- Split View
-
Views
-
Cite
Cite
Y.-H. Yang, C.-F. Hung, C.-R. Hsu, L.-C. Wang, Y.-H. Chuang, Y.-T. Lin, B.-L. Chiang, A nationwide survey on epidemiological characteristics of childhood Henoch–Schönlein purpura in Taiwan, Rheumatology, Volume 44, Issue 5, May 2005, Pages 618–622, https://doi.org/10.1093/rheumatology/keh544
- Share Icon Share
Abstract
Objective. To evaluate the annual incidence and other epidemiological characteristics of Henoch–Schönlein purpura (HSP) among children in Taiwan.
Methods. The records of patients were derived from the research database of the Bureau of National Health Insurance, Taiwan, Republic of China, from January 1999 to December 2002. Children younger than 17 yr of age with the diagnosis of HSP were included into this study. Data for each patient including sex, age, date of onset and length of hospitalization were recorded and analysed.
Results. A total of 2759 cases were included with an annual incidence of 12.9 (11.8–13.4) per 100 000 children <17 yr of age. The occurrence of HSP had a peak at the age of 5 to 6 yr. In this study, 1118 (40.5%) patients had been hospitalized at some stage. There were 1454 males and 1305 females, for a male to female ratio of 1.11. Males had a higher annual incidence before the age of 10 yr (P = 0.04), and had a lower incidence than females at older ages (P = 0.02). Disease onset was more common in autumn and winter, and no apparent change in seasonal pattern was noted over 4 yr.
Conclusions. Insurance claim data provide useful information on the epidemiology of HSP in Taiwan. Childhood HSP in Taiwan, with an incidence of 12.9 per 100 000 children, occurs commonly in autumn and winter; and at the age of 5 to 6 yr. The characteristics presented in this study may provide valuable data for understanding and further studies of HSP.
Henoch–Schönlein purpura (HSP) is a kind of systemic small vessel vasculitis characterized by non-thrombocytopenic palpable purpura, arthritis, bowel angina and haematuria/proteinuria. Although the real pathogenesis is still unknown, it is well-recognized as a specific clinicopathological entity by the vascular deposition of immunoglobulin A (IgA)-containing immune complexes and elevated serum IgA levels. This disease is usually self-limiting. Involvement of internal organs such as kidney, intestine and central nervous system is a major complication [1–3]. Although HSP is not uncommon in children, there are few large-scale epidemiological studies of childhood HSP, especially nationwide surveys [4–6]. Like most previous studies that have been limited to hospital-based or regional-based sampling data, the latest survey conducted in the West Midlands, UK has shown the annual incidence of HSP to be around 20.4 per 100 000 children aged <17 yr [7].
In Taiwan, the National Health Insurance (NHI) programme—a compulsory social insurance programme enrolling the entire population—has been established and operating since 1995. In this system people living in the Taiwan area are provided with the right to equal opportunity of access to healthcare services including disease prevention, clinical care, hospitalization, residential care and social rehabilitation. Up to June 2002 there had been 21 750 489 individuals enrolled in the NHI system with a coverage rate of around 96%, meaning that the medical records of 96% of residents can be obtained from the NHI system [8, 9].
Using the data from Bureau of the NHI (BNHI), the aim of our study was to ascertain the incidence of childhood HSP and to identify other epidemiological characteristics such as age distribution, gender differences and seasonal variations in HSP among children in Taiwan.
Methods
Data source
The data used in this study were derived from research database of the BNHI, from 1 January 1999 to 31 December 2002. The BNHI data contain every medical claim record (in-patient care and out-patient care) including ID number, gender, date of birth, date of visit, length of hospitalization and International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM) diagnosis. The medical institutions in Taiwan are classified into four levels: clinics, district hospitals, regional hospitals and academic hospitals. During the study period, around 92–93% of these medical institutions located in the Taiwan area including Taiwan Island, Penghu Island, Kinmen Island, Matsu Island and other small islands belonging to the Republic of China (ROC) have participated in the NHI programme [8]. The demographic data on age and gender relating to the child population of the Taiwan area were obtained from the Department of Statistics, Ministry of the Interior, ROC [10].
This is an epidemiological study and the data were derived from insurance claim database. Therefore, ethical approval and informed patient consent were not required for this study.
Patient selection
Children (<17 yr of age) ever diagnosed with HSP (ICD-9-CM, 287.0) during the study period were included as the study population. The diagnosis of HSP confirmed by physicians in Taiwan, including paediatricians, family doctors and dermatologists, was based on the criteria defined by the American College of Rheumatology (ACR) [11], which require individuals to have at least two of four criteria: age less than or equal to 20 yr at disease onset, palpable purpura without thrombocytopenia, acute abdominal pain or gastrointestinal bleeding and biopsy showing granulocytes in the walls of small arterioles or venules. Physicians were able to distinguish HSP from other forms of vasculitis according to these criteria with a sensitivity of 87.1% and a specificity of 87.7% [11]. Each patient in the BNHI database is assigned a unique ID number, which allows identification of patients with multiple medical visits. For patients with multiple medical visits including clinic services and hospitalization, the date of the first visit was recorded as the date of disease onset.
