1.34 (0.43, 3.90) for Tac versus MMF. The probability (95% credible interval) of inducing a creatinine remission at 6 months was Tac 56% (19%, 88%); MMF 51% (23%, 79%); and CYC 37% (28%, 47%). The probability of inducing a proteinuric remission was Tac 41% (23%, 63%); MMF 34% (23%, 50%); CYC 26% (20%, 32%); azathioprine 10% (1%, 55%); prednisone 11% (2%, 38%). None of the results were conclusive when examined in a sensitivity analysis. Conclusion. There is currently insufficient evidence to determine which of these immunosuppressive agents is superior. The probability of renal remission is 50% or lower at 6 months.
    The Journal of Rheumatology 09/2014; 41(10). DOI:10.3899/jrheum.140050 · 3.17 Impact Factor
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: Key Clinical Message Systemic lupus erythematosus should be included in the differential diagnosis of every adolescent with pancytopenia. An accurate diagnosis with the appropriate therapy is vital and can cause lasting reversal of this condition. Systemic lupus erythematosus should be included in the differential diagnosis of every adolescent with pancytopenia. An accurate diagnosis with the appropriate therapy is vital and can cause lasting reversal of this condition.
    09/2014; 2(6). DOI:10.1002/ccr3.123
  • [Show abstract] [Hide abstract]
    ABSTRACT: Objective: To determine the impact of prenatal exposure to maternal anti-Ro antibodies, slow fetal heart rate and/or chronic dexamethasone therapy for immune-mediated congenital atrio-ventricular heart block (CAVB) on cognitive and academic performance at school age.Methods: Prospective, blinded assessment of cognitive functioning of 3 cohorts of children aged 6-16 years with in-utero exposure to maternal anti-Ro antibodies and with: 1) no CAVB and no prenatal dexamethasone (n=14); 2) CAVB without prenatal treatment (n=10); and 3) CAVB with prenatal dexamethasone therapy (n=16). Domains assessed included intelligence, visual perceptual and visual motor skills, auditory and visual attention, verbal learning and memory, visual memory, executive function and behavior.Results: All cohorts scored within the normal range and were not significantly different in intelligence scores, verbal comprehension, perceptional reasoning, working memory, and processing speed. For prenatally treated children with CAVB, there were no significant associations between neurocognitive function scores, minimal fetal heart rate (range: 47 – 80 beats/min) and the duration and dosage (2-15 weeks; 56 – 824 mg) of dexamethasone therapy.Conclusion: CAVB and transplacental prenatal treatment with dexamethasone was not associated with neurocognitive impairment in school-age children. Larger numbers are required to validate our observation and assessment of other cognitive abilities is warranted. © 2014 American College of Rheumatology.
    08/2014; 66(8). DOI:10.1002/art.38675
  • [Show abstract] [Hide abstract]
    ABSTRACT: Studies of paediatric patients with membranous lupus nephritis (MLN) have yielded variable results, mostly due to the inclusion of mixed, i.e. proliferative nephritis. The aim of this study was to describe clinical and laboratory findings at the diagnosis of paediatric non-proliferative MLN, report the outcome and identify predictors of remission.
