Analysis of a pediatric rheumatology clinic population

This analysis evaluates the role of a pediatric rheumatology clinic in assessing children with suspected rheumatic diseases and establishes relative disease frequencies in a clinic population. The study population comprised 875 children referred to a pediatric rheumatology clinic serving a population of 290,000 children. The mean annual referral rate was 113 patients. A diagnosis was established in 580 (66%) of whom 337 (58%) had a rheumatic disease. Of those with a rheumatic disease 156 (46%) had juvenile rheumatoid arthritis, 104 (31%) a spondyloarthropathy, 62 (18%) a connective tissue/collagen vascular disorder and 15 (5%) a variety of other conditions. Of the 243 diagnosed as having a nonrheumatic disease 79 (33%) had a mechanical or traumatic cause for musculoskeletal symptoms, 33 (14%) had an infection, 15 (6%) a neoplastic disorder and 71 (29%) a variety of other disorders. In addition, 45 children (19%) were evaluated because of family histories of rheumatic diseases or questionably abnormal symptoms or signs; after evaluation all these children were considered to be normal. The remaining group comprised 295 subjects (34%) for whom a definite diagnosis has not been made. In addition to diagnosing and caring for children with rheumatic disorders a pediatric rheumatology clinic serves to identify nonrheumatic conditions and provides information concerning relative frequencies and epidemiologic characteristics of childhood rheumatic diseases.     

Profile of a pediatric rheumatology practice in Israel

BACKGROUND: Several studies from Western countries have analyzed the profile of pediatric rheumatology practices. Due to differences in demography and health care systems the profile in Israel may differ from those countries. OBJECTIVE: To describe the profile of a pediatric rheumatology practice in Israel. METHODS: All new patients seen during the course of 2000 as part of my pediatric rheumatology practice in Northern Israel were registered at their initial encounter. Recorded were demographic data, referral patterns, diagnoses, and disease-related data. Diagnoses were grouped together by types of condition. RESULTS: 242 new patients were seen. 39% of the patients had a rheumatic condition, 39% had non-inflammatory conditions, 12% had periodic fever syndromes and for 10% no definitive diagnosis was determined. 14% had chronic rheumatic diseases. The time until diagnosis was significantly greater and more physicians were involved in the evaluation of periodic fever syndromes than in other disease groups. Seventeen (7%) patients had juvenile rheumatoid arthritis (JRA). The minimum estimated incidence of JRA was 8.8 per 100,000 children. CONCLUSIONS: Most patients seen did not have classic inflammatory rheumatic diseases, similar to data from other Western countries. Distinctive to Israel and the Middle East, periodic fever syndromes comprise a large proportion of the pediatric rheumatology practice. These syndromes are relatively difficult for community physicians to diagnose.     

Immunity to soluble retinal antigen in patients with uveitis accompanying juvenile rheumatoid arthritis

Studies of immunity to bovine soluble retinal antigen (antigen S) were carried out using serum and peripheral blood lymphocytes from children with juvenile rheumatoid arthritis and chronic anterior uveitis (JRA-uveitis), children with JRA alone, children with nonrheumatic diseases, and controls who had no ocular or rheumatic disease. Enzyme-linked immunosorbent assay and the lymphocyte transformation assay were used to determine immunity. Antibody to antigen S was present significantly more frequently in children with JRA-uveitis than in children with JRA alone, children with nonrheumatic disorders, or controls. These latter groups did not differ in positivity for this antibody. Lymphocyte transformation occurred more frequently in children with JRA-uveitis than in children with JRA alone or controls. Children with JRA alone and controls had similar frequencies of lymphocyte transformation positivity. Enzyme-linked immunosorbent assay positivity and lymphocyte transformation positivity tended to occur in different children. Children with JRA-uveitis who had HLA-B35 had the highest frequency of antibody to antigen S. Immunity to antigen S may be the result of ocular damage by mechanisms other than a pathogenic mechanism per se.     

