- Email: [email protected]
Newly diagnosed idiopathic thrombocytopenic purpura in childhood: an observational study
ARTICLES
Newly diagnosed idiopathic thrombocytopenic purpura in childhood: an observational study Thomas Kühne, Paul Imbach, Paula H B Bolton-Maggs, Willi Berchtold, Victor Blanchette, George R Buchanan, for the Intercontinental Childhood ITP Study Group
Summary Background Diagnosis and management of idiopathic thrombocytopenic purpura (ITP) have been based primarily on expert opinion and practice guidelines rather than on evidence. We have used a registry to prospectively survey the presenting features and the diagnostic evaluation and management practices used for children with ITP worldwide. Methods We used the Intercontinental Childhood ITP Registry which had been widely advertised. 209 physicians from 136 institutions in 38 countries participated by submitting data for each of their newly diagnosed patients. Data from 2031 children with ITP was registered between June, 1997, and May, 2000, and we analysed 6-month follow-up data from 1496 children. Findings There was a peak in occurrence of childhood ITP during spring and a nadir in the autumn. Mean initial platelet count was 15·4⫻109/L (SD 19·7). 1447 (73%) of 1976 children were admitted to hospital. Initial management consisted of no drug treatment in 612 (31%), intravenous immunoglobulin in 576 (29%), corticosteroids in 651 (33%), or both in 137 (7%) patients. Intracranial haemorrhage was reported in two patients. Interpretation The variable approaches to management of childhood ITP demonstrate the need for prospective clinical trials, which should be feasible within such a study group. Lancet 2001; 358: 2122–25
Divisions of Oncology/Haematology, University Children’s Hospital, Postfach, Römergasse 8, CH-4005 Basel, Switzerland (T Kühne MD, P Imbach MD); Royal Liverpool Children’s Hospital, Liverpool, UK (P H B Bolton-Maggs FRCPCH); University of Applied Sciences of Aargau, Brugg, Switzerland (W Berchtold PhD); The Hospital for Sick Children, Toronto, Canada (V Blanchette FRCP); and The University of Texas Southwestern Medical Center at Dallas, Dallas, Texas, USA (George R Buchanan MD) Correspondence to: Dr Thomas Kühne (e-mail: [email protected])
2122
Introduction Management of childhood idiopathic thrombocytopenic purpura (ITP) is based more on opinion and local practices than on evidence and has been the subject of fierce debate for decades. Although practice guidelines have been developed and published,1,2 compliance with the guidelines is poor3,4 and the validity of one set of guidelines has been openly challenged.5 A network of shared consensus protocols and trials, such as is in place for malignant diseases, does not exist for this benign condition. The lack of an evidence-based approach is a disadvantage for patients, with diverse medical and socioeconomic consequences, reflected by the array of simple and sophisticated diagnostic procedures and treatments, from no therapy to multiple expensive platelet-enhancing agents with various adverse effects.6,7 The natural history of the disease has been described in several retrospective studies.2 However, prospective data have only been obtained from randomised clinical trials with short follow-up and fairly small numbers of patients. We aimed to establish an international network of physicians involved in the diagnosis and management of children with ITP, to prospectively collect clinical data regarding the natural history and management of childhood ITP, and to compare results with previously published reports.
Patients and methods The Intercontinental Childhood ITP Registry was established in June 1997 by the Intercontinental Childhood ITP Study Group to prospectively investigate the pathophysiology, clinical course, management, and outcome of children with ITP. The group advertised the Registry on its web-page (www.unibas.ch/itpbasel, accessed Dec 7, 2001), at national and international haematology meetings, in haematology journals, and by regularly mailed newsletters to prospective participants. Each patient older than 4 months and younger than 16 years with newly diagnosed ITP seen at a participating centre was eligible for registration. However, participation was voluntary and therefore it is assumed that not each consecutive patient was registered. Participating physicians registered patients by completing and submitting an entry sheet. This first single-page questionnaire requested the address of the physician and data from the patient including date of birth, sex, date of diagnosis, platelet count at diagnosis, whether a bone marrow aspirate had been done, whether the patient had been admitted to hospital, and whether treatment (intravenous immunoglobulin, corticosteroids, or both) had been given or not. The entry sheet was submitted as soon as possible after the diagnosis of ITP to the coordinating centre in Basel, Switzerland. After 6 months the coordinating centre faxed a second questionnaire to each physician to obtain follow-up data, when it was possible to distinguish the acute from the chronic form of ITP, defined by a persistence of thrombocytopenia (<150⫻109/L) for 6 months or longer. The second questionnaire requested the follow-up platelet count, and
THE LANCET • Vol 358 • December 22/29, 2001
For personal use. Only reproduce with permission from The Lancet Publishing Group.