    Pediatric Nephrology 07/2014; 30(1). DOI:10.1007/s00467-014-2908-2 · 2.88 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: Background/Purpose:Biologic therapies have revolutionized the management of rheumatic diseases of childhood. However, these medications are associated with adverse effects including the possible development of neoplasms. The aim of our study was to determine the rate as well as risk factors for the development of neoplasms in patients with JIA treated with biologics.Methods:We performed a retrospective review of the rheumatology biologic registry (RBR) at The Hospital for Sick Children (SickKids), Toronto, Canada, one of the largest quarternary pediatric referral centres in North America. Demographic and neoplastic data, clinical course and medication history were extracted from the RBR and clinical charts. Unless specified, data was expressed as median (IQR). The study was approved by the SickKids Ethics Review Board.Results:The cohort consisted of 357 patients with rheumatic diseases who were on one or more biologics between January 1997 and August 2013. Juvenile Idiopathic Arthritis (JIA) was the most common diagnosis [302/357 (84.5%)]. A total of 6/357 (1.68%) patients developed a neoplasm: 4 with JIA, 1 each with idiopathic uveitis and polyarteritis nodosa. (Refer to t87) 1Table 1. Patient 1Patient 2Patient 3Patient 4Patient 5Patient 6NeoplasmNasopharyngeal carcinoma1. Tonsillar lymphoproliferative disorder (LD) in 2006Clear cell renal carcinomaHepatosplenic lymphomaSmall blue cell sarcoma, undifferentiatedPilomatricoma 2. Renal carcinoma in 2011 Rheumatic Disease (RD)Idiopathic UveitisPolyarteritis NodosaPoly JIA (RF negative)Systemic JIAOligo JIA (extended)Oligo JIA (extended)Age of onset of RD (years), Cytotoxic agentsMTX (4.7)MTX (1.2);MTX (6.2);MTX (3.6);MTX (7.2);CSA (3.2);(Time (years)) AZA (NA);LFN(0.2);CSA (0.9);LFN (0.2)AZA (4.3);Abbreviations: -Methotrexate (MTX), CYC (1.5)CSA (1.5);Tacrolimus (5.0); MTX (6.0);-Azathioprine (AZA), AZA (1.2);Etoposide (6 cycles over 34 days); LFN (2.8)-Leflunomide (LFN), SSA (0.9)Thalidomide (0.6) Cyclophosphamide (CYC), -Cyclosporine (CSA), -Sulfasalazine (SSA) -Not available (NA) Biologic(s)Infliximab (3.3)Infliximab (7.5)Etanercept (4.8)Etanercept (0.1);Etanercept (5.2)Infliximab (8.2);(Time (years)) Infliximab (1.7); Etanercept (1.5) Anakinra (1.9) Time to neoplasm (years)5.31. 14.115.811.816.712.7 2. 18.6 Time from DMARD to neoplasm (years)4.71. 6.817.611.88.412.6 2. 11.4 Time from Biologic to neoplasm (years)3.31. 4.810. 2. 9.3 Family history of neoplasmMaternal grandfather: lung cancer; Maternal grandmother: cervical cancerMother died due to breast cancer.NANANAStrong family history of colon cancerAmongst patients with neoplasms, the median (IQR) age at rheumatic disease onset was 3.4 (1.7–7.3) years and the age of diagnosis of the neoplasm was 16.3 (15.3–17.9) years. All cancers were malignant except in patient 6 which was of a benign nature. On average, patients had exposure to 3 DMARDs (range 1–5), with methotrexate as the commonest DMARD used in 5/6 patients prior to diagnosis of the neoplasm. The time from the disease onset to initiation of DMARD was 3.9 (0.4–7.9) years. The time from the DMARD initiation and the diagnosis of a neoplasm was 10.1 (7.2–12.3) years. For methotrexate alone, the duration of use was 5.4 (3.9–6.1) years. All patients diagnosed with a neoplasm, had received both, DMARD(s) and biologic(s). 2 patients were on infliximab alone, 2 on etanercept alone and 2 on multiple biologics of which etanercept and infliximab were both used. The time between onset of a biologic and diagnosis of a neoplasm was 5.0 (3.6–7.7) years. Two patients (2 and 4) also received medications which are associated with the development of the neoplasms (cyclophosphamide or etoposide).Conclusion:Majority of neoplasms were of uncommon morphology and site as well as aggressive in nature. The neoplasms developed late after drug exposure with a median duration to diagnosis of neoplasm of 10.1 years on a DMARD and 5.5 years on a biologics. Whilst DMARDs and biologics are effective for the management of rheumatic diseases, patients with refractory disease requiring ongoing drug therapies should have malignancy/ neoplasm surveillance included as part of routine clinical care. Comparison of patients on biologics with and without neoplasms is in progress.
