• E-mail
• Print
• Comment
• Font Size
• Digg
• del.icio.us
• Discuss article

Idiopathic Thrombocytopenic Purpura: A 10-Year Natural History Study at the Childrens Hospital of Alabama

Posted on: Friday, 29 October 2004, 03:00 CDT

Summary

Childhood idiopathic thrombocytopenic purpura (ITP) is a common disorder. However, single-institution, long-term, natural history data are limited. The objective of this paper is to review presenting features, response to therapy, and natural history of ITP treated at a single pediatric academic medical center. A retrospective chart review was made for all children (ages birth-18 years) diagnosed with ITP (ICD 287.3) and treated at the Childrens Hospital of Alabama/University of Alabama at Birmingham between 1993 and 2003. Four hundred nine patients were identified (49% male, 51% female; mean age: 5.85 years; range: 1 month-17 years). There was no seasonal variation of presentation. The mean platelet count was 19k (0-120k). Bone marrow aspiration (BMA) was performed in 72% but altered the diagnosis or therapy in no patient. Treatment consisted of corticosteroids in 256 (92% response), intravenous immunoglobulin (IVIG) in 125 (87% response), Win-Rho D in 58 (91% response), and no therapy in 71 (100% response). Response was defined as increase in platelet count to > 50k. There was no difference in response to any therapy. No patients died. One patient presented with a CNS hemorrhage at presentation, responded to therapy, and survived. Twenty-three of 409 patients (6%) experienced clinical bleeding requiring hospitalization or blood transfusion. Chronic ITP (persistence > 6 months) was noted in 99 patients (24%). Chronic patients presented at an older age (7.8 vs 5.2 years for acute only, p<0.001), and with higher platelet counts (27k vs 17k, p<0.001). The risk of chronic ITP was partially predicted by presenting platelet count > 50k and age > 10 years, or both; 50% of patients presenting with these features developed chronic ITP vs 24% overall rate. Splenectomy was curative in 30/31 (97%) patients. There was no postsplenectomy sepsis. Of 99 patients with chronic ITP, 25 responded to splenectomy, 37 resolved at a mean of 20.3 months after diagnosis (7-96 months), 36 had persistent mild thrombocytopenia (50k-125k), and 1 failed to respond to any treatment including splenectomy. Overall, 91% of cases resolved with therapy or observation. ITP is a common pediatric disease presenting at any age with low morbidity and mortality. Most cases can be managed by pediatricians without hematology referral. Several equally successful therapeutic options exist. Chronic cases present at an older age with higher platelet counts. Up to 50% of cases of chronic ITP will resolve with ongoing follow-up. The overall prognosis in childhood ITP is excellent.

Clin Pediatr. 2004;43:691-702

Introduction

Idiopathic thrombocytopenic purpura (ITP) is a common childhood hematologic disorder.1-3 All pediatricians in their career will diagnose, treat, and follow up children with this disease. Pediatricians in urban areas often refer their patients to pediatric hematologists for advice and treatment and serve primarily as coordinators of referral and care. In contrast, pediatricians in rural communities or at significant distances from pediatric subspecialists face the often frightening task of diagnosing, treating, and following up children with ITP themselves, utilizing phone consultation and experience without direct involvement of a hematologist. In this study, I review a decade's experience with childhood ITP in a pediatric referral center in a rural state. The insight gained from this natural history study should aid practicing pediatricians in their day-to-day treatment of children with this disorder.

Materials and Methods

Cases were identified by reviewing hospital and/or clinic records for all patients with the discharge diagnosis of idiopathic thrombocytopenic purpura (ITP) (ICD 287.3) treated in the University of Alabama at Birmingham (UAR) Division of Pediatric Hematology- Oncology at the Childrens Hospital of Alabama from July 1993 to June 30, 2003. Before review, approval for the study was obtained from the UAB Institutional Review Board for Human Use, which included a Health Insurance Portability and Accountability Act (HIPAA) waiver of consent. Childrens Hospital maintains an electronic database containing all inpatient discharges dated from January 1, 1994 to June 30, 2003 and all outpatient discharges/visits dated from January 1, 1999 to June 30, 2003. The UAB Pediatric Hematology- Oncology Division maintains medical records of patient visits dated from 1980 to 2003. These sources were searched to ensure a comprehensive dataset and to complete the review for the period before establishment of the hospital electronic database. Following validation of diagnoses and correction of duplicate entries, a total of 492 patients with the diagnosis of ITP were identified for further medical record review. Further review excluded 83 patients owing either to errors in coding or alternative diagnoses. Alternative diagnoses excluded from further review are shown in Table 1. Thus, in total, 409 patients with the diagnosis of ITP form the basis of this report.

