Henoch-Schonlein purpura (HSP) is more common in children than adults. However, the current ACR classification criteria* were developed in predominantly adult populations in order to distinguish between primary systemic vasculitides. These criteria may, therefore, have limitations when applied to childhood populations.

Aims: To estimate the incidence of HSP in children using the ACR criteria*, to assess the value of the criteria in defining this population, and to identify the presenting clinical features of HSP in an incidence cohort of children.

Methods: A resident population of 1.1 million children was surveyed over 2 years. Data were collected by monthly questionnaires sent to 321 hospital consultants, a single questionnaire sent to 2860 general practitioners, and review of 406 further case notes with hospital diagnostic codes for vasculitis. Cases fulfilling the ACR criteria* were included. However, children with isolated palpable purpura (PP) were only included when the PP was in the classical distribution.

Results: 463 children fulfilled diagnostic criteria. All cases fulfilled the age criterion, and had PP. Few cases fulfilled biopsy (1%) or gastrointestinal bleeding (2%) criteria. Moderate thrombocytopenia (105-142 x 109/l) excluded 20 children with classical PP, including 12/20 with arthritis +/or abdominal pain. A hierarchy of clinical features was recognized: PP (100%); arthritis (75%); arthritis + PP only (37%); abdominal pain (35%); classical triad (28%); PP only (14%); abdominal pain + PP only (5.6%). The sex ratio (M:F) was 1.2:1 overall, but arthritis was twice as common in boys (2.4:1) unlike isolated PP (1.08:1), or severe disease (1.07:1). Urinalysis was normal in 61%, with significant renal disease on biopsy in 1%.

Conclusions: The data suggest that the current ACR criteria are inappropriate for the paediatric age group. PP was the only ACR criteria identifying >90% of HSP in a childhood population. New criteria for children should be developed for prospective testing.

*Mills et al. Arthritis Rheum 1990; 33:1114-1121.

METHODS: Retrospective study of unselected children (age 14 and younger) diagnosed with HSP at the single hospital for a defined population in NW Spain between 1980 and 1999. Patients were classified according to the criteria proposed by Michel et al in 1992. To assess renal outcome only those patients with a follow-up of at least 1 year were reviewed.

RESULTS: Seventy-eight children (median age 5.5 years) were classified with HSP. Girls slightly outnumbered boys. Cases occurred more commonly in fall and winter. A previous history of upper respiratory infection was frequent. The overall average annual incidence rate was 10.45/100,000. Abdominal pain and or joint manifestations preceded the onset of purpura in 31% of the cases. All children developed nonthrombocytopenic palpable purpura during the course of the disease. Arthritis was seen in 65%, gastrointestinal bleeding in 31% and renal manifestations in 54% within the first 3 months after the onset of symptoms. All but one of the patients with renal manifestations had hematuria, associated with proteinuria in 19 of 41 cases. Sixty-nine cases were followed for at least 1 year. After a median follow-up of 7 years, 8 (11.6%) patients had persistent hematuria with proteinuria. However, none of them had developed renal insufficiency. Hematuria at the onset of the disease or renal manifestations within the first 3 months after the onset of the symptoms and relapses were significantly more common in the group of patients with renal sequelae. The presence of nephrotic syndrome during the course of the disease was generally associated with renal sequelae. In contrast, the age at the onset of the disease was not associated with a different disease severity and outcome. Using discriminant analysis the best predictor for renal sequelae was the presence of nephrotic syndrome during the first 3 months after the onset of symptoms. Using this model 65 of 69 patients with a follow-up of at least 1 year were classified correctly for the risk to develop renal sequelae.

CONCLUSIONS: In NW Spain, HSP is a common, generally benign and self-limited, vasculitis. Some children, however, in particular those who develop nephrotic syndrome, suffer permanent renal involvement.

Methods: Prospective monthly questionnaires were sent to 321 paediatricians, rheumatologists and other relevant consultants, with a mean return rate of 71.1% over 3 years (1996-99). Case ascertainment was verified by review of a further 406 case notes not identified by the questionnaire but with diagnostic codes for vasculitis, and a single questionnaire to 2860 primary care doctors (GPs) (return rate 59.7%) at the end of the study.

Results: 586 new cases of vasculitis fulfilling established diagnostic criteria were collected prospectively from the 1.2 million children resident in the region. All Asian children originated from the Indian subcontinent, none were of oriental origin. Asian children had a higher incidence of KD and rarer primary systemic vasculitis(PSV) than white children. Asian and black children had a higher incidence of systemic lupus erythematosus (SLE), juvenile dermatomyositis (JDMS).