    03/2014; 66(S11). DOI:10.1002/art.38596
  • [Show abstract] [Hide abstract]
    ABSTRACT: Background/Purpose:To assess efficacy and safety of SC golimumab (GLM) in polyarticular pediatric juvenile idiopathic arthritis pts (aged 2 to <18 yrs) with active arthritis despite MTX for ≥3 months.Methods:In GO-KIDS, a 3-part randomized double-blind, PBO-controlled, withdrawal trial in pts with active JIA with a polyarticular course (≥5 active joints) and disease duration of ≥6 months despite current MTX (10–30 mg/m2/wk). In Part 1(wk 0–12), all pts received open-label (OL) 30 mg/m2 GLM SC (max 50 mg) q4 wks with stable MTX dose. At wk 16, pts with ACR JIA 30 response entered Part 2 (wk 16–48). In Part 2, pts were randomized to continue GLM or switch to PBO q4 wks. Upon Part 2 completion at wk 48 or ACR JIA flare in Part 2, pts received OL GLM in Part 3. Primary endpoint was proportion of ACR JIA 30 responders at wk 16 without a JIA flare in Part 2 using wk 16 as baseline measurement. Secondary outcomes were ACR JIA 30/50/70/90 response rates and inactive disease rates at wk 16 and wk 48 by group and safety.Results:173 pts (Caucasian 87.9%, 75.7% females; age [median/range] 12 yrs/2–17 yrs) with moderately active disease were enrolled (Table 1); 19 (11%) were d/c in Part 1 (lack of efficacy n = 14, AE n = 4, withdrawal of consent n = 1). In Part 1, 151 of 173 (87.3%) achieved an ACR JIA 30 response and 36.1% inactive disease status (Table 2). There were 154 pts randomized double-blind in Part 2 (PBO n = 76; continued GLM n = 78). At the end of Part 2, no significant differences between groups in JIA ACR non-flare rates and trial did not meet primary endpoint [PBO-grp vs. GLM-grp: 52.6% vs. 59.0%, p = 0.41) nor were there differences for major secondary endpoints (Table 2). In contrast, sustained response in both groups (PBO, GLM) relative to wk 0 (Table 1 and 2). 6.5% (10 of 154) of pts d/c therapy after wk16 through wk 48. Through wk 48, AEs, serious AE (SAE), and serious infections were reported in 87.9%, 13.3%, and 2.9% of all randomized pts, respectively. Most common SAE was JIA exacerbation. No deaths, active TB or malignancies were reported (1). GLM-treated pt experienced SAE of toxic hepatitis and 1 had SAE of hepatic enzyme increase. One pt switched from PBO to GLM experienced SAE of serum sickness reaction, resulting in GLM d/c. Proportion of pts with ≥1 injection site reaction was 8.1%. None of the reactions were serious, severe, or led to GLM d/c.Table 1. At Baseline (All enrolled patients n=173)At wk16 (All randomized patients prior to randomization, n=154)At wk48Values are medians (interquartile range)PBO + MTX (n=76)GLM + MTX (n=78) Physicians global assessment5.40 (3.90; 7.00)0.50 (0.10; 1.30)0.30 (0.00; 1.00)0.30 (0.00; 1.30)Patient/parent global assessment4.50 (2.80; 6.10)0.90 (0.30, 2.30)0.60 (0.20; 1.65)0.45 (0.10; 1.70)Number of active joints12.0 (8.0; 18.0)1.0 (0.0; 3.0)1.0 (0.0; 3.0)0.0 (0.0; 2.0)Number of joints with limited range of motion8.0 (6.0; 15.0)1.0 (0.0, 4.0)0.5 (0.00; 4.0)1.0 (0.0; 3.0)Physical function by CHAQ0.94 (0.38; 1.50)0.25 (0.00; 0.75)0.13 (0.00; 0.63)0.00 (0.00; 0.63)ESR (mm/hr)16.00 (8.00; 28.00)9.00 (5.00; 19.00)9.50 (5.00; 16.50)10.00 (5.00; 19.00)Methotrexate (mg/wk)15t(5.00; 30.00)15t(5.00; 30.00)15t(5.00; 30.00)15t(5.00; 30.00)Table 2. ACR JIA response and flare ratesPART 1 [WK 0ndash;16] Pts experiencing flares/those considered treatment failures due to protocol violations are considered ACR non-responders from the flair/failure visit to wk 48aObserved dataPercentage of ACR JIA responders at end of Part 1 [WK 16] (n = 173)JIA ACR 3087.3%t(151) JIA ACR 5079.2%t(137) JIA ACR 7065.9%t(114) JIA ACR 9036.4%t(63) Inactive disease36.1%t(62) PART 2 [WK 16ndash;48] Percentage of ACR JIA 30 responders without flare in Part 2 (n = 154)PBO + MTX (n=76)52.6%t(40)P=0.41 GLM + MTXt(n=78)59.0%t(46) Inactive disease21.6%/39.7% Clinical remission11.8%/12.8% Percentage of ACR JIA responders at WK48 (vs. wk0) by treatment in Part 2‡ [PBO +MTX/GLM +MTX]JIA ACR 3095.9%/89.0% JIA ACR 5091.8%/86.3% JIA ACR 7078.1%/78.1% JIA ACR 9053.4%/56.2% Conclusion:JIA pts with active polyarticular dis ease demonstrated rapid response to GLM with 16 wks of OL treatment, resulting in inactive disease in 36% of the pts. The lack of differences in flare rates between GLM and PBO arms from wk 16 to 48 among OL GLM responders needs further evaluation. Safety profile was acceptable and injections well tolerated.