Onset to first visit intervals in childhood rheumatic diseases

OBJECTIVE: To determine time intervals between onset of symptoms of a childhood rheumatic disease and first visit to a pediatric rheumatology clinic and to evaluate factors influencing onset to first visit intervals. METHODS: Onset to first visit intervals were analyzed in 836 children representing the 10 most common diseases in a pediatric rheumatology clinic population of 1093. RESULTS: Among 836 subjects, 469 (56.1%) could identify month of symptom onset. Among patients with juvenile rheumatoid arthritis (JRA) 125 of 195 (64.1%) with pauciarticular, 58 of 105 (55.2%) with polyarticular, and 28 of 36 (77.8%) with systemic subtypes were able to determine time interval between symptom onset and first visit. Month intervals were confidently established in 80 of 250 with a spondyloarthropathy (32.4%), 19 of 52 (36.5%) with psoriatic arthropathy, 65 of 72 (90.3%) with Henoch-Sch?nlein purpura (HSP), 50 of 56 (89.3%) with Kawasaki disease, 22 of 34 (64.7%) with systemic lupus erythematosus, 13 of 18 (72.2%) with dermatomyositis, and 9 of 18 (50%) with localized scleroderma. Determination of onset was significantly more likely in HSP than in other diagnostic categories except systemic JRA, and more likely in Kawasaki disease than other disease categories except systemic JRA and dermatomyositis. In the group of 469, 287 (61.2%) were seen within 2 months of symptom onset and 447 (95.3%) within 1 year of symptom onset. CONCLUSION: Diseases ordinarily typified by an abrupt and acute onset of symptoms were referred most promptly, suggesting that acuity of symptoms at disease onset is the factor that most influences promptness of referral. Prospective studies are required to establish how onset to first visit intervals might influence disease outcomes and to devise best practice referral guidelines.     

Underutilization of anticoagulation therapy in chronic atrial fibrillation

Atrial fibrillation, the most common chronic arrhythmia, results in an increased risk of stroke. Anticoagulation therapy can reduce this risk, but appears to be underused. The objective of this study was to examine the use of warfarin and prevalence of stroke in patients with rheumatic, nonrheumatic valvular and nonvalvular atrial fibrillation. Between January 1993 and December 1998, 457 chronic atrial fibrillation patients with continuous follow-up in our hospital were identified as having rheumatic heart disease (n = 114): nonrheumatic valvular disease (n = 65); or nonvalvular disease (n = 278). Warfarin was used less often in patients with nonrheumatic valvular (16.7%) and nonvalvular diseases (20.1%) than in those with rheumatic heart disease (81.6%, p < 0.001). In contrast, the prevalence of stroke among patients with nonvalvular disease was 40.3% which was similar to the 33.3% found in patients with rheumatic heart disease but significantly higher than the 24.6% found in patients with nonrheumatic valvular disease (p < 0.05). A history of stroke did not alter the trend of use of warfarin among the three groups of patients. Only 20.6% of patients on warfarin received monthly monitoring of prothrombin time. In conclusion, the anticoagulation therapy in our patients with chronic atrial fibrillation, regardless of their associated valvular diseases, is significantly underutilized. This underuse could account for a high prevalence of stroke. This risk of stroke, however, is less in patients with nonrheumatic valvular discase than in those with nonvalvular atrial fibrillation.     

Impact of TennCare reform on care of patients with rheumatic diseases

BACKGROUND: As a consequence of cost-cutting changes, termed "TennCare reform," to Tennessee's state-funded health care plan, a number of beneficiaries have either had their benefits substantially reduced or have been disenrolled entirely. The purpose of this study was to determine the impact of these reforms on the health of patients with chronic rheumatic diseases. METHODS: We determined differences with adherence to scheduled appointments in an urban academic rheumatology clinic in the 3 months before and in the 3 months after TennCare reform. A telephone survey was conducted to determine plans for future rheumatic care among those patients scheduled to be seen in this clinic in the 3 months after TennCare reform. RESULTS: Overall, 402 of the 601 patients scheduled for a rheumatology clinic appointment before TennCare reform (67%) adhered to their scheduled rheumatology appointment, compared with 362 of 595 patients (61%) scheduled for an appointment after TennCare reform, a difference that was statistically significant (P = 0.034). After TennCare reform, patients who did not adhere with clinic follow-up were more likely to have been disenrolled from TennCare (P < 0.001). By telephone survey, among those who were disenrolled from TennCare, almost half indicated that they did not know where they were going to receive future rheumatic disease care. Among those who retained TennCare, less than half indicated that they could afford to purchase all of their medications. CONCLUSIONS: TennCare reform has significantly limited care for patients with rheumatic diseases. The long-term consequences of this merit future study.     