Information extracted from the medical record for each patient included the following: name, date of birth, medical record number, date of presentation, presenting platelet count, age at presentation, preexisting medical conditions or diagnoses, diagnostic procedures (especially bone marrow evaluation), treatment given, response to therapy, complications of therapy, complications of ITP, whether or not the illness became chronic, use of splenectomy for treatment, and final outcome.

Definition of response to therapy was defined as follows. A complete response (CR) was defined as return to normal platelet count (greater than 150 10^sup 9^/L) during or after therapy. A partial response (PR) was defined as an increase in the platelet count to greater than 50 10^sup 9^/L. No response (NR) was defined as no response to therapy with continued platelet count below 50 10^sup 9^/L.^sup 4^ Chronic ITP was defined as persistent thrombocytopenia with or without continued therapy for 6 months following initial diagnosis.

Statistical measures of differences between groups are described individually for each comparison in the results section. Calculations of means, medians, modes, standard deviations, and routine statistical measures were performed by the mathematical component of the Microsoft Excel Spreadsheet computer program. Statistical significance was defined as a probability value less than or equal to 0.05.

Results

Patients Characteristics

Four hundred and nine children met the inclusion criteria and are the basis of further review. Forty-nine percent (200 patients) were male and 51% (209 patients) were female. Fifteen percent of patients were identified from inpatient records and 85% from outpatient records. Figure 1 illustrates the age at presentation of the patient group. The youngest patient was 1 month of age at presentation and the oldest was 17 years of age (although the Childrens Hospital and the UAB Division of Pediatric Hematology-Oncology accept patient referrals up to the age of 21 years, no patients older than 17 years were identified). The mean age of presentation was 5.85 years, standard deviation (s.d.) 4.56 years. The median age was 5 years and the mode age of presentation 1 year (12-24 months).

Since childhood ITP frequently follows viral infections, we evaluated whether or not a seasonal difference in presentation existed. Figure 2 shows the distribution of cases by month of diagnosis. The mean number of cases diagnosed in individual months in the 10 years of study was 83.5 cases (thus 3.85 per month per year). Although there was a tendency toward more cases diagnosed from January to June, no significant difference by month of diagnosis was noted over the 10year study period (95% confidence intervals by month 19.5-47.5 [s.d.] 7.17 cases).

Table 1

DIAGNOSES EXCLUDED FROM FURTHER REVIEW (83=17%)

The presenting platelet count at diagnosis ranged from a low of zero to a high of 120 10^sup 9^/L. The mean presenting platelet count was 19.46 10^sup 9^/L with an s.d. of 23.46 10^sup 9^/L. The median presenting platelet count was 10 10^sup 9^/L while the mode presenting platelet count was 2 10^sup 9^/L.

Bone marrow aspiration was performed as part of the initial diagnostic evaluation in 296 cases (72%). In no case did the bone marrow aspiration result in a change from the clinical diagnosis of ITP. No patient experienced a complication from bone marrow aspiration. In the early years of the study, bone marrow aspirations were performed with local anesthesia alone without other sedation. In the later years of study various sedation methods, including conscious sedation with oral midazolam, general anesthesia, or intravenous sedation with propofol, were utilized.

Concurrent diagnoses were noted at the time of presentation in 37 of the patients (9%) (Table 2). The most common were seizure disorder and mental retardation. In 4 patients a recent varicella infection was noted.

Treatment and Response

Specific treatment for thrombocytopenia was given to 338 of the 409 patients (83%). In 71 patients, observation alone was the preferred clinical option. Of the 71 untreated patients, by definition all responded with a return to normal platelet count. F\igure 3 shows the range of therapies and response for the 338 treated patients. In some cases, multiple treatments were administered to the same patient. Thus the total number of treated patients in Figure 3 exceeds the 338 total patients. Responses were defined as partial or complete as outlined in the methods section. The response data shown in Figure 3 represent complete and partial responses combined. Therapies included corticosteroids (typically oral prednisone, but in some case oral or parenteral dexamethasone) or immune globulin. For response calculations all steroid therapies are grouped together and listed under PDN (prednisone) in the figure; 256 patients were treated with steroids, and 235 had a response (92%). Intravenous immunoglobulin (IVIG) was utilized as treatment in 125 patients with 109 (87%) responding to that therapy. Dosing varied but in general was 1-2 g/kg body weight. The anti-D immunoglobin product, Win Rho D, was used to treat 58 patients with 53 responding (91%). The combined results for IVIG/Anti-D were 162 responses in 183 patient treatments for an overall response rate of 89%. Statistical evaluation of the different therapies, as measured by the standard errors of the differences between percentages,5 found no significant difference between any treatment response (p of steroids vs combined IVIG/AntiD = 0.31, p of steroids vs IVIG alone = 0.317

Figure 1. Age of patients at time of initial presentation/ diagnosis. Arrow marks the mean age of 5.85 yrs.