    03/2014; 66(S11). DOI:10.1002/art.38569
  • [Show abstract] [Hide abstract]
    ABSTRACT: Background/Purpose:Little is known about the long-term morbidity and mortality of childhood-onset SLE (cSLE) after transition to adult care; however, linking clinical data to administrative databases enables study of previously unknown outcomes. Our objectives were to describe long-term outcomes in young adults with cSLE, including: i) cardiovascular outcomes; ii) renal outcomes; iii) joint replacement surgery due to avascular necrosis; iv) malignancies; and v) pregnancies and births.Methods:A retrospective chart review of all cSLE patients (<18 years at diagnosis) diagnosed between Jan 1, 1984 and Dec 31, 2011 and followed for ≥1 year was conducted after contacting all pediatric and adult rheumatologists and nephrologists practicing in Ontario. Clinical data and Ontario Health Insurance Plan (OHIP) numbers were securely transferred to the Institute for Clinical Evaluative Sciences (ICES). OHIP numbers were transformed into an encrypted ICES key number (IKN) used to link the cohort to multiple administrative datasets to determine the outcomes of interest. Outcomes were determined using validated definitions. Means ± standard deviations are given unless noted.Results:622 cSLE patients are included in this inception cohort. The cohort was predominantly female (81%) with mean follow-up of 10.7 ± 7.3 years, representing 6629 person-years of disease. Age at diagnosis was 12.8 ± 3.2 years. There were 34 (6%) hospitalizations for cerebrovascular events, at mean age 18.1 ± 7.7 years, and disease duration 6.7 ± 7.0 years. Fewer than 6 hospitalizations for myocardial infarctions occurred at a median disease duration of 10.7 years. 273 (44%) patients had biopsy-proven lupus nephritis, with end stage renal disease occurring in 20 (7.3%) of these patients at mean age 31.4 ± 8.0 years, and disease duration 17.5 ± 8.7 years. Eight patients underwent renal transplant. Twenty-two (3.5%) patients have had 32 hip replacement surgeries, at mean age 24.2 ± 6.6 years and disease duration 10.1 ± 6.7 years. Fewer than 6 patients have had knee and/or ankle replacement surgery. Fourteen malignancies occurred in 13 (2%) patients, at mean age 28.2 ± 9.2 years and disease duration 16.5 ± 7.2 years. Nine were female reproductive and/or hematologic malignancies. Obstetrical outcomes were examined in females ages 15–45 years. Of 222 pregnancies, there were 82 (37%) live births, occurring at mean age 26.4 ± 4.9 years and disease duration 12.1 ± 5.5 years. Fewer than 6 pregnancies resulted in twins. There were 90 (41%) therapeutic abortions, 39 (18%) miscarriages, and 8 (4%) still births.Conclusion:Myocardial infarction and end-stage renal failure rates in this cohort with universal healthcare appear to be lower than reported in other populations. Cerebrovascular disease, hip replacement surgery and malignancies were significant morbidities in this cohort of young adults. Many women were able to conceive and deliver live infants; however, the miscarriage and stillbirth rates were greater than expected for young women.
    03/2014; 66(S11). DOI:10.1002/art.38579
  • [Show abstract] [Hide abstract]
    ABSTRACT: Background/Purpose:It had been suggested that prenatal exposure to hydroxychloroquine reduced the risk of cardiac NLE. The primary aim was to assess if prenatal exposure to antimalarials (AM) decreased the risk of cardiac NLE. The secondary aim was to analyze the effect of AM exposure on the risk of non-cardiac NLE.Methods:A retrospective cohort study was performed on a large single-center cohort of children exposed to anti-Ro and/or anti-La antibodies on whom prospective data has been collected since 1984. Inclusion criteria were: 1) first child born from a woman positive for anti-Ro and/or anti-La antibodies with a diagnosis of cutaneous lupus, systemic lupus erythematosus, Sjogren's syndrome, dermatomyositis or rheumatoid arthritis, 2) the mother underwent fetal chocardiography screening during pregnancy and/or the child had a postnatal ECG and 3) the child was ≥6 months old as of October 2013. We used Bayesian analysis for the association between prenatal use of AM and NLE.Results:The study population consisted of 265 children, of whom 72 were exposed to AM throughout gestation. Full laboratory data was available on 216 infants: 101 (46.8%) developed NLE and 115 (53.2%) remained unaffected. Forty nine children were classified as having no cardiac NLE but could not be diagnosed as true unaffected children as one or more blood test components were missing. These children were only included when the outcome studied was cardiac NLE. On univariable analysis and under a non-informative prior, the probability that prenatal AM exposure would be protective (RR < 1) was 97.1% for cardiac and 66.9% for non-cardiac NLE. On multivariable analysis, the RRs obtained were numerically higher, but still suggestive of a protective effect (1). Using a more stringent RR cutoff (RR < 0.75), the effect on the outcome cardiac NLE remained significant, unlike for non-cardiac NLE for which probabilities dropped significantly.Table 1. Bayesian Analyses of the Effect of Prenatal Exposure to Antimalarial on the Risk of NLEOutcomePriorSample size (N)RR (95% CrI)Probability of RR <1 (%)Probability of RR <0.75 (%)aIndependent variable: prenatal exposure to AMbIndependent variables: prenatal exposure to AM, anti-Ro ≥50 U/mL, anti-La ≥50 U/mL and maternal diagnosis (Sjogren's syndrome vs all other diagnosis)cNLE: neonatal lupus; RR (95% CrI): relative risk (95% credible interval)Univariable analyses† Cardiac NLENon-informative2650.17 (0.01, 1.05)97.193.8Cardiac NLEInformative2650.34 (0.17, 0.66)99.998.9Non-cardiac NLENon-informative2140.93 (0.64, 1.29)66.912.7Multivariable analyses‡ Cardiac NLENon-informative2300.31 (0.01, 1.91)87.880.0Cardiac NLEInformative2300.36 (0.18, 0.72)99.898.1Non-cardiac NLENon-informative1900.94 (0.65, 1.31)64.011.6Conclusion:In this study of the largest single-center cohort of children born to anti-Ro and/or anti-La antibody positive women with a connective tissue disease, we have shown that the probability that AM exposure was associated with a decreased risk of cardiac NLE was >87%. A clinically significant beneficial effect from AM exposure on non-cardiac NLE features was not present. These findings need to be confirmed in an independent cohort.