ARTICLES
Results From June 1, 1997, to May 31, 2000, 2190 children were enrolled by 209 physicians from 136 institutions in 38 countries. We included data from 2031 children with newly diagnosed ITP. The remaining 159 (7%) children were ineligible for analysis because they were aged less than 4 months or 16 years old or older. Of eligible patients, 6-month follow-up data on 1510 (74%) children were received. Of those 14 (0·9%) were excluded because of incomplete data, thus we analysed follow-up data from 1496 (74%) patients. Mean time from diagnosis to the follow-up report was 170·6 days (95% CI 166·3–174·8). We arbitrarily subdivided participating countries into eight groups representing continental regions, to detect common or differing patterns in the data (table). The median number of registered patients per institution was 4 (range 1–121). 120 (57%) of 209 physicians registered less than six patients, 59 (28%) registered 6–20 patients, and 30 (14%) registered more than 20 patients. Eight physicians (4%) registered more than 50 patients, accounting for a total of 653 (30%) patients. Participating physicians were mainly paediatric haematologists or oncologists (86%) holding a hospital based full-time academic position (66%) or a hospital position not affiliated with a university (20%). Continent
Number of countries
Number of Number of Total number institutions patients of patients (range)
North America South America Western Europe Eastern Europe Africa Western Asia Eastern Asia Australia Total
3 2 10 12 2 3 5 1 38
23 9 37 21 2 10 29 5 136
2–117 4–56 1–60 1–65 2–29 1–121 1–90 1–34 1–121
536 215 491 369 31 252 215 81 2190
*The countries of each group are as follows—North America: Canada, Jamaica, and USA; South America: Argentina and Brazil; Western Europe: Austria, Germany, Italy, Malta, Portugal, Scotland, Sweden, Switzerland, The Netherlands, and UK; Eastern Europe: Belarus, Bulgaria, Czech Republic, Croatia, Greece, Hungary, Poland, Romania, Russia, Slovakia, Turkey, and Yugoslavia; Africa: Egypt and South Africa; Western Asia: Iran, Israel, and Kyrgyzstan; Eastern Asia: China, India, Japan, Korea, and Vietnam; and Australia.
Number of countries, institutions, and patients participating in the ITP Registry*
THE LANCET • Vol 358 • December 22/29, 2001
Girls Boys
160
Number of patients
140 120 100 80 60 40 20 0
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 Age (years)
Figure 1: Age of children with ITP
The mean age (SD) of all registered children was 5·7 (4·1) years (figure 1). Boys had a mean age of 5·5 (4·0) and girls had a mean age of 5·9 (4·2). In the age groups less than 1, 1–5, and older than 5 years there were 149 (7%), 946 (47%), and 920 (46%) patients, respectively. There were 1104 boys (54·8%, 95% CI 52·6–57·0%) and 909 girls (45·2%, 43·0–47·4%)—ITP was diagnosed with a peak incidence in spring/early summer and a nadir in autumn (figure 2). The initial mean platelet count of all children was 15·4⳯109/L (SD 19·7). At 6 months the mean platelet count was 209·7⫻109/L (125·9). Chronic ITP was seen in 31% (95% CI 28·9–33·6%) of children in equal numbers of girls and boys. Mean age was 6·9 (4·3) years in children with chronic ITP, and 5·1 (3·8) years in children with acute ITP. In those aged less than 1, 1–5, or older than 5 years, 5%, 38%, and 57% of children, respectively, had chronic ITP. Children with chronic ITP had a mean platelet count at initial diagnosis of 18·4⫻109/L (21·4) and those with acute ITP of 13·8⫻109/L (19·0). A bone marrow aspirate was obtained in 985 (50%) of the 1984 children for whom data was available. Bone marrow aspiration was done in 234 (39%) of 606 untreated children and in 751 (56%) of 1339 treated children. 214 (38%) of 565 children treated with intravenous immunoglobulin, 444 (69%) of 644 children treated with corticosteroids, and 93 (72%) of 130 children treated with intravenous immunoglobulin and corticosteroids had a bone marrow aspirate. Number of patients per month
established whether the patient had had bleeding during the 6-month period, whether intracranial haemorrhage (ICH) had occurred, and whether the patient had received additional platelet enhancing treatment or not (intravenous immunoglobulin, corticosteroids, or both) during the first 6 months. Diagnosis of ITP was based on a platelet count of less than 150⫻109/L, a normal haemoglobin concentration, white blood cell count, and blood smear (except for thrombocytopenia) and absence of underlying conditions such as HIV infection, systemic lupus erythematosus, and lymphoproliferative disorders. Bone marrow aspiration was optional. The Registry had been approved by institutional review boards according to local requirements. Data in this study are multi-level. Patients are clustered within institutions, which are within countries, which have then been grouped in geographical regions. Thus, there are four levels of cluster. Additionally, the varying number of patients per institution, and institutions per country make the interpretation of statistical analyses difficult. Therefore the data will be presented in a descriptive form without statistical analysis.
220 200 180 160 0 1
2
3
4
5
6 7 Month
8
9
10 11 12
Figure 2: Seasonal occurrence of ITP To analyse the incidence of diagnosis by month, data from southern hemisphere patients were transformed as follows: month–north=(month–south + 6); if month–north exceeds 1 year, 12 is subtracted. Thus, “July” in the southern hemisphere is equivalent to “January” in the northern hemisphere.
2123
For personal use. Only reproduce with permission from The Lancet Publishing Group.
ARTICLES
A platelet count of more than 150⳯109/L at 6 months was achieved in 68% of untreated children, in 73% of children on intravenous immunoglobulin (303/415 children), and in 66% (346/527 children) of children on corticosteroids. Intracranial haemorrhage was reported in two children among the 1496 children for whom 6-month follow-up forms were submitted. One patient was a 1·5year-old girl in whom an ICH occurred 4 months after the diagnosis of ITP with a favourable outcome without neurological sequelae. The other patient was a 15-yearold girl with an initial platelet count of 16⳯109/L and initially treated with corticosteroids. The time when ICH occurred and the outcome of the latter patient remains unknown.
IVIG+CS CS IVIG No treatment
100 90 Proportion of patients (%)
80 70 60 50 40 30
Discussion
20 10 0 ica
er
rth
No
Am
h
ut
So
ica
er
Am W
rn
te
es
e
op
n
er
st
Ea
a
e
op
r Eu
r Eu
ia
ric
Af
t
n er
es
W
ia
As
n er
As
st
Ea
lia
ra
t us
A
Figure 3: Management of children with ITP in eight geographical groups IVIG=intravenous immunoglobulin, CS=corticosteroids. Management consisted of no treatment and treatment—ie, intravenous immunoglobulin, corticosteroids, or both.