Table: Incidence (95% CI)/100,000 children/year. *p<0.01 between ethnic groups.

% paediatricpopulation HSP(2 year data)n=463 KDn=73 SLEn=28 JDMSn=14 rarer PSVn=8

All cases 20.4 (18.6-22.4) 2.1 (1.7-2.7) 0.8 (0.5-1.2) 0.4 (0.2-0.7) 0.24 (0.1-0.5)

Asian 10% 24.0 (18.2-31.2) 4.9* (2.8-7.8) 5.6 (3.0-9.5) 0.6 (0.1-2.1) 1.7* (0.5-4.4)

Black 3% 6.2* (1.7-16.0) 2.1 (0.3-7.5) 3.1 (0.6-9.1)) 2.1* (0.3-7.5) -

White 86% 17.8 (16.0-19.8) 1.9 (1.4-2.4) 0.4* (0.2-0.7) 0.3 (0.2-0.6) 0.14 (0-0.4)

The annual incidence of HSP was higher than previously described, rising to 22.1/100,000 under 14 years (95% CI 19.2-25.0), and 70.3*/100,000 between the ages of 4-6 years (95% CI 54.2-86.4). The annual incidence of KD rose to 5.5/100,000 in children under 5 (95% CI 3.1-7.8), and 14.6* /100,000 (95% CI 1.3-27) in Asian children under 5 years. In girls over 11 years the annual incidence of SLE was 12.3* /100,000 (95% CI 7.3-17.3). Further analysis revealed clustering (at p=0.05 level) of KD (time: 180 days, space 1500 metres) and HSP (time: 120 days, space 1500 metres).

Conclusion: Our study has recognised a high incidence of vasculitis in non-oriental Asian and black children, and a higher incidence of HSP than previously reported. Genetic and environmental influences may be involved.

Methods: A retrospective chart review was conducted of all patients admitted at our center with a diagnosis of KD between January 1st, 1985 and December 31st, 1999. Patients who either fulfilled KD diagnostic criteria or were deemed to have atypical or incomplete KD, and were treated in the acute phase with low dose ASA and any of the 4 IVIG regimens that were standard at the time, were included.

Results: Study entry criteria were met by 221 patients. There were 81 girls and 140 boys (1:1.7) with a mean age at diagnosis of 2.97 years (range 0.1-13). One hundred and eighty (81.45%) fulfilled KD diagnostic criteria, 38 (17.19%) had incomplete KD, and 3 (1.36%) had atypical KD. Twenty (9.05%) patients developed CAL. In those treated with the IVIG regimen of 2 g/kg, 10 of 139 (7.19%) developed CAL. Thirty-nine patients (17.65%) had other early cardiac abnormalities including myocardial dysfunction, pericarditis, and valvular abnormalities. Total fever duration was longer in those with CAL (p=0.0006) and in those with other early cardiac abnormalities (p=0.012), and this likely represented a more intense inflammatory process. Of the 196 patients who had ECGs, 27 (13.78%) were abnormal. No serious events occurred. There was 1 death (0.45%) which was unlikely related to KD.

Conclusion: Treatment with low dose ASA and IVIG in the acute phase of KD is associated with a similar incidence of CAL to high dose ASA and IVIG. Other early cardiac abnormalities and ECG abnormalities occur commonly. All cardiac events are associated with longer duration of total fever. The significance of these findings and the long-term outcome of this cohort is currently under study.

Methods: A retrospective chart review was conducted of all patients admitted at our center with a diagnosis of KD between January 1985 and December 1999. Patients who either fulfilled KD diagnostic criteria or were deemed to have atypical or incomplete KD, and who were treated with low dose ASA and IVIG were included. From these patients, those who had undergone serial platelet testing (at least 2 values within 7 days of each other) and had demonstrated thrombocytosis (platelet count > 450,000) at any point were studied further.

Results: A total of 221 patients were included. Of these, 165 fulfilled platelet monitoring criteria and underwent further study. The mean day of illness in which thrombocytosis was first documented was 9.32 (range 2 - 22) with a mean platelet count of 565,394 (range 451,000 - 989,000). Interestingly, 49 (29.69%) patients had an elevated platelet count at the time of presentation. The mean day of presentation in these patients was 6.76 (range 2 - 17), while the mean day of illness for commencement of treatment of the total cohort was 6.61 (range 2 - 17).