Autoantibodies to the chromosomal protein HMG-17 in juvenile rheumatoid arthritis.

OBJECTIVE. To determine the antibody profiles in sera from patients with juvenile rheumatoid arthritis (JRA). METHODS. Immunoblotting using nuclear extracts and recombinant high-mobility group (HMG) nonhistone chromosomal proteins. RESULTS. Antibodies directed against HMG-17 were found in 47% of antinuclear antibody (ANA)-positive patients with pauciarticular-onset JRA and in 16% of ANA-positive patients with polyarticular-onset JRA. HMG-17 values of 6% and 8%, respectively, were detected in ANA-negative patients with JRA and in those with nonrheumatic diseases. CONCLUSION. There is evidence for a high prevalence of anti-HMG-17 antibodies in sera of patients with pauciarticular-onset JRA.     

Aortic valve prolapse with aortic regurgitation assessed by Doppler color-flow echocardiography.

The incidence of and the Doppler color-flow echocardiographic characteristics of aortic valve prolapse with nonrheumatic aortic regurgitation were examined. Aortic valve prolapse was observed in 21 of 243 patients (15 men and 6 women) with aortic regurgitation as detected by Doppler color-flow echocardiography (rheumatic, 112; nonrheumatic, 131) in 1247 consecutive patients. Patients with aortic valve prolapse included three patients with essential hypertension and one with annuloaortic ectasia. The remaining 17 patients (7% of those with aortic regurgitation) had no other associated cardiovascular disease (idiopathic aortic valve prolapse). Prolapse of the mitral or the tricuspid valve or both was associated with aortic valve prolapse in seven patients. Aortic regurgitation jet was markedly deviated from the axis of left ventricular outflow tract toward the anterior mitral leaflet or the interventricular septum in 17 of 21 (81%) patients with aortic valve prolapse, whereas 28 of 110 (25%) patients with nonrheumatic aortic regurgitation without prolapse and 17 of 112 (15%) patients with rheumatic aortic regurgitation without prolapse showed the deviation of regurgitant jet (p < 0.001). In conclusion, idiopathic aortic valve prolapse is one of the significant causes of aortic regurgitation, and a marked deviation of regurgitant jet is a characteristic Doppler color-flow echocardiographic finding of aortic regurgitation that results from aortic valve prolapse.     

Pathology associated to the Baker's cysts: a musculoskeletal ultrasound study

The purpose of this study was to know the pathology associated with Baker's cyst (BC) in a rheumatology clinic and to evaluate the incidence, characteristics, and complications of BC. We reviewed the rheumatology ultrasound laboratory charts of patients with BC from Oct 2006 through Dec 2008. Demographic and disease data were also collected. Of the 1,120 patients who underwent ultrasound studies, 145 (12.9%) were found to have 180 BCs. The associated diseases were as follows: 91 (50.6%) osteoarthritis (OA) of the knee, 37 (20.6%) rheumatoid arthritis (RA), 25 (13.9%) gout, 14 (7.8%) seronegative spondyloarthropathy (SpA), and 13 (7.2%) pyrophosphate arthropathy. We found ruptured BCs in 12 patients, whose associated pathologies were in the following: four RA, four OA of the knee, two gout, one SpA, and one pyrophosphate arthropathy. The most frequent associated arthropathy of BCs was OA (50.6%), followed by RA (20.6%). However, in the cases of ruptured BC, the inflammatory pathology (66.7%) is more frequent than the degenerative one (33.3%).     

Association between vitiligo and spondyloarthritis

OBJECTIVE: To establish if spondyloarthritis (SpA) and vitiligo occur together more frequently than by chance. METHODS: All consecutive patients with SpA seen in a 6 month period were evaluated for vitiligo by an experienced dermatologist. The control group included the 2 consecutive patients without SpA seen after each patient with SpA. RESULTS: Two hundred thirty-four patients with SpA (131 men, 103 women; mean age 59 +/- 18.3 yrs) were seen in the study period. Of these, 43 had ankylosing spondylitis (AS), 112 psoriatic arthritis (PsA), 14 SpA associated with inflammatory bowel disease, 64 undifferentiated SpA, and one reactive arthritis. The 468 control patients (360 women, 108 men; mean age 68.5 +/- 2 yrs) had various degenerative and inflammatory rheumatic diseases. Eight (3.4%) patients out of 234 with SpA had type A vitiligo. In the control group, 5 (1.06%) out of 468 had type A vitiligo. The difference was statistically significant (p < 0.05). Of the 8 patients with coexisting vitiligo and SpA, 4 had PsA, 2 primary AS, one AS associated with Crohn's disease, and one undifferentiated SpA. Of the 5 patients with vitiligo in the control group, one had rheumatoid arthritis, one S ogren's syndrome, one palindromic rheumatism, one crystal arthropathy, and one osteoarthritis. CONCLUSION: Our results suggest that vitiligo and SpA do not coexist by chance and that vitiligo should be included in the list of diseases associated with SpA.     