Of the 1995 patients for whom data on hospital admission was available, 1447 (73%) were admitted to hospital, including 78% of those with an initial platelet count of less than 20⳯109/L. The mean initial platelet count in admitted and non-admitted children was 11·6 (SD 14·2), and 25·2⳯109/L (26·6) respectively. Untreated and treated patients were admitted in 47% and 85% of cases, respectively. Patients on intravenous immunoglobulin, corticosteroids, and intravenous immunoglobulin plus corticosteroids were admitted in 91%, 77%, and 95%, respectively. Of the 1976 children for whom data was available on treatment, 612 (31%) did not receive drug treatment, and 1364 (69%) did, either with intravenous immunoglobulin (576, 29%), corticosteroids (651, 33%), or intravenous immunoglobulin plus corticosteroids (137, 7%). 107 children (7%) received either intravenous immunoglobulin or corticosteroids or both despite an initial platelet count of more than 20⫻109/L. Figure 3 shows how treatment varied in different parts of the world. The mean initial platelet count of untreated children was 26·6⫻109/L (SD 26·8), and of children receiving therapy 10·4⫻109/L (11·5). The mean initial platelet count of patients treated with intravenous immunoglobulin, corticosteroids, and intravenous immunoglobulin plus corticosteroids was 8·0⫻109/L (7·8), 13·3 (15·6), and 8·1 (11·0), respectively. During the first 6 months, antiRh(D) was administered initially in five children and as additional therapy during the 6-month follow-up period in 27 (2%) children. The percentage of children receiving or not receiving initial drug treatment at the time of the diagnosis of ITP was in acute ITP 68% and 32%, respectively, and in chronic ITP 68%, and 33%, respectively. Intravenous immunoglobulin was given in children with acute ITP in 49% (303/620 children), and in children with chronic ITP in 40% (112/278 children). Corticosteroids were administered in children with acute ITP in 51% (317/620 children) and in children with chronic ITP in 60% (166/278 children).
2124
The Intercontinental Childhood ITP Registry serves as a network for facilitating collaborative research in ITP with the advantage of its prospective nature involving investigators in many countries, making it a unique investigative resource. The small numbers of patients recruited by 74% of institutions may be due to inconsistent registration, but is more likely caused by the relative rarity of the disease, as ITP occurs only in about one in 25 000 children yearly.8 Several features of childhood ITP have been confirmed, such as age distribution with a peak between 1 and 6 years,9 presenting platelet count of less than 20⫻109/L, and the fact that most children remit rapidly with or without drug therapy. Additionally, unexpected results have been observed, such as details of outcome and sex distribution with a higher number of boys, as previously described by our group10 and others.11 Whether a bone-marrow aspirate should be undertaken or not has been a matter of long-lasting debate. It has been shown that isolated thrombocytopenia does not necessarily represent an indication for doing a bone marrow aspirate,12,13 but if atypical features of ITP are present or if corticosteroids are considered as initial therapy, a bone marrow aspirate is usually recommended.1,2,14,15 The frequency of hospital admission reported in this survey reflects the recommendation by the American guidelines, which state that it is appropriate for a child with severe, life-threatening bleeding regardless of the platelet count and for a child with a platelet count of less than 20⳯109/L.2 In an assessment of the UK practice guidelines, 83% of patients were admitted,3 and in a survey of diagnostic and management strategies, age and type of bleeding (dry or wet haemorrhage) directed admission—ie, 85% of respondents reported that they would admit a 5-year-old patient with wet purpura, but only 41% admitted patients with dry purpura.4 Thus, further prospective studies are necessary to collect clinical data on morbidity related to bleeding. This requirement is addressed in a follow-up registry to be opened soon. The outcome of children with ITP at 6 months seemed not to be affected by initial treatment strategy. This observation confirms previously published prospective trials.16–18 In a retrospective single-centre study, 51 children with ITP did not receive platelet-enhancing treatment and only four received short-term corticosteroids because of extensive mucosal bleeding.19 The outcome was similar to that of treated children, with chronic ITP occurring in 13% of the patients, and without the occurrence of life-threatening bleeding. The high incidence of chronic ITP in our patients might represent an artefact of under-reporting of patients with rapidly resolving acute ITP at 6 months. Children who developed
THE LANCET • Vol 358 • December 22/29, 2001
For personal use. Only reproduce with permission from The Lancet Publishing Group.
ARTICLES
chronic ITP had higher initial platelet counts at diagnosis and were older than children with the acute form. By contrast with pubished reports,2 we did not report more girls than boys with chronic ITP. Intracranial haemorrhage was reported only twice among 1496 children, of whom 69% were treated, which is far below the often quoted incidence of 1%2,20,21 suggesting a very low risk of this complication during the first 6 months after diagnosis. However, this result must be interpreted cautiously due to incomplete follow-up data (74%) and the possibility that some patients with intracranial haemorrhage at diagnosis might not have been registered even if such patients presented to participating institutions. Lilleyman suggested that the risk of intracranial haemorrhage increases with time, and that it is closer to 0·1–0·2% in the first few days after diagnosis.8,22 Most children with intracranial haemorrhage were reported in retrospective studies, whereas only one child with intracranial haemorrhage has been described in a prospective study.23 The key value of the Registry is the establishment of an international network, with the capability of accruing large numbers of patients with ITP, and confirming or refuting results derived by smaller studies. As a result of this collaborative network future studies will address the hypothesis that major haemorrhage, defined by intracranial or other overt internal or mucous membrane bleeding resulting in anaemia or requiring local treatment,24 is uncommon in children with ITP and only rarely encountered after diagnosis and initial therapy. Data such as these will serve as a basis for conducting clinical trials answering the question—which subgroup of children with ITP requires drug treatment. Contributors T Kühne designed the registry, took part in data analysis, and contributed to the writing of the report. W Berchtold took part in data analysis. P Imbach, P H B Bolton-Maggs, V Blanchette, and G R Buchanan reviewed the study report and contributed to the writing of the report.