Conclusion: The mean day of illness in which thrombocytosis occurred was likely lower than 9.32 days because patients did not have daily platelet counts and only the earliest documented thrombocytosis was included. Thrombocytosis can occur earlier than previously described in patients with KD. Therefore, introduction of anti-platelet (low) dose ASA during the acute phase of the disease seems appropriate.

Results: 73 children with KD and 20 with IKD were identified from 164 reported to the study. 58 additional children referred as possible KD lacked sufficient criteria for either group and were rejected. The estimated annual incidence under 5 years old for KD was 5.5 per 100,000 (95% CI 4.2-7.1 per 100,000), compared to 2.4 per 100,000 (95% CI 1.4-3.9 per 100,000) for IKD, with significantly higher incidence rates in non-oriental Asian children in both groups. 3 children with KD and 1 child with IKD developed persistent CAA (comparative CAA risk IKD:KD = 0.8). One child in the rejected group developed CAA (comparative CAA risk rejects:KD = 0.4). In IKD, misery was more common (IKD:KD 2.5:1) whereas lymphadenopathy (IKD:KD 0.4:1) and conjunctivitis (IKD:KD 0.4:1) were reported less. In all groups late peeling of the extremities was more frequently recognized than early erythema and oedema. There was no significant difference in the mean time to presentation (10.6(0-54):9.6(0-37) days), the ethnic mix (comparative risk in Asian children 0.73) or the median age (3.2:3.5 years) between KD and IKD respectively. Application of risk analysis from Beiser et al2 accurately identified the children with CAA in both KD and IKD groups.

Conclusions: Inclusion of children missing one criterion identifies an additional 40% of children with a comparable risk of CAA, and similar population characteristics to children with KD, who also have other non-criteria features of the illness.

1. American Heart Association Committee. Am J Dis Child. 1990;144:1219-1220.

2. Beiser et al. Am J Cardiol 1998; 81: 116-1120.

In this study, we show that responses to LCWE possess all the hallmarks of a superantigen (SAg)-mediated response. LCWE-induced activation of naive T lymphocytes requires antigen presentation but not processing by accessory cells. As in the case of other known SAgs, the response to LCWE is MHC non-classically restricted, and there exists a hierarchy of differing class II MHC in the ability to present LCWE. Furthermore, the most compelling evidence of antigenic activity of LCWE is its ability to induce the characteristic activation and proliferation of T cells expressing specific T cell receptor variable families 2, 4, 6 and 14, followed by the deletion of these reactive T cells. Most importantly, superantigenic activity of LCWE correlates with its ability to induce coronary vasculitis upon injection into mice. Only LCWE preparations possessing superantigenic properties mediate cellular infiltrations in the cardiac tissue leading to coronary arteritis in vivo.

Taken together, these findings demonstrate that LCWE contains a novel superantigen, the activity of which may lead to disease pathogenesis. Since LCWE-mediated coronary arteritis in mice is currently the best animal model of KD, understanding of the mechanisms leading from immune activation to coronary artery lesions in this model will be of great benefit in designing new treatment strategies, reducing vascular damage, and, more importantly, the significant morbidity and mortality associated with KD.

Kawasaki disease (KD) is the most common cause of multisystem vasculitis in children in the developed world. Although KD is widely believed to be the result of an infectious agent, its etiology is still unknown. Past research has reported the possible role for bacterial superantigens in the induction of KD. Our laboratory has characterized a novel superantigen found in Lactobacillus casei cell wall extract (LCWE) which is responsible for triggering the immune response leading to KD-like coronary vasculitis in injected mice.

In this study, we investigated the T helper subsets responsible for disease induction by studying expression of signature cytokines. LCWE was injected into susceptible mouse strains and RT-PCR and immunohistochemistry was performed on peripheral lymphoid tissue at set time-points. T lymphocyte activation and cytokine production was determined. Furthermore, peripheral cytokine levels and T cell activation was investigated and correlated to that in the heart end organ.

Hypothesis: Anticoagulation does not affect the natural history of GAs in patients with KD.

Objectives: To determine: 1)Whether anticoagulation affects remodeling of coronary aneurysms. 2)If anticoagulation is cardioprotective for myocardial ischemia.

Method: A retrospective cohort study of all patients with KD who had a 2D echocardiogram at HSC between 05/90 and 12/00 was performed.