    03/2014; 66(S11). DOI:10.1002/art.38570
  • [Show abstract] [Hide abstract]
    ABSTRACT: Background/Purpose:Neonatal lupus erythematosus (NLE) is a passively transferred autoimmune disease that occurs in babies born to mothers with anti-Ro and anti-La antibodies. The most serious complication of NLE is congenital heart block (CHB). In pregnancies of mothers with a known autoimmune condition and positive anti-Ro antibodies, the incidence of heart block is approximately 1-2% of live births. We have previously shown that only mothers with moderate-high titre antibodies are at risk to deliver a child with CHB. However, the rate of anti-Ro positive antibody pregnant women in an otherwise healthy population is unknown or is their risk for delivering a child with CHB.Objectives: Determine the rate of anti-Ro/La antibodies in the general pregnant population. Determine if the incidence of CHB is increased in healthy mothers with positive Ro/La antibodies compared to mothers with known autoimmune disease and positive anti-Ro/La antibodies.Methods:Antibody testing was performed on 15198 pregnant women who were having concurrent Maternal Serum Screening in the Metropolitan Toronto area. Maternal self-reported outcomes of prenatal, pregnancy, and post-natal medical conditions were reported, along with fetal outcomes of pre and post-natal illnesses. Autoantibody titres were stratified into negative, low, moderate, and high positive.Results:1152/151598 (7.6%) of the mothers who had anti-Ro antibodies and 179/15198 (1.18%) had moderate-high titres (at risk to deliver a child with CHB). 779 (5%) had anti-La antibodies with the majority being low titre. During the course of the study there were 13 cases of CHB that were unrelated to our maternal sample population- 10 to well mothers and 3 to mothers with an autoimmune disease. All of these women mothers had moderate-high titre anti-Ro antibodies, while only 31% had moderate-high titre anti-La antibodies. During the course of the study 64 pregnant women with a known autoimmune disease and anti-Ro antibodies (at risk to deliver a child with CHB) were prospectively followed. 3/64 delivered a child with CHB. All 3 of these mothers had moderate-high titre anti-Ro antibodies while 41/61 mothers who delivered a child without CHB had moderate-high titre anti-Ro antibodies. Therefore 3/44 (6.9%) mothers with moderate-high titre anti-Ro antibodies and an autoimmune disease delivered a child with CHB.Conclusion:The incidence of CHB is reported to be between 1:10/N15,000 pregnancies. Therefore, based on our data showing 1.18% of otherwise well pregnant woman had moderate-high titre anti-Ro antibodies (at risk to deliver a child with CHB), for each child with CHB we predict that 117—174 children without CHB will be delivered to otherwise healthy mothers. In contrast, in mothers with a known autoimmune disease and moderate-high anti-Ro antibody titre, we found a 6.9% incidence of CHB and therefore for each child with CHB there were 14 children without CHB born. Therefore the risk for a woman with a known autoimmune disease and moderate-high titre anti-Ro antibodies was approximately 10x that of otherwise healthy pregnant women. These data therefore suggest that the anti-Ro antibody repertoire differs between these 2 groups of pregnant women.