Rheumatological medicine literacy among Middle Eastern populations

Background: It has been observed for years that many Middle Eastern patients with autoimmune rheumatic disorders, are more likely to be delayed in seeing a rheumatologist for their symptoms and that the rheumatology services are in general under‐utilized by the population. Aim: To explore if patients with autoimmune rheumatic disorders were truly delayed in seeing rheumatologists and to explore the possible reasons for that delay should any delay be documented. Subjects and methods: Patients suffering from chronic autoimmune rheumatic disorders were interviewed and were asked to answer a questionnaire that assesses their initial set of actions when they had their first symptoms of disease, how much time they took to see a rheumatologist and their background knowledge about rheumatology as a specialty before and after they saw a rheumatologist. Results: Seventy‐eight patients, 57 (73%) females and 21 (27%) males were included in this study. Their ages ranged from 11 to 72 years with a mean of 38.9 ± 13 years. Patients’ explanations for their initial symptoms were ‘evil eye doing’, disease, exertion, cold weather and trauma in 44%, 37%, 20%, 16% and 8% respectively. Ninety‐six percent of patients had to make a total of 166 consultations first at other specialties before they were finally advised or directed to see a rheumatologist. Non‐rheumatologist referrals to rheumatologists happened in only 33% of the time. The duration from the onset of the disease until patients finally came to see a rheumatologist ranged from 0.5 weeks to 432 weeks with a mean of 51 ± 88 weeks. Conclusions: General health literacy and knowledge of the rheumatology scope of service is extremely limited among Middle Eastern patients. Most patients with autoimmune rheumatic diseases make their initial consultations at clinics other than rheumatology clinics and non‐rheumatologists have been shown to consistently not refer patients with rheumatic diseases to rheumatologists. Wrong diagnosis is attributed to rheumatology symptoms by non‐rheumatologists 82% of the time. Level of education of patients, has no impact on their choice of the right specialty to be consulted for their disease symptoms.     

Prevalence of self-reported spondyloarthritis features in a cohort of patients with inflammatory bowel disease

BACKGROUND:

Musculoskeletal symptoms belonging to the spectrum of ‘seronegative spondyloarthritis’ (SpA) are the most common extraintestinal manifestations in patients with inflammatory bowel disease (IBD) and may lead to important disease burden. Patients with suspected SpA should be referred to a rheumatologist for further evaluation.

OBJECTIVE:

To investigate the self-reported prevalence of musculoskeletal SpA features in a cohort of patients with IBD and to compare this with actual referrals to a rheumatologist.

METHODS:

Consecutive patients with IBD visiting the outpatient clinic were interviewed by a trained research nurse about possible SpA features using a standardized questionnaire regarding the presence or history of inflammatory back pain, peripheral arthritis, enthesitis, dactylitis, psoriasis, uveitis and response to nonsteroidal anti-inflammatory drugs. All patient files were verified for previous visits to a rheumatologist and any rheumatic diagnosis.

RESULTS:

At least one musculoskeletal SpA feature was reported by 129 of 350 (36.9%) patients. No significant differences between patients with Crohn disease and ulcerative colitis were found. Review of medical records showed that 66 (51.2%) patients had ever visited a rheumatologist. Axial SpA was diagnosed in 18 (27.3%) patients, peripheral SpA in 20 (30.3%) patients and another rheumatic disorder in 14 (21.2%) patients.

CONCLUSION:

Musculoskeletal SpA features are frequently present in patients with IBD. However, a substantial group of patients is not evaluated by a rheumatologist. Gastroenterologists play a key role in early referral of this often debilitating disease.     