4
5
6
7
8 9 10
11 12
13
14
15
16
17
18
Acknowledgments We thank all our colleagues, who have generously submitted data of their patients, and all the nurses and staff of the hospitals who were engaged in the Registry. We thank Heide Luther, administrator, and Ursula Längin, past-administrator for their assistance with the administration of the data and for preparation of the paper. This study was supported through an unrestricted grant by Norvartis Pharma AG.
References 1
2
3
Eden OB, Lilleyman JS. Guidelines for management of idiopathic thrombocytopenic purpura: the British Paediatric Haematology Group. Arch Dis Child 1992; 67: 1056–58. George JN, Woolf SH, Raskob GE, et al. Idiopathic thrombocytopenic purpura: a practice guideline developed by explicit methods for The American Society of Hematology. Blood 1996; 88: 3–40. Bolton-Maggs PHB, Moon I. Assessment of UK practice for management of acute childhood idiopathic thrombocytopenic purpura against published guidelines. Lancet 1997; 350: 620–23.
THE LANCET • Vol 358 • December 22/29, 2001
19
20
21 22 23
24
Vesely S, Buchanan GR, Cohen A, Raskob G, George J. Self-reported diagnostic and management strategies in childhood idiopathic thrombocytopenic purpura: results of a survey of practicing pediatric hematology/oncology specialists. J Pediatr Hematol Oncol 2000; 22: 55–61. Buchanan GR, de Alarcon PA, Feig SA, et al. Acute idiopathic thrombocytopenic purpura—management in childhood. Blood 1997; 89: 1464–65. Jayabose S, Mahmoud M, Levendoglu-Tugal O, et al. Corticosteroid prophylaxis for neurologic complications of intravenous immunoglobulin G therapy in childhood immune thrombocytopenic purpura. J Pediatr Hematol Oncol 1999; 21: 514–17. Gaines AR. Acute onset hemoglobinemia and/or hemoglobinuria and sequelae following Rho(D) immune globulin intravenous administration in immune thrombocytopenic purpura patients. Blood 2000; 95: 2523–29. Lilleyman JS. Management of childhood idiopathic thrombocytopenic purpura. Br J Haematol 1999; 105: 871–75. Lusher JM, Zuelzer WW. Idiopathic thrombocytopenic purpura in childhood. J Pediatr 1966; 68: 971–79. Kühne T, Berchtold W, Be TV, Binh TV, Imbach P. Ethnicity and environment may affect the phenotype of immune thrombocytopenic purpura in children. Pediatr Res 2000; 48: 374–79. Robb LG, Tiedeman K. Idiopathic thrombocytopenic purpura: predictors of chronic disease. Arch Dis Child 1990; 65: 502–06. Dubansky AS, Boyett JM, Falletta J, et al. Isolated thrombocytopenia in children with acute lymphoblastic leukemia: a rare event in a Pediatric Oncology Group Study. Pediatrics 1989; 84: 1068–71. Calpin C, Dick P, Poon A, Feldman W. Is bone marrow aspiration needed in acute childhood idiopathic thrombocytopenic purpura to rule out leukemia? Arch Pediatr Adolesc Med 1998; 152: 345–47. Halperin DS, Doyle JJ. Is bone marrow examination justified in idiopathic thrombocytopenic purpura? Am J Dis Child 1988; 142: 508–11. Reid MM. Bone marrow examination before steroids in thrombocytopenic purpura or arthritis. Acta Paediatr 1992; 81: 1052–53. Sartorius JA. Steroid treatment of idiopathic thrombocytopenic purpura in children: preliminary results of a randomized cooperative study. Am J Pediatr Hematol Oncol 1984; 6: 165–69. Buchanan GR, Holtkamp CA. Prednisone therapy for children with newly diagnosed idiopathic thrombocytopenic purpura. A randomized clinical trial. Am J Pediatr Hematol Oncol 1984; 6: 355–61. Blanchette VS, Luke B, Andrew M, et al. A prospective, randomized trial of high-dose intravenous immune globulin G therapy, oral prednisone therapy, and no therapy in childhood acute immune thrombocytopenic purpura. J Pediatr 1993; 123: 989–95. Dickerhoff R, Von Ruecker A. The clinical course of immune thrombocytopenic purpura in children who did not receive intravenous immunoglobulins or sustained prednisone treatment. J Pediatr 2000; 137: 629–32. Aronis S, Platokouki H, Mitsiki A, Haidus S, Constantopoulos A. Seventeen years of experience with chronic idiopathic thrombocytopenic purpura in childhood. Pediatr Hematol Oncol 1994; 11: 487–98. Blanchette VS, Turner C. Treatment of acute idiopathic thrombocytopenic purpura. J Pediatr 1985; 108: 326–27. Lilleyman JS. Intracranial hemorrhage in idiopathic thrombocytopenic purpura. Arch Dis Child 1994; 71: 251–53. Imbach P, Wagner HP, Berchtold W, et al. Intravenous immunoglobulin versus oral corticosteroids in acute immune thrombocytopenic purpura in childhood. Lancet 1985; 2: 464–68. Medeiros D, Buchanan GR. Major hemorrhage in children with idiopathic thrombocytopenic purpura: immediate response to therapy and long-term outcome. J Pediatr 1998; 133: 334–39.