    03/2014; 66(S11). DOI:10.1002/art.38458
  • [Show abstract] [Hide abstract]
    ABSTRACT: Background/Purpose:Childhood-onset systemic lupus erythematous (cSLE) has a more severe clinical course with higher disease activity and involvement of major organs compared to adult-onset SLE. Avascular necrosis (AVN) is a significant morbidity causing pain and disability, sometimes requiring joint replacement surgery. Our objective was to examine the frequency and risk factors for symptomatic AVN in cSLE.Methods:A single center matched case-control design was used. 617 patients with cSLE followed at SickKids Lupus Clinic between July 1982 and June 2013 were included in the study. The AVN cohort consisted of 37 patients identified with clinical findings of symptomatic AVN and diagnosis confirmed by one or more imaging modalities. Three controls were matched to each AVN patient by date and age at diagnosis (± 1 year). Baseline clinical, laboratory and treatment characteristics were compared between the patients with AVN and the controls by univariate analyses and if statistically significant, were included in a multivariable logistic regression model.Results:A total of 37/617 (6%) patients developed symptomatic AVN during follow-up at SickKids. The majority was female (30/37, 81%) and of Asian descent (20/37, 54%). The mean age at diagnosis was 16.1 years (±2.1), with median time to AVN of 1.52 years. Only 2/37 (5%) of patients developed AVN prior to the onset of puberty. The hip was the most commonly involved joint (26/37, 70%) with bilateral involvement in 49% (18/37) of patients. Compared to the matched non-AVN cohort, patients with AVN had higher incidence of CNS disease (p = 0.003), nephritis (p < 0.001), acute or chronic renal failure (p = 0.029), but less frequent photosensitivity (p = 0.006). Patients with AVN also required greater cumulative prednisone from cSLE diagnosis to AVN (364 ± 53 vs. 232 ± 36 mg/kg, p < 0.001), higher prednisone dose at time of AVN diagnosis (±3 months, 0.30 ± 0.25 vs. 0.19 ± 0.24 mg/kg, p = 0.012), maximal daily prednisone dose (1.25 ± 0.36 vs. 0.71 ± 0.53 mg/kg, p < 0.001) and more frequent use of pulse methylprednisolone therapy (39% vs. 10%, p < 0.001). The median prednisone dose at time of AVN was 0.27 vs. 0.09 mg/kg. Multivariable regression analysis confirmed nephritis, CNS disease, maximal daily prednisone dose and use of pulse methylprednisolone as significant predictors of symptomatic AVN development. Overall disease activity from SLE diagnosis to AVN diagnosis as measured by adjusted mean SLEDAI was not significantly different (6310 ± 976 vs. 4994 ± 440, p = 0.165).Conclusion:cSLE patients with severe organ involvement including nephritis and CNS disease, higher maximal daily dose of prednisone, and more frequent use of pulse methylprednisolone are more likely to develop symptomatic AVN.
    03/2014; 66(S11). DOI:10.1002/art.38434
  • [Show abstract] [Hide abstract]
    ABSTRACT: Background/Purpose:1) To determine the longitudinal disease activity trajectory of an inception cohort of Juvenile Dermatomyositis (JDM) patients, 2) To identify predictor(s) for longitudinal disease activity trajectory of JDM.Methods:A single-center, inception cohort of juvenile dermatomyositis (JDM) patients (age of diagnosis <18 years old), recruited within 1 year of their initial presentation was studied. The patients were accrued from January 1991 to December 2010, in a JDM subspecialty clinic. Patients were evaluated at every clinic visit for disease activity with the modified Disease Activity Score (mDAS). Baseline disease features–DASm, swallowing difficulty, skin ulcer, calcinosis, duration of symptoms before evaluation, age and gender–were evaluated as predictors of the longitudinal trajectory of JDM. Time-varying treatment effects were also tested. Longitudinal trajectory modeling was performed with mixed random effects modeling (random slopes).Results:Ninety-five JDM patients (33 males, 35%) were studied. The median age of onset was 7.8 (25th–75th percentile (P): 4.9–12.1) years. The median mDAS score at presentation was 7 (25th–75th P: 6–9), the median skin subscale score was 3 (25th–75th P: 2–4), the median musculoskeletal subscale was 4 (25th–75th P: 3–6). The median duration of follow-up was 4.6 (25th–75th P: 2.4–6.9) years. All patients in the cohort were treated with a standardized protocol comprising prednisone alone (before 2000) or methotrexate (MTX) and prednisone (after 2000). The disease trajectory of mDAS in this inception cohort followed a rapid reduction of activity within the first 2 years with a minor flare in disease activity beyond 2 years. Baseline nailfold abnormality predicted a slower improvement in mDAS trajectory. Higher baseline mDAS predicted faster improvement in mDAS trajectory. Treatment with MTX and prednisone predicted reductions in disease activity 3 and 6 months later respectively.Conclusion:Disease activity trajectory of JDM showed a general rapid reduction of mDAS followed by a small flare of disease activity beyond 2 years. Baseline nailfold abnormality and mDAS predicted future evolution of disease trajectory. The effects of treatment were observed only 3–6 months after.