Figure 2. Number of ITP patients by month of presentation/ diagnosis. Data from all 10 years collated and presented in summary form.

Complications of therapy were infrequent but potentially significant. Seven patients developed acute varicella infections while taking corticosteroids. The steroid-induced immunosuppression prompted antiviral therapy with acyclovir in these 7 patients, and all survived without sequelae. In addition, 1 patient developed "massive obesity" on steroid therapy. Thus, the combined overall complication rate with corticosteroid therapy was 8/256 or 3%. The only documented complication with rVIG/Anti-D was aseptic meningitis, which occurred in 4 of 125 patients or 3%. In at least 2 cases the symptoms of aseptic meningitis (headache and stiff neck) prompted radiologic and clinical evaluation for possible central nervous syndrome hemorrhage. All patients recovered completely without sequelae.

Table 2

CONCURRENT PATIENT DIAGNOSES

Morbidity and Mortality

None of the 409 patients died of complications of the ITP or its treatment. Significant morbidity of the thrombocytopenia was noted in 23 patients (6%). Table 3 lists the noted morbidity. The most common complication was epistaxis. Eight patients required hospitalization or intervention for severe epistaxis. Oral mucosal bleeding, menorrhagia, hematemesis, and hematuria were also noted. Five patients experienced bleeding so severe that packed red blood cell transfusion was necessary.

Figure 3. Treatment response.

Table 3

COMPLICATIONS OF THROMBOCYTOPENIA

The most feared complication of ITP is intracranial hemorrhage. This complication was noted in 1 patient (0.24%). She was an 11- year-old who presented to her local emergency room with local seizures. Radiographic evaluation showed a left tempo: oparietal hemorrhage. Laboratory investigation indicated severe thrombocytopenia (platelet count 4 10^sup 9^/L). History revealed concurrent aspirin use in the days before presentation. She was treated with platelet transfusion, IVIG, intravenous corticosteroids, and emergency splenectomy. The thrombocytopenia responded to therapy and the child survived.

Acute vs Chronic ITP

Ninety-nine of the 409 patients (24%) had persistent or recurrent disease for more than 6 months and are thus defined as having chronic ITP. The other 310 patients (76%) completely resolved their disease within 6 months arid meet the criteria for acute ITP. The acute ITP subset was compared with the chronic ITP subset to determine what features, if any, at presentation could predict the development of chronic disease.

Age at Presentation. Figure 4 shows the age of presentation of patients according to their final status as acute or chronic ITP. Patients who later were found to have chronic ITP presented at a significantly older age than patients with acute ITP (7.8 4.8 years for chronic vs 5.2 4.3 years for acute, p < 0.001, by standard error of the difference between 2 means). The median age of presentation for acute patients was 4 years, whereas the median age of presentation of chronic patients was 8 years. The mode age for both groups was 1 year (12-24 months) of age. Figure 4 demonstrates that although the mean and median ages at presentation were significantly higher for the chronic group, chronic disease can occur in the youngest of patients, and patients presenting as teenagers are still likely to have self-limited, acute disease.

Figure 4. Acute vs chronic ITP by age of presentation/diagnosis. Note that chronic patients presented at a significantly older age than acute patients.

Presenting Platelet Count: The mean presenting platelet count for the chronic patients (27 10^sup 9^/L) was significantly higher than that for the acute group (17 10^sup 9^/L) (p < 0.001 determined by standard error of the difference between 2 means). However, the standard deviations for both groups are large: 30 10^sup 9^/L for chronic and 21 10^sup 9^/L for acute, resulting in significant overlap. The median platelet count (13 10^sup 9^/L for chronic vs 9 10^sup 9^/L for acute) and mode platelet count (4 10^sup 9^/L for chronic vs. 2 10^sup 9^/L for acute) were not different.

Gender: No difference was noted in the incidence of chronic ITP in male vs female patients.

Predicting Chronic ITP: Overall, the patients who developed chronic ITP tended to present at an older age and with a higher platelet count than those subsequently noted to have acute ITP. However, utilizing the presenting age or platelet count alone could not predict at the time of diagnosis patients who subsequently went on to develop chronic disease. Table 4 demonstrates the predictive results of combining a presenting platelet count greater than 50 10^sup 9^/L with presenting age of 10 years or greater. In the overall patient group, as stated above, 76% of patients manifested acute ITP and 24% had chronic ITP. In contrast, patients presenting at age 10 years or older had a 41% rate of chronic ITP. Patients presenting with a platelet count of 50 10^sup 9^/L or higher had a 45% rate of chronic ITP, and in the patient subset with both age 10 years or older and platelet count of 50 10^sup 9^/L or higher, the rate of chronic ITP was 50%. The rate of chronic ITP was significantly higher in each of these patient subsets as noted in the table and measured by the standard error of the difference between percentages. However, even with both of these characteristics predicting at the time of presentation, the patients destined to manifest chronic symptoms is at best a 50-50 proposition.