2125
For personal use. Only reproduce with permission from The Lancet Publishing Group.
Newly diagnosed idiopathic thrombocytopenic purpura in childhood: an observational study Thomas Kühne, Paul Imbach, Paula H B Bolton-Maggs, Willi Berchtold, Victor Blanchette, George R Buchanan, for the Intercontinental Childhood ITP Study Group
Summary Background Diagnosis and management of idiopathic thrombocytopenic purpura (ITP) have been based primarily on expert opinion and practice guidelines rather than on evidence. We have used a registry to prospectively survey the presenting features and the diagnostic evaluation and management practices used for children with ITP worldwide. Methods We used the Intercontinental Childhood ITP Registry which had been widely advertised. 209 physicians from 136 institutions in 38 countries participated by submitting data for each of their newly diagnosed patients. Data from 2031 children with ITP was registered between June, 1997, and May, 2000, and we analysed 6-month follow-up data from 1496 children. Findings There was a peak in occurrence of childhood ITP during spring and a nadir in the autumn. Mean initial platelet count was 15·4⫻109/L (SD 19·7). 1447 (73%) of 1976 children were admitted to hospital. Initial management consisted of no drug treatment in 612 (31%), intravenous immunoglobulin in 576 (29%), corticosteroids in 651 (33%), or both in 137 (7%) patients. Intracranial haemorrhage was reported in two patients. Interpretation The variable approaches to management of childhood ITP demonstrate the need for prospective clinical trials, which should be feasible within such a study group. Lancet 2001; 358: 2122–25
Divisions of Oncology/Haematology, University Children’s Hospital, Postfach, Römergasse 8, CH-4005 Basel, Switzerland (T Kühne MD, P Imbach MD); Royal Liverpool Children’s Hospital, Liverpool, UK (P H B Bolton-Maggs FRCPCH); University of Applied Sciences of Aargau, Brugg, Switzerland (W Berchtold PhD); The Hospital for Sick Children, Toronto, Canada (V Blanchette FRCP); and The University of Texas Southwestern Medical Center at Dallas, Dallas, Texas, USA (George R Buchanan MD) Correspondence to: Dr Thomas Kühne (e-mail: [email protected])
2122
Introduction Management of childhood idiopathic thrombocytopenic purpura (ITP) is based more on opinion and local practices than on evidence and has been the subject of fierce debate for decades. Although practice guidelines have been developed and published,1,2 compliance with the guidelines is poor3,4 and the validity of one set of guidelines has been openly challenged.5 A network of shared consensus protocols and trials, such as is in place for malignant diseases, does not exist for this benign condition. The lack of an evidence-based approach is a disadvantage for patients, with diverse medical and socioeconomic consequences, reflected by the array of simple and sophisticated diagnostic procedures and treatments, from no therapy to multiple expensive platelet-enhancing agents with various adverse effects.6,7 The natural history of the disease has been described in several retrospective studies.2 However, prospective data have only been obtained from randomised clinical trials with short follow-up and fairly small numbers of patients. We aimed to establish an international network of physicians involved in the diagnosis and management of children with ITP, to prospectively collect clinical data regarding the natural history and management of childhood ITP, and to compare results with previously published reports.
Patients and methods The Intercontinental Childhood ITP Registry was established in June 1997 by the Intercontinental Childhood ITP Study Group to prospectively investigate the pathophysiology, clinical course, management, and outcome of children with ITP. The group advertised the Registry on its web-page (www.unibas.ch/itpbasel, accessed Dec 7, 2001), at national and international haematology meetings, in haematology journals, and by regularly mailed newsletters to prospective participants. Each patient older than 4 months and younger than 16 years with newly diagnosed ITP seen at a participating centre was eligible for registration. However, participation was voluntary and therefore it is assumed that not each consecutive patient was registered. Participating physicians registered patients by completing and submitting an entry sheet. This first single-page questionnaire requested the address of the physician and data from the patient including date of birth, sex, date of diagnosis, platelet count at diagnosis, whether a bone marrow aspirate had been done, whether the patient had been admitted to hospital, and whether treatment (intravenous immunoglobulin, corticosteroids, or both) had been given or not. The entry sheet was submitted as soon as possible after the diagnosis of ITP to the coordinating centre in Basel, Switzerland. After 6 months the coordinating centre faxed a second questionnaire to each physician to obtain follow-up data, when it was possible to distinguish the acute from the chronic form of ITP, defined by a persistence of thrombocytopenia (<150⫻109/L) for 6 months or longer. The second questionnaire requested the follow-up platelet count, and
THE LANCET • Vol 358 • December 22/29, 2001
For personal use. Only reproduce with permission from The Lancet Publishing Group.