Statistical analysis
The length of hospitalization and the number of cases in each season are presented as mean ± standard deviation. These descriptive continuous data were compared using Student's t-test. We also compared the categorical variables and proportions by using the χ2 test and the univariate regression test. Statistical significance was considered as a value of P<0.05. These statistical analyses were conducted using SPSS 10.0 software.
Results
Overall
During the study period, children younger than 17 yr of age accounted for 21.7% of the whole population, and 97.7% of them were Chinese [10]. A total of 2759 children with the diagnosis of HSP were identified from the database of BNHI over 4-yr study period. The annual incidence ranged from 11.8 to 13.4 per 100 000 children <17 yr old with an overall incidence of 12.9 per 100 000. In the treatment of HSP, hospitalization is not always necessary; 1118 (40.5%) patients in this study had ever been hospitalized. For those who had histories of hospitalization, the mean duration of admission was 5.1 days. We found that the rate of hospitalization for childhood HSP was statistically increasing year by year (1999: 35.7%; 2000: 40%; 2001: 42.3%; 2002: 44.9%; P = 0.011).
Age and gender variation
Figures 1A and B show the distribution of number of cases and annual incidence according to patients' age. Most cases occurred between 3 and 10 yr (73.9% of total cases). The number and incidence had a similar trend that was increasing gradually from 0 to 6 yr, and the peak incidence by year of age was 5 to 6 yr old (25.4% of total cases, annual incidence: 26.6 per 100 000 children aged 5 yr, 27.9 per 100 000 children aged 6 yr). After the age of 7 yr, the number of cases and incidence decreased. During the study period, there were 1454 males and 1305 females identified with a male-to-female ratio of 1.11. As shown in Fig. 1B, before the age of 10 yr the annual incidence in males exceeded that in females for most ages. In contrast, the incidence of females became slightly higher than that of males in the older ages. According to this trend, we divided children into two groups: a younger group (children aged <10 yr) and an older group (children aged ≥10 yr but <17 yr). The annual incidence of males was higher than that of females in the younger group (17.9 vs 16.4 per 100 000 children, P = 0.04), and was lower in older group (6.5 vs 7.8 per 100 000 children, P = 0.02). The rate of hospitalization of male patients was higher than female patients (43.0 vs 37.7%, P = 0.0046). However, the overall annual incidence (13.0 vs 12.7 per 100 000 children, P = 0.49) and length of hospitalization (5.0 ± 4.6 vs 5.3 ± 4.3 days, P = 0.27) between males and females was not significantly different.
Seasonal variation
Figure 2 showed that HSP could occur year-round. In this study, all 4 years showed a similar seasonal pattern (Fig. 2A). The number of cases deceased after March, stayed low in June and July, and increased from August (Fig. 2B). If we divided the year into four seasons, HSP occurred more commonly in autumn and winter (September to February) than in spring and summer (March to August) (388.8 ± 37.0 vs 301.0 ± 24.5 cases, P = 0.008).
Discussion
Because the disease course of HSP is usually benign and self-limited, hospitalization is not always indicated. Using the data from BNHI, which contains records of both hospitalization and clinic services, this study provided general population-based and nationwide epidemiological information on childhood HSP in Taiwan from January 1999 to December 2002. The result showed that the hospitalization rate for HSP in Taiwan was 40.5%. Therefore, if we only use the more easily available hospital discharge data to investigate epidemiology, 60% of patients will be missed.
The annual incidence of HSP was recently studied by Gardner-Medwin et al. [7]. Using questionnaires in one region of the United Kingdom, they found the estimated annual incidence from 1996 to 1998 to be 20.4 per 100 000 children <17 yr of age. Asian children accounting for 10% of the study population had a higher incidence (24.0 per 100 000) than White and Black. Among them, most were Indian (53%) followed by Pakistani (33%) and Bangladeshi (7%). Only 3% of these Asian children were Chinese [7]. In our study, 97.7% of the study population was Chinese [10]. Our findings suggested a lower annual incidence of HSP in children younger than 17 yr of age (12.9 per 100 000) than had been previously estimated. The differences in the annual incidence rates between these two studies may be due to different ethnic origin and/or having different environmental factors, and different ascertainment of cases (questionnaire data vs insurance claim data) [7].
The change of payment system for NHI [8] and more easily available medical facilities may affect physicians' behaviour and may have contributed to the increasing hospitalization rate during the 4-yr study period. Another possibility is that the number of severe cases increased gradually. In Taiwan, it is suggested that HSP children with vital organ involvement and those who suffer from a more refractory disease course are hospitalized for further management [12]. In order to confirm this speculation, it is necessary to perform further studies such as assessment of the complications of these in-patients.
The male to female ratios of childhood HSP varied among different studies. Most series have described a higher incidence in males (the ratio ranged from 1.2 to 1.6) [6, 7, 12, 13], while some studies have shown a female predominance [14, 15]. These variations may be attributable to the small sample sizes in the majority of previous studies and different time frames, races and geographical areas from which the data were recorded and analysed. In this study, the overall male to female ratio was 1.11. The annual incidence in males was statistically higher than in females before the age of 10, and became lower with increasing age. This trend may indicate that younger male children and older female children are slightly more susceptible to HSP. Our study also revealed a higher hospitalization rate in male patients; however, the length of hospitalization in males and females was not different. Therefore, further study is needed to ascertain whether the disease course in male patients is more complicated and severe than in female patients.