Results: Twenty-two of 997 patients (2.2%) had GAs. Patients were divided into two groups:1)Therapeutic anticoagulation with warfarin (with or without antiplatelet agents); 2)No anticoagulation (with or without antiplatelet agents). There was a male predominance in both the total KD and GA populations. There were no significant differences between the groups for sex distribution (11M:2F vs. 7M:2F), and age at diagnosis (5.2 vs. 3.8 yr). Mean duration of follow-up was significantly longer for the no anticoagulation group (6.9 vs. 13.3 yr, p=0.008).

Anticoagulation (n=13) No Anticoagulation (n=9) p-value

# GAs regressed to <0.8 cm 8 (36%) 9 (53%) 0.31

Ischemic events (myocardial infarction, coronary artery obstruction, or perfusion defects on thallium or sestamibi stress testing) occurred in 5 patients in the anticoagulation group (4 early, 1 late event). Two late ischemic events occurred in the no anticoagulation group.

Conclusions: Regression of GAs is observed with or without therapeutic anticoagulation. Early ischemic events occur independently of warfarin therapy. The incidence of late events appears to be unaffected by anticoagulation. Long term (>20 years) follow-up studies are needed to address this important issue. Clinical evidence of the potential protective effects of anticoagulation must be balanced with the necessary lifestyle changes and bleeding complications with lifelong therapy.

Results Significantly raised ORs (95% C.I.) were found for farm exposure in the Index Year in PSV [3.15(1.70-5.83)] and WG [3.59(1.83-7.03)]. Exposure to livestock (cows, sheep, chickens) was significantly associated with PSV [3.78(1.17-12.22)], WG, mPA and CSS in contrast to crops [PSV, OR: 2.43(0.99-5.94)]. Direct contact with livestock gave an OR of 3.60(1.33-9.70) in PSV. Working in high silica exposure jobs in the Index Year gave raised ORs for PSV [3.62(1.41-9.31)],WG [3.45(1.16-10.25)] and CSS [5.6(1.34-23.46)]. Employment of >40 years in these jobs gave an OR of 2.57(1.01-6.58) in PSV and 4.58(1.18-17.83) for any duration in mPA. A history of a high solvent exposure occupation was significantly associated with PSV [2.35(1.03-5.37)], WG [3.69(1.54-8.85)] and cANCA [3.43(1.22-9.68)]. There were no significant differences for age, sex, social class, urban/rural residence, pets, metal exposure or specific occupational codes.

1. Watts et al; BJR; 1998; 37; suppl., 86; 2. Duna G.F et al; Clin Exp Rheum; 1998; 16; 669-674

Background: Because PAN, MPA, WG and CSS are rare diseases, only a few population-based studies have been conducted to estimate their precise frequencies, particularly in urban populations.

Disease Prevalence/1,000,000 Incidence/1,000,000

PAN 22.3 (13.8 - 33.3) 0.9 (0.03 - 5.6)

MPA 11.7 (6.2 - 21.0) 3.7 (1.1 - 10.2)

WG 25.7 (17.0 - 38.1) 3.7 (1.1 - 10.2)

SCS 7.3 (2.8 - 14.4) 0.9 (0.03 - 5.6)

All 4 70.2 (54.6 - 88.4) 17.4 (9.9 - 27.2)

Conclusion: Compared to previous estimates based mostly on rural populations, our results suggest, in particular, a slightly lower frequency of WG that more probably reflects methodological and/or diagnostic differences rather than specific environmental etiological factors.

Methods: Patients meeting the clinical descriptive criteria for BACNS(J Rheum 20:2046,1993) who were evaluated and treated by a single investigator were included. Data on demographics, signs and symptoms, laboratory studies, neuroimaging, brain biopsy, treatment, and complications was recorded. Short and long term outcomes were assessed. The long-term assessment included a phone interview that utilized the Barthel index and a specifically designed cognitive index.

Results: 16 cases met the inclusion criteria for this study. Mean age was 40 years (10-66) with female to male ratio of 4.3/1and a mean follow up period of 35 months (0-128). Headache was the most common presenting symptoms seen in 88% of the cases followed by focal symptoms (63%), and diffuse symptoms (44%). All patients had highly abnormal cerebral angiography and MRI abnormalities were present in 77%. Severe CSF abnormalities was present in only one patient (7%). All patients recieved less than 6 months of glucocorticoids and none recieved cytotoxic agents. Follow-up cerebral angiography showed marked improvement in all. 94% of the patients demonstrated recovery, 6% relapsed and there were no deaths. 71% of patients assessed by the Barthel index showed no disability and 29% had only mild disability. Patients also had favorable response when assessed by the cognitive index.