ARTICLES
Results From June 1, 1997, to May 31, 2000, 2190 children were enrolled by 209 physicians from 136 institutions in 38 countries. We included data from 2031 children with newly diagnosed ITP. The remaining 159 (7%) children were ineligible for analysis because they were aged less than 4 months or 16 years old or older. Of eligible patients, 6-month follow-up data on 1510 (74%) children were received. Of those 14 (0·9%) were excluded because of incomplete data, thus we analysed follow-up data from 1496 (74%) patients. Mean time from diagnosis to the follow-up report was 170·6 days (95% CI 166·3–174·8). We arbitrarily subdivided participating countries into eight groups representing continental regions, to detect common or differing patterns in the data (table). The median number of registered patients per institution was 4 (range 1–121). 120 (57%) of 209 physicians registered less than six patients, 59 (28%) registered 6–20 patients, and 30 (14%) registered more than 20 patients. Eight physicians (4%) registered more than 50 patients, accounting for a total of 653 (30%) patients. Participating physicians were mainly paediatric haematologists or oncologists (86%) holding a hospital based full-time academic position (66%) or a hospital position not affiliated with a university (20%). Continent
Number of countries
Number of Number of Total number institutions patients of patients (range)
North America South America Western Europe Eastern Europe Africa Western Asia Eastern Asia Australia Total
3 2 10 12 2 3 5 1 38
23 9 37 21 2 10 29 5 136
2–117 4–56 1–60 1–65 2–29 1–121 1–90 1–34 1–121
536 215 491 369 31 252 215 81 2190
*The countries of each group are as follows—North America: Canada, Jamaica, and USA; South America: Argentina and Brazil; Western Europe: Austria, Germany, Italy, Malta, Portugal, Scotland, Sweden, Switzerland, The Netherlands, and UK; Eastern Europe: Belarus, Bulgaria, Czech Republic, Croatia, Greece, Hungary, Poland, Romania, Russia, Slovakia, Turkey, and Yugoslavia; Africa: Egypt and South Africa; Western Asia: Iran, Israel, and Kyrgyzstan; Eastern Asia: China, India, Japan, Korea, and Vietnam; and Australia.
Number of countries, institutions, and patients participating in the ITP Registry*
THE LANCET • Vol 358 • December 22/29, 2001
Girls Boys
160
Number of patients
140 120 100 80 60 40 20 0
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 Age (years)
Figure 1: Age of children with ITP
The mean age (SD) of all registered children was 5·7 (4·1) years (figure 1). Boys had a mean age of 5·5 (4·0) and girls had a mean age of 5·9 (4·2). In the age groups less than 1, 1–5, and older than 5 years there were 149 (7%), 946 (47%), and 920 (46%) patients, respectively. There were 1104 boys (54·8%, 95% CI 52·6–57·0%) and 909 girls (45·2%, 43·0–47·4%)—ITP was diagnosed with a peak incidence in spring/early summer and a nadir in autumn (figure 2). The initial mean platelet count of all children was 15·4⳯109/L (SD 19·7). At 6 months the mean platelet count was 209·7⫻109/L (125·9). Chronic ITP was seen in 31% (95% CI 28·9–33·6%) of children in equal numbers of girls and boys. Mean age was 6·9 (4·3) years in children with chronic ITP, and 5·1 (3·8) years in children with acute ITP. In those aged less than 1, 1–5, or older than 5 years, 5%, 38%, and 57% of children, respectively, had chronic ITP. Children with chronic ITP had a mean platelet count at initial diagnosis of 18·4⫻109/L (21·4) and those with acute ITP of 13·8⫻109/L (19·0). A bone marrow aspirate was obtained in 985 (50%) of the 1984 children for whom data was available. Bone marrow aspiration was done in 234 (39%) of 606 untreated children and in 751 (56%) of 1339 treated children. 214 (38%) of 565 children treated with intravenous immunoglobulin, 444 (69%) of 644 children treated with corticosteroids, and 93 (72%) of 130 children treated with intravenous immunoglobulin and corticosteroids had a bone marrow aspirate. Number of patients per month
established whether the patient had had bleeding during the 6-month period, whether intracranial haemorrhage (ICH) had occurred, and whether the patient had received additional platelet enhancing treatment or not (intravenous immunoglobulin, corticosteroids, or both) during the first 6 months. Diagnosis of ITP was based on a platelet count of less than 150⫻109/L, a normal haemoglobin concentration, white blood cell count, and blood smear (except for thrombocytopenia) and absence of underlying conditions such as HIV infection, systemic lupus erythematosus, and lymphoproliferative disorders. Bone marrow aspiration was optional. The Registry had been approved by institutional review boards according to local requirements. Data in this study are multi-level. Patients are clustered within institutions, which are within countries, which have then been grouped in geographical regions. Thus, there are four levels of cluster. Additionally, the varying number of patients per institution, and institutions per country make the interpretation of statistical analyses difficult. Therefore the data will be presented in a descriptive form without statistical analysis.
220 200 180 160 0 1
2
3
4
5
6 7 Month
8
9
10 11 12
Figure 2: Seasonal occurrence of ITP To analyse the incidence of diagnosis by month, data from southern hemisphere patients were transformed as follows: month–north=(month–south + 6); if month–north exceeds 1 year, 12 is subtracted. Thus, “July” in the southern hemisphere is equivalent to “January” in the northern hemisphere.
2123
For personal use. Only reproduce with permission from The Lancet Publishing Group.
ARTICLES
A platelet count of more than 150⳯109/L at 6 months was achieved in 68% of untreated children, in 73% of children on intravenous immunoglobulin (303/415 children), and in 66% (346/527 children) of children on corticosteroids. Intracranial haemorrhage was reported in two children among the 1496 children for whom 6-month follow-up forms were submitted. One patient was a 1·5year-old girl in whom an ICH occurred 4 months after the diagnosis of ITP with a favourable outcome without neurological sequelae. The other patient was a 15-yearold girl with an initial platelet count of 16⳯109/L and initially treated with corticosteroids. The time when ICH occurred and the outcome of the latter patient remains unknown.
IVIG+CS CS IVIG No treatment
100 90 Proportion of patients (%)
80 70 60 50 40 30
Discussion
20 10 0 ica
er
rth
No
Am
h
ut
So
ica
er
Am W
rn
te
es
e
op
n
er
st
Ea
a
e
op
r Eu
r Eu
ia
ric
Af
t
n er
es
W
ia
As
n er
As
st
Ea
lia
ra
t us
A
Figure 3: Management of children with ITP in eight geographical groups IVIG=intravenous immunoglobulin, CS=corticosteroids. Management consisted of no treatment and treatment—ie, intravenous immunoglobulin, corticosteroids, or both.