In the present study, case numbers and annual incidence increased as patient age increased from 0 to 6 yr; after 7 yr of age the number and incidence decreased gradually. In Taiwan, the rate of attendance at kindergartens of children aged from 0 to 6 yr also gradually increases (0–2 yr old, 0.2%; 3–4 yr old, 7.4%; 4–5 yr old, 25.7%; 5–6 yr old, 37.4%; from 1999 to 2002) [16]. The close contact with schoolmates among younger children is suggested to be a relevant risk factor for many infectious diseases, especially respiratory infections [17–19]. The age-related epidemiological variations in HSP, together with the disease clustering in autumn and winter and the histories of preceding upper respiratory infections recorded in many HSP patients [1, 20, 21], provide clues to the possibility that HSP is infection-related.
There are some limitations in this study. Thorough medical records for each patient are not available from the BNHI research database. Therefore, the clinical presentations such as the percentage of cases with renal involvement, complications, therapeutic strategies and prognosis in these patients are unknown. Patients may visit clinics several times and/or be hospitalized in the same episode, without medical records; we also could not conclude the recurrence rate and the interval between two episodes of childhood HSP.
In conclusion, this study provides the first general population-based nationwide estimations for childhood HSP. The annual incidence in Taiwan is lower than in the recent report from the United Kingdom [7]. HSP affects mainly children between 3 and 10 yr of age, with a peak incidence in children aged 5 to 6 yr. Like previous series, most cases occur in autumn and winter. These epidemiological characteristics are valuable not only for disease prevention but also for the further aetiological studies of HSP.
We thank the National Health Research Institutes for providing data. This study is based in part on data from the National Health Insurance Research Database provided by the Bureau of National Health Insurance, Department of Health and managed by National Health Research Institutes, ROC. The interpretation and conclusions contained herein do not represent those of Bureau of National Health Insurance, Department of Health or National Health Research Institutes.
The authors have declared no conflicts of interest.
References
Kawasaki Y, Suzuki J, Nozawa R, Suzuki S, Suzuki H. Efficacy of methylprednisolone and urokinase pulse therapy for severe Henoch–Schonlein nephritis.
Bissonnette R, Dansereau A, D'Amico P, Pateneaude JV, Paradis J. Perforation of large and small bowel in Henoch–Schonlein purpura.
Stewart M, Savage JM, Bell B, McCord B. Long term renal prognosis of Henoch–Schonlein purpura in an unselected childhood population.
Watts RA, Jolliffe VA, Grattan CE, Elliott J, Lockwood M, Scott DG. Cutaneous vasculitis in a defined population—clinical and epidemiological associations.
Saulsbury FT. Henoch–Schonlein purpura in children. Report of 100 patients and review of the literature.
Gardner-Medwin JM, Dolezalova P, Cummins C, Southwood TR. Incidence of Henoch–Schonlein purpura, Kawasaki disease, and rare vasculitides in children of different ethnic origins.
Bureau of National Health Insurance, ROC.
Department of Statistics, Ministry of the Interior, ROC.
Mills JA, Michel BA, Bloch DA et al. The American College of Rheumatology 1990 criteria for the classification of vasculitis.
Lin SJ, Huang JL. Henoch–Schonlein purpura in Chinese children and adults.
Dawod ST, Akl KF. Henoch–Schoenlein syndrome in Qatar: the effects of steroid therapy and paucity of renal involvement.
Calvino MC, Llorca J, Garcia-Porrua C, Fernandez-Iglesias JL, Rodriguez-Ledo P, Gonzalez-Gay MA. Henoch–Schonlein purpura in children from northwestern Spain: a 20-year epidemiologic and clinical study.
al Harbi NN. Henoch–Schoenlein syndrome in children: experience from southern part of Saudi Arabia.
Department of Statistics, Ministry of the Education, ROC.
Khetsuriani N, Bisgard K, Prevots DR et al. Pertussis outbreak in an elementary school with high vaccination coverage.
Cockerill FR 3rd, MacDonald KL, Thompson RL et al. An outbreak of invasive group A streptococcal disease associated with high carriage rates of the invasive clone among school-aged children.
Chang LY, King CC, Hsu KH et al. Risk factors of enterovirus 71 infection and associated hand, foot, and mouth disease/herpangina in children during an epidemic in Taiwan.
Blanco R, Martinez-Taboada VM, Rodriguez-Valverde V, Garcia-Fuentes M, Gonzalez-Gay MA. Henoch–Schonlein purpura in adulthood and childhood: two different expressions of the same syndrome.
Author notes
Department of Pediatrics, National Taiwan University Hospital, College of Medicine, National Taiwan University, and 1Department of Biostatistics, Clinical Protocol Office, Koo Foundation Sun Yat-Sen Cancer Center, Taipei, Taiwan.
Comments