Conclusion: There is a subset of PACNS characterized clinically by acute presentation, most commonly with headache, normal to mild CSF findings with female predominance, and radiologically by highly abnormal cerebral angiography that is reversible after treatment, and requiring less intensive treatment than has been traditionally used.

Materials and Methods: DNA samples isolated from 70 patients with TA and 191 healthy controls were studied. DNA typing for HLA-B specificity was investigated by using PCR and sequence specific oligonucleotide probe (SSOP) analysis.

Results: Among the HLA-B alleles only B*52 has shown a significant association with TA (13/70 vs. 13/191, p=0.009, OR: 3.1).The comparison between subtypes according to angiographic classification did not reveal any subtype association with any allele. In this cohort, out of 17 patients with renal artery stenosis only 2 had B*39.

Conclusion: In this study, the previously reported association of TA with B*52 in Japanese and other populations was confirmed in patients from Turkey, whereas no association with B*51 and B*39 is observed. Repeatedly demonstrated association with B*52 might have implications about the functional relevance of this molecule in TA pathogenesis, which has to be further explored.

Takayasu arteritis (TA) is a rare, chronic idiopathic granulomatous vasculitis of the aorta and its branches, predominantly affecting young women (<40 yo).

OBJECTIVE: To determine the diagnostic features of children diagnosed with TA.

METHODS: We identified 117 patients <18 yrs old from published reports on TA using the National Library of Medicine PubMed system. We systematically analyzed demographic and clinical data at presentation. We then identified five patients with TA cared for at CHOP within the last 10 years and compared them to published cases.

DISCUSSION: ACR criteria for diagnosis of TA includes at least 3 of the following: onset < 40 years of age; claudication of extremities; decreased brachial artery pulse; blood pressure difference >10 mmHg between the arms; subclavian or aorta bruit; or an arteriogram abnormality (sensitivity 90.5%, specificity 97.8%). Failure to recognize the early signs of TA in children leads to a delayed diagnosis (19 mos vs. 10 mos for adults), more severe hypertension, more congestive heart failure (66%), and higher mortality (30-35% vs. 5-15% in adults). Pediatric hypertension is uncommon (<1% prevalence), but both hypertension and elevated ESR are found in most patients with TA. Thus, our review suggests that this combination in pediatric patients should merit further screening for TA, particularly in those with systemic complaints and/or abnormal pulses. Our data further extends the literature by comparing angiographic findings in TA with MRI/MRA of the thoracic and abdominal aorta and great vessels, using gadolinium contrast to image stenotic arteries with thickened enhanced vessel walls. As described in adults, MRI/MRA is emerging as a non-invasive tool to diagnosis TA.

Materials and Methods: Sixty-nine patients (F/M: 60/9, mean age at diagnosis: 31.9 years) were diagnosed with TA according to the angiogram (Classification of International Conference on TA- Tokyo,1994) and fulfilled the 1990 classification criteria of ACR, were assessed according to a predefined protocol.

Results: Most common clinical findings were the absence or weakness of pulses (81%), vascular bruits (77%), pain in the extremites (71%), claudication (69%) and hypertension (42%). Central nervous system manifestations due to vascular insufficiency occurred in 14 patients (20%), 6 of whom developed ischemic cerebro-vascular disease. Other common symptoms were renal artery stenosis (25%), erythema nodosum (14%), visual disturbances (14%), aortic insufficiency (13%) and pulmonary hypertension (4%). Angiographic findings in 66 patients revealed that 49% (n=32) of the patients were type I, 41% (n=27) type V, 8 % type IIa (n=5) and one patient each type IIb and III. A total of 16 patients (23%) had undergone angioplastic revascularization procedures. One patient had pulmonary tuberculosis. 7 patients out of 20 developed a delayed skin reactivity to PPD, 5 of whom had a history of house-hold tuberculosis. Forty-six patients are currently on steroid therapy, and 32(46%)of these are also taking methotrexate (MTX) or azathiopirin (AZA).

Conclusion: Types III and IV are uncommon in TA patients from Turkey, probably explaining the lower frequencies of renal artery stenosis and hypertension compared to other series. PPD positivity and active tuberculosis were also not frequently observed. Combination therapy with MTX or AZA are more commonly prescribed compared to earlier series.