Of the 1995 patients for whom data on hospital admission was available, 1447 (73%) were admitted to hospital, including 78% of those with an initial platelet count of less than 20⳯109/L. The mean initial platelet count in admitted and non-admitted children was 11·6 (SD 14·2), and 25·2⳯109/L (26·6) respectively. Untreated and treated patients were admitted in 47% and 85% of cases, respectively. Patients on intravenous immunoglobulin, corticosteroids, and intravenous immunoglobulin plus corticosteroids were admitted in 91%, 77%, and 95%, respectively. Of the 1976 children for whom data was available on treatment, 612 (31%) did not receive drug treatment, and 1364 (69%) did, either with intravenous immunoglobulin (576, 29%), corticosteroids (651, 33%), or intravenous immunoglobulin plus corticosteroids (137, 7%). 107 children (7%) received either intravenous immunoglobulin or corticosteroids or both despite an initial platelet count of more than 20⫻109/L. Figure 3 shows how treatment varied in different parts of the world. The mean initial platelet count of untreated children was 26·6⫻109/L (SD 26·8), and of children receiving therapy 10·4⫻109/L (11·5). The mean initial platelet count of patients treated with intravenous immunoglobulin, corticosteroids, and intravenous immunoglobulin plus corticosteroids was 8·0⫻109/L (7·8), 13·3 (15·6), and 8·1 (11·0), respectively. During the first 6 months, antiRh(D) was administered initially in five children and as additional therapy during the 6-month follow-up period in 27 (2%) children. The percentage of children receiving or not receiving initial drug treatment at the time of the diagnosis of ITP was in acute ITP 68% and 32%, respectively, and in chronic ITP 68%, and 33%, respectively. Intravenous immunoglobulin was given in children with acute ITP in 49% (303/620 children), and in children with chronic ITP in 40% (112/278 children). Corticosteroids were administered in children with acute ITP in 51% (317/620 children) and in children with chronic ITP in 60% (166/278 children).
2124
The Intercontinental Childhood ITP Registry serves as a network for facilitating collaborative research in ITP with the advantage of its prospective nature involving investigators in many countries, making it a unique investigative resource. The small numbers of patients recruited by 74% of institutions may be due to inconsistent registration, but is more likely caused by the relative rarity of the disease, as ITP occurs only in about one in 25 000 children yearly.8 Several features of childhood ITP have been confirmed, such as age distribution with a peak between 1 and 6 years,9 presenting platelet count of less than 20⫻109/L, and the fact that most children remit rapidly with or without drug therapy. Additionally, unexpected results have been observed, such as details of outcome and sex distribution with a higher number of boys, as previously described by our group10 and others.11 Whether a bone-marrow aspirate should be undertaken or not has been a matter of long-lasting debate. It has been shown that isolated thrombocytopenia does not necessarily represent an indication for doing a bone marrow aspirate,12,13 but if atypical features of ITP are present or if corticosteroids are considered as initial therapy, a bone marrow aspirate is usually recommended.1,2,14,15 The frequency of hospital admission reported in this survey reflects the recommendation by the American guidelines, which state that it is appropriate for a child with severe, life-threatening bleeding regardless of the platelet count and for a child with a platelet count of less than 20⳯109/L.2 In an assessment of the UK practice guidelines, 83% of patients were admitted,3 and in a survey of diagnostic and management strategies, age and type of bleeding (dry or wet haemorrhage) directed admission—ie, 85% of respondents reported that they would admit a 5-year-old patient with wet purpura, but only 41% admitted patients with dry purpura.4 Thus, further prospective studies are necessary to collect clinical data on morbidity related to bleeding. This requirement is addressed in a follow-up registry to be opened soon. The outcome of children with ITP at 6 months seemed not to be affected by initial treatment strategy. This observation confirms previously published prospective trials.16–18 In a retrospective single-centre study, 51 children with ITP did not receive platelet-enhancing treatment and only four received short-term corticosteroids because of extensive mucosal bleeding.19 The outcome was similar to that of treated children, with chronic ITP occurring in 13% of the patients, and without the occurrence of life-threatening bleeding. The high incidence of chronic ITP in our patients might represent an artefact of under-reporting of patients with rapidly resolving acute ITP at 6 months. Children who developed
THE LANCET • Vol 358 • December 22/29, 2001
For personal use. Only reproduce with permission from The Lancet Publishing Group.
ARTICLES
chronic ITP had higher initial platelet counts at diagnosis and were older than children with the acute form. By contrast with pubished reports,2 we did not report more girls than boys with chronic ITP. Intracranial haemorrhage was reported only twice among 1496 children, of whom 69% were treated, which is far below the often quoted incidence of 1%2,20,21 suggesting a very low risk of this complication during the first 6 months after diagnosis. However, this result must be interpreted cautiously due to incomplete follow-up data (74%) and the possibility that some patients with intracranial haemorrhage at diagnosis might not have been registered even if such patients presented to participating institutions. Lilleyman suggested that the risk of intracranial haemorrhage increases with time, and that it is closer to 0·1–0·2% in the first few days after diagnosis.8,22 Most children with intracranial haemorrhage were reported in retrospective studies, whereas only one child with intracranial haemorrhage has been described in a prospective study.23 The key value of the Registry is the establishment of an international network, with the capability of accruing large numbers of patients with ITP, and confirming or refuting results derived by smaller studies. As a result of this collaborative network future studies will address the hypothesis that major haemorrhage, defined by intracranial or other overt internal or mucous membrane bleeding resulting in anaemia or requiring local treatment,24 is uncommon in children with ITP and only rarely encountered after diagnosis and initial therapy. Data such as these will serve as a basis for conducting clinical trials answering the question—which subgroup of children with ITP requires drug treatment. Contributors T Kühne designed the registry, took part in data analysis, and contributed to the writing of the report. W Berchtold took part in data analysis. P Imbach, P H B Bolton-Maggs, V Blanchette, and G R Buchanan reviewed the study report and contributed to the writing of the report.