    03/2014; 66(S11). DOI:10.1002/art.38442
  • Mohammed Olfat, Earl D. Silverman, Deborah M. Levy
    [Show abstract] [Hide abstract]
    ABSTRACT: Autoimmune thrombocytopenia (AITP) and hemolytic anemia (AIHA) are common in childhood-onset systemic lupus erythematosus (cSLE) and may be refractory to conventional therapies. Our objectives were to: (a) examine our experience; (b) determine the rate and durability of response to rituximab; and (c) evaluate its safety in our cSLE population with refractory cytopenias. We performed a single-center retrospective cohort study of cSLE patients with refractory AITP or AIHA treated with rituximab between 2003 and 2012. Outcomes included the time to complete clinical response, time to B-cell depletion, duration of response and time to flare. Adverse events were also analyzed. Twenty-four (6%) of 394 cSLE patients received rituximab for refractory cytopenia. The indication was AITP in 16 (67%), AIHA in five (21%) and both in three (13%) patients. The median (interquartile range (IQR)) time from cytopenia onset to rituximab therapy was 16 (7-27) months for AITP and 10 (2-29) months for AIHA. Complete response following the first course of rituximab occurred at a median (IQR) of 48 (14-103) days, only one patient failed to respond. Five (21%) patients had one or more flare episodes at 22 (15-27) months. Infusion reactions were rare and one infection with herpes zoster required hospitalization in the first 12 months. Three of four patients with low IgG levels prior to the first rituximab course developed persistent hypogammaglobulinemia, and three patients have required intravenous immunoglobulin replacement. Rituximab appears to be a well-tolerated, safe and long-lasting therapy for cSLE patients with refractory AITP and/or AIHA. Caution should be exercised when considering rituximab for patients with preexisting hypogammaglobulinemia. © The Author(s) 2015 Reprints and permissions: sagepub.co.uk/journalsPermissions.nav.
    03/2014; 66(S11). DOI:10.1002/art.38435
  • [Show abstract] [Hide abstract]
    ABSTRACT: Background/Purpose:Macrophage activation syndrome (MAS), a life-threatening inflammatory emergency of children with rheumatic diseases, is increasingly recognized in pediatric systemic lupus erythematosus (pSLE). The challenge is to differentiate active pSLE from MAS in order to make correct treatment decisions. The purpose of this study is to generate a decision tree for the recognition of MAS in newly diagnosed pSLE and test the performance of these proposed criteria in an independent pSLE cohort.Methods:A retrospective cohort study of consecutive patients requiring admission to SickKids Hospital with newly diagnosed, active pSLE between January 2002 and July 2007 (training cohort) was performed. All patients met ≥4/11 ACR criteria. Data collection on: 1) Clinical features including fever, CNS dysfunction, splenomegaly, hepatomegaly and hemorrhage; 2) laboratory parameters; CBC, ESR, CRP, C3, C4, ferritin, AST, ALT, LDH, albumin, bilirubin, triglycerides, LDL, HDL, urea, creatinine, sodium, coagulation parameters including INR, PTT, fibrinogen and D-Dimer, and soluble IL-2 receptor (sIL-2R) and CD163. Patients were assigned to one of 2 cohorts exclusively (MAS/non-MAS). Putative predictor variables were compared between cohorts. A decision tree analysis for diagnosis of MAS in pSLE was constructed using recursive partitioning, and decision rules were subsequently applied to an independent cohort of newly diagnosed, active pSLE diagnosed and admitted to SickKids between July 2007 and July 2013 (testing cohort) to determine the sensitivity and specificity of the proposed criteria.Results:The training cohort consisted of 56 pSLE patients: 9 (16%) diagnosed with MAS and 47 non-MAS patients. Splenomegaly was more common in the non-MAS cohort, with no other differences in clinical characteristics between cohorts. Of all the available laboratory data, ALT ≥ 45 units/L, neutrophils < 1.65 × 103/mm3 and ferritin < 836 µg/L identified 56% of the patients with MAS (R2 = 0.75) with 100% specificity. The testing cohort consisted of 9 (20%) MAS and 37 non-MAS pSLE patients. The proposed thresholds for ALT, neutrophil count and ferritin demonstrated a sensitivity of 11% and specificity of 97%.Conclusion:MAS is a life threatening complication of pSLE. Early recognition is challenging yet critical for appropriate therapy. PSLE patient with and without MAS have many similar clinical and laboratory features. Using statistical modeling, the initial criteria developed maintained excellent specificity but poor sensitivity for distinguishing MAS from active SLE at diagnosis in the testing cohort. Hence, new validated models are required for the early recognition of MAS among pSLE patients. Future analyses are planned to address these issues.