Rheumatology Education in United States Medical Schools

Although rheumatology manpower in United States medical schools has dramatically increased in the past decade, 13% of medical schools did not have a full-time staff rheumatologist in 1980. Thirty-eight percent of medical schools had 2 or less full-time rheumatologists. Staff rheumatologists and rheumatology fellows provided the majority of medical student education in the clinical aspects of rheumatic disease; however, rheumatologists in less than 50% of medical schools taught in the basic science curriculum or in related fields such as collagen biochemistry, metabolic bone disease, and orthopedic intervention in arthritis. The staff rheumatologists' time commitment to medical student education was inversely proportional to the rheumatology faculty size. At medical schools with no rheumatologists, however, there was little, if any, formal education in the rheumatic diseases. Most subjects are taught in systems-oriented lectures. Education is currently limited to the common rheumatic conditions such as bursitis and back pain. Only 62% of medical schools provide a structured course on the musculoskeletal examination. Elective rotations in rheumatology, usually offered in the third or fourth year, are currently being provided to only 15% of U.S. medical students.     

Autoantibodies to the chromosomal protein HMG-17 in juvenile rheumatoid arthritis

Objective. To determine the antibody profiles in sera from patients with juvenile rheumatoid arthritis (JRA). Methods. Immunoblotting using nuclear extracts and recombinant high-mobility group (HMG) nonhistone chromosomal proteins. Results. Antibodies directed against HMG-17 were found in 47% of antinuclear antibody (ANA)–positive patients with pauciarticular-onset JRA and in 16% of ANA-positive patients with polyarticular-onset JRA. HMG-17 values of 6% and 8%, respectively, were detected in ANA-negative patients with JRA and in those with nonrheumatic diseases. Conclusion. There is evidence for a high prevalence of anti—HMG-17 antibodies in sera of patients with pauciarticular-onset JRA.     

Fever of unknown origin caused by adult juvenile rheumatoid arthritis: the diagnostic significance of double quotidian fevers and elevated serum ferritin levels

Fever of unknown origin (FUO) in adults is a commonly encountered clinical problem. Treatable causes of FUO in the adult should be the primary focus of the diagnostic workup. Neoplasms have replaced infectious diseases as being the most common cause of FUO in adults, and collagen vascular diseases are now relatively rare. The most important collagen vascular diseases presenting as an FUO include Takayasu's arteritis, Kikuchi's disease, polymyalgia rheumatica, and adult juvenile rheumatoid arthritis (JRA) (adult Still's disease). There are no specific diagnostic tests for these disorders, which commonly present as prolonged fevers that are not easily diagnosed (i.e., FUO). Adult JRA is a rare but important cause of FUO in adults. Typically, patients with adult Still's disease present with liver/spleen involvement, posi-articular arthritis, ocular involvement, and evanescent salmon-colored truncal rash. An important diagnostic finding in adult JRA is the presence of a double quotidian fever, which occurs in few other disorders. Only visceral leishmaniasis and adult JRA are causes of FUO in adults associated with double quotidian fevers. Highly elevated serum ferritin levels are the most important nonspecific diagnostic finding associated with adult JRA. We present a case of FUO caused by adult JRA presenting with diffuse polyarticular migrating arthritis, evanescent rash, and splenomegaly. The diagnosis of adult JRA was suggested by these findings in association with a double quotidian fever and a highly elevated serum ferritin level. Clinicians should appreciate the diagnostic significance of fever patterns and the diagnostic significance of elevated serum ferritin levels in patients with FUO.     

Ethnic differences in risk for pediatric rheumatic illness in a culturally diverse population

OBJECTIVE: To analyze the differences of occurrence of pediatric rheumatic disease among various ethnic groups in a culturally diverse isolated geographic area. METHODS: A retrospective study of pediatric rheumatic diseases in a multiethnic area during a 6 year period. RESULTS: A group of 922 patients was categorized based on predominant ethnicity, and their risk of having acute rheumatic fever (ARF), juvenile rheumatoid arthritis (JRA), and systemic lupus erythematosus (SLE) was studied. Odds ratios (OR) were computed for each illness with Caucasians as the reference group. Results indicated that Polynesians were overrepresented among patients with ARF, having elevated OR that were significantly different from Caucasians (22.5-120.7, p < 0.0001). For SLE, the highest OR were obtained for Samoans, Filipinos, and Japanese. In contrast, for JRA, Filipinos and Japanese had OR less than one, and no Samoans were diagnosed with JRA, possibly indicating a protective effect against developing JRA. CONCLUSION: This unique retrospective study examined the ethnic variations of expression of certain rheumatic diseases in an isolated region. Results reveal that certain ethnic groups are at risk for ARF and SLE, but are protected against JRA. These findings suggest investigating possible immunogenetic similarities and differences in these illnesses.     