4
5
6
7
8 9 10
11 12
13
14
15
16
17
18
Acknowledgments We thank all our colleagues, who have generously submitted data of their patients, and all the nurses and staff of the hospitals who were engaged in the Registry. We thank Heide Luther, administrator, and Ursula Längin, past-administrator for their assistance with the administration of the data and for preparation of the paper. This study was supported through an unrestricted grant by Norvartis Pharma AG.
References 1
2
3
Eden OB, Lilleyman JS. Guidelines for management of idiopathic thrombocytopenic purpura: the British Paediatric Haematology Group. Arch Dis Child 1992; 67: 1056–58. George JN, Woolf SH, Raskob GE, et al. Idiopathic thrombocytopenic purpura: a practice guideline developed by explicit methods for The American Society of Hematology. Blood 1996; 88: 3–40. Bolton-Maggs PHB, Moon I. Assessment of UK practice for management of acute childhood idiopathic thrombocytopenic purpura against published guidelines. Lancet 1997; 350: 620–23.
THE LANCET • Vol 358 • December 22/29, 2001
19
20
21 22 23
24
Vesely S, Buchanan GR, Cohen A, Raskob G, George J. Self-reported diagnostic and management strategies in childhood idiopathic thrombocytopenic purpura: results of a survey of practicing pediatric hematology/oncology specialists. J Pediatr Hematol Oncol 2000; 22: 55–61. Buchanan GR, de Alarcon PA, Feig SA, et al. Acute idiopathic thrombocytopenic purpura—management in childhood. Blood 1997; 89: 1464–65. Jayabose S, Mahmoud M, Levendoglu-Tugal O, et al. Corticosteroid prophylaxis for neurologic complications of intravenous immunoglobulin G therapy in childhood immune thrombocytopenic purpura. J Pediatr Hematol Oncol 1999; 21: 514–17. Gaines AR. Acute onset hemoglobinemia and/or hemoglobinuria and sequelae following Rho(D) immune globulin intravenous administration in immune thrombocytopenic purpura patients. Blood 2000; 95: 2523–29. Lilleyman JS. Management of childhood idiopathic thrombocytopenic purpura. Br J Haematol 1999; 105: 871–75. Lusher JM, Zuelzer WW. Idiopathic thrombocytopenic purpura in childhood. J Pediatr 1966; 68: 971–79. Kühne T, Berchtold W, Be TV, Binh TV, Imbach P. Ethnicity and environment may affect the phenotype of immune thrombocytopenic purpura in children. Pediatr Res 2000; 48: 374–79. Robb LG, Tiedeman K. Idiopathic thrombocytopenic purpura: predictors of chronic disease. Arch Dis Child 1990; 65: 502–06. Dubansky AS, Boyett JM, Falletta J, et al. Isolated thrombocytopenia in children with acute lymphoblastic leukemia: a rare event in a Pediatric Oncology Group Study. Pediatrics 1989; 84: 1068–71. Calpin C, Dick P, Poon A, Feldman W. Is bone marrow aspiration needed in acute childhood idiopathic thrombocytopenic purpura to rule out leukemia? Arch Pediatr Adolesc Med 1998; 152: 345–47. Halperin DS, Doyle JJ. Is bone marrow examination justified in idiopathic thrombocytopenic purpura? Am J Dis Child 1988; 142: 508–11. Reid MM. Bone marrow examination before steroids in thrombocytopenic purpura or arthritis. Acta Paediatr 1992; 81: 1052–53. Sartorius JA. Steroid treatment of idiopathic thrombocytopenic purpura in children: preliminary results of a randomized cooperative study. Am J Pediatr Hematol Oncol 1984; 6: 165–69. Buchanan GR, Holtkamp CA. Prednisone therapy for children with newly diagnosed idiopathic thrombocytopenic purpura. A randomized clinical trial. Am J Pediatr Hematol Oncol 1984; 6: 355–61. Blanchette VS, Luke B, Andrew M, et al. A prospective, randomized trial of high-dose intravenous immune globulin G therapy, oral prednisone therapy, and no therapy in childhood acute immune thrombocytopenic purpura. J Pediatr 1993; 123: 989–95. Dickerhoff R, Von Ruecker A. The clinical course of immune thrombocytopenic purpura in children who did not receive intravenous immunoglobulins or sustained prednisone treatment. J Pediatr 2000; 137: 629–32. Aronis S, Platokouki H, Mitsiki A, Haidus S, Constantopoulos A. Seventeen years of experience with chronic idiopathic thrombocytopenic purpura in childhood. Pediatr Hematol Oncol 1994; 11: 487–98. Blanchette VS, Turner C. Treatment of acute idiopathic thrombocytopenic purpura. J Pediatr 1985; 108: 326–27. Lilleyman JS. Intracranial hemorrhage in idiopathic thrombocytopenic purpura. Arch Dis Child 1994; 71: 251–53. Imbach P, Wagner HP, Berchtold W, et al. Intravenous immunoglobulin versus oral corticosteroids in acute immune thrombocytopenic purpura in childhood. Lancet 1985; 2: 464–68. Medeiros D, Buchanan GR. Major hemorrhage in children with idiopathic thrombocytopenic purpura: immediate response to therapy and long-term outcome. J Pediatr 1998; 133: 334–39.
2125
For personal use. Only reproduce with permission from The Lancet Publishing Group.