    03/2014; 66(S11). DOI:10.1002/art.38431
  • Annals of the Rheumatic Diseases 01/2014; 72(Suppl 3):A728-A728. DOI:10.1136/annrheumdis-2013-eular.2159 · 9.27 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: Systemic lupus erythematosus (SLE) frequently has neuropsychiatric involvement including affective disorders, psychosis, and cognitive dysfunction. Evidence suggests that anorexia nervosa (AN) in adolescents with SLE may be triggered by steroid-induced changes in weight and body shape. We propose that AN may be another manifestation of neuropsychiatric SLE and should be considered in this patient population. A retrospective chart review identified 7 children/adolescents diagnosed with SLE and the Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition diagnostic criteria for AN, restrictive subtype, at the Hospital for Sick Children in Toronto between January 1989 and January 2011. One patient developed AN 15 months after being diagnosed with SLE that was attributed to prednisone-induced weight gain and cushingoid appearance. Of the remaining 6 patients, the median age at onset of AN symptoms was 12.2 years and diagnosis of AN was 13.6 years. The median age at SLE diagnosis was 14.2 years with median time after onset of AN symptoms of 20 months (7.5-32 months). All patients had evidence of joint symptoms and a positive antinuclear antibody, and 50% had lymphopenia at the time of SLE diagnosis. Treatment of SLE resulted in improvement of AN in all patients. The timing of the clinical presentation of AN in relation to the diagnosis of SLE and response to SLE treatment suggests that AN may be a novel presentation of neuropsychiatric SLE. Patients with AN who present with or develop joint symptoms, a positive antinuclear antibody, or lymphopenia should be investigated and followed for possible SLE.
    PEDIATRICS 01/2014; 133(2). DOI:10.1542/peds.2012-3048 · 5.30 Impact Factor
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: Epidemiological associations suggest that vitamin D status may play a role in inflammation and progression of atherosclerosis. Using frozen serum, carotid intima medial thickness (CIMT) measurements and other existing data from the Atherosclerosis Prevention in Pediatric Lupus Erythematosus (APPLE) trial, we assessed interactions between serum 25-hydroxyvitamin D (25(OH)D), atorvastatin randomisation and CIMT progression rate.

Publication Stats

8k Citations
1,374.24 Total Impact Points


  • 1989–2015
    • SickKids
      • • Division of Rheumatology
      • • Division of Cardiology
      • • Department of Paediatrics
      • • Department of Diagnostic Imaging
      Toronto, Ontario, Canada
    • Yale-New Haven Hospital
      New Haven, Connecticut, United States
  • 1987–2015
    • University of Toronto
      • • Hospital for Sick Children
      • • Department of Paediatrics
      • • Division of Rheumatology
      • • Laboratory Medicine Program
      Toronto, Ontario, Canada
  • 2011
    • Cincinnati Children's Hospital Medical Center
      • Division of Rheumatology
      Cincinnati, OH, United States
  • 2009
    • The University of Chicago Medical Center
      Chicago, Illinois, United States
    • University of Oklahoma Health Sciences Center
      • Department of Pediatrics
      Oklahoma City, Oklahoma, United States
    • University Health Network
      Toronto, Ontario, Canada
  • 2007–2009
    • IRCCS Istituto G. Gaslini
      Genova, Liguria, Italy
  • 2002–2009
    • University of Cincinnati
      Cincinnati, Ohio, United States
  • 1995–2006
    • Dalhousie University
      • Department of Pediatrics
      Halifax, Nova Scotia, Canada
  • 1996
    • University of Saskatchewan
      • Department of Pediatrics
      Saskatoon, Saskatchewan, Canada
    • University of Ottawa
      • Department of Pediatrics
      Ottawa, Ontario, Canada
  • 1992–1996
    • Toronto Western Hospital
      Toronto, Ontario, Canada
  • 1993
    • University of Texas Health Science Center at Houston
      • Department of Internal Medicine
      Houston, Texas, United States
  • 1992–1993
    • Baylor College of Medicine
      • Department of Pediatrics
      Houston, Texas, United States
  • 1985
    • Stanford University
      Palo Alto, California, United States
    • Stanford Medicine
      Stanford, California, United States
  • 1984
    • Children's Hospital Los Angeles
      Los Angeles, California, United States