Increase in CD5+ B cells in juvenile rheumatoid arthritis. Relationship to IgM rheumatoid factor expression and disease activity.

OBJECTIVE. To investigate the association between CD5+ B cell expression and IgM rheumatoid factor (IgM-RF) in juvenile rheumatoid arthritis (JRA). METHODS. CD5+ B cell levels analyzed by flow cytometry and IgM-RF expression determined by enzyme-linked immunosorbent assay were compared in children with JRA, children with other collagen vascular diseases, and healthy controls. RESULTS. Children with polyarticular JRA had expanded populations of CD5+ B cells, and expansion of CD5+ B cells and IgM-RF both correlated with disease activity. CONCLUSION. The results indicate that an expanded CD5+ B cell population leads to IgM-RF production in patients with polyarticular JRA, as well as patients with RA.     

Autoantibodies to nonhistone chromosomal proteins HMG-1 and HMG-2 in sera of patients with juvenile rheumatoid arthritis

IgG antibodies against the high mobility group (HMG) nonhistone chromosomal proteins HMG-1 and/or HMG-2 were detected in the sera of 49 (39%) of 126 antinuclear antibody (ANA)-positive patients with juvenile rheumatoid arthritis (JRA), by immunoblotting. Clinical diagnosis classified these patients in 2 major groups, 105 with pauciarticular-onset JRA and 21 with polyarticular-onset JRA. Anti-HMG-1 and/or anti-HMG-2 antibodies were found in 8 (25%) of 32 pauciarticular-onset JRA patients with uveitis and in 34 (47%) of 73 patients without uveitis, whereas anti-HMG-1 and/or anti-HMG-2 antibodies were found in 4 (24%) of 17 children with polyarticular-onset JRA without uveitis. Among 53 sera from ANA-negative JRA patients, 3 (6%) were positive for anti-HMG-1 and/or anti-HMG-2 antibodies, whereas no reactivity to HMG-1 or HMG-2 proteins was observed in 48 sera from age-matched children with nonrheumatic diseases.     

Referrals from General Practice to an Outpatient Rheumatology Clinic: Disease Spectrum and Analysis of Referral Letters

Our objective was to study the demographic characteristics of patients referred from general practitioners to a rheumatology outpatient clinic and to analyse the content and quality of the referral letters. During a 12-month period 346 randomly chosen referral letters of new patients from GPs to a rheumatology outpatient clinic were evaluated. The mean age of the 346 referred patients (73.1% females and 26.9% males) was 45.5 years and 17.8% were 60 or older. Mean disease duration at the time of referral was 50.9 months (1–432 months). Only about 10% of the patients referred had a disease duration of 1 month or less. The current clinical problem was appropriately presented in 95% of the referral letters. In only 0.9% of referrals had there been a prior phone consultation. Altogether, 95.1% of the referrals were as a result of diagnosis or treatment, and in nearly half the cases a diagnosis of inflammatory rheumatic disease was suggested. In 23% of the letters the result of clinical examinations were missing.  Laboratory tests such as serum rheumatoid factor, antinuclear antibodies and HLA-B27 were used by GPs to screen for rheumatic disease in general. A lack of correlation between clinical manifestations and subsequently requested laboratory examinations was frequently found in the referral letters, exemplified by the use of HLA-B27 in rheumatoid arthritis and serum rheumatoid factors in ankylosing spondylitis. These results show that among GPs the threshold for referring patients to a rheumatology outpatient clinic appears rather high, and that patients are subjected to long observation periods before referral. A more frequent use of phone consultations and an improvement in the diagnostic skills of GPs may positively influence the selection of patients for referral and shorten the long waiting lists in rheumatology. This need for improvement was further strengthened by GPs’ inappropriate use of laboratory tests. Received: 14 June 1999 / Accepted: 5 April 2000