Table 4

PREDICTING CHRONIC ITP

Table 5

PREDICTING ACUTE ITP

If the same data are examined from the standpoint of predicting a low risk of chronic disease, a few points are noted. For example, in patients with age at presentation less than 10 years and presenting platelet count less than 50 10^sup 9^/L, only 43 of 272 patients manifested chronic disease for a 16% risk. Further, in patients presenting as infants or young children, the rate of chronic disease was low. For example, in patients presenting under age 12 months, only 2 of 28 experienced chronic disease (7%), and in children less than 24 months at initial presentation, the rate was only 11% (9/ 84) (see Table 5).

Splenectomy in ITP

Overall, in our patient group, 31 patients were treated with splenectomy (7.5%). The mean time from diagnosis to splenectomy was 17 months (range: immediate to 72 months). As expected, most of the patients who underwent splenectomy had chronic ITP (26 of 31). However, 5 patients with acute ITP (illness of less than 6 months' duration) underwent splenectomy. In each case, the indication for early splenectomy was bleeding uncontrolled by other therapy. The sites of bleeding and age at splenectomy in the acute cases were as follows: 3 patients with severe epistaxis at ages 3, 3, and 4 months from diagnosis, respectively; 1 patient with menorrhagia at 4 months from diagnosis; and the child with central nervous system hemorrhage, discussed above, who underwent splenectomy at initial presentation. All 5 patients with acute ITP had a complete response to splenectomy.

Twenty-six patients underwent splenectomy for chronic ITP. All but 1 of these patients had a complete response to the surgery. Overall, 30 of 31 patients in our dataset responded to splenectomy (97% response rate). No patient experienced a postsplenectomy septic complication.

Patient Outcome

Of the 409 patients identified in this retrospective review, 310 responded to therapy or resolved spontaneously within 6 months (acute ITP). The other 99 patients had persistent or recurrent symptoms past 6 months and thus had chronic ITP. Of the patients with chronic ITP, 25 underwent splenectomy with resolution. An additional 37 patients resolved their disease spontaneously, with or without ongoing medical management, at a mean of 20.3 months from initial diagnosis (s.d.: 18.4 months, range: 7 to 96 months). Thirty- six patients had a partial response to therapy and at last follow- up had a platelet count between 50 and 125 10^sup 9^/L without clinical bleeding. Only 1 of the 99 patients with chronic ITP failed to respond to all therapy, including splenectomy. That patient had persistent severe thrombocytopenia without clinical bleeding at last follow-up. Thus, the overall complete response rate of the 409 patients in the study was 91% (310 acute + 62 chronic). Combined par\tial and complete responses were noted in 99% (408/409). The outcome is summarized in Figure 5.

Figure 5. Outcome of total patient group. CR=complete response.

Discussion

Idiopathic thrombocytopenic purpura (ITP) is a common pediatrie hematologic disorder. The estimated incidence is 2-8 per 100,000 population per year.6-8 It is likely that all pediatricians will encounter children with this disorder at some point in their practice. The illness is frightening to parents and providers alike. A pediatrician's response to the child with apparent ITP should be based on knowledge of the illness and its natural history. Long term longitudinal studies such as the one we report offer guidance on evaluation and therapy options.9-11 Key questions for the practicing pediatrician include the following: (1) What diagnostic evaluation is necessary to confirm the diagnosis of ITP? (2) Is referral to a pediatric hematologist necessary? (3) Should the patient receive specific treatment? If so, what and for how long? (4) What can I tell the family to expect? Are there any specific instructions that are necessary?

The initial diagnostic evaluation for children with the classic presenting signs of ITP should consist of a thorough history, physical examination, and complete blood count with review of a peripheral blood smear. A child presenting in the correct age group (2-10 years) with sudden onset of petechiae and purpura, isolated thrombocytopenia, otherwise normal findings from physical examination, and no evidence of atypical white blood cells on the blood smear rarely represents another disease process. Worrisome findings that should prompt further thought include the following: history suggesting a chronic bleeding disorder, history of bone or joint pain, and presence of pathologic lymphadenopathy, hepatomegaly, splenomegaly, concurrent anemia or neutropenia, and presence of atypical white blood cells on the blood smear or other physical examination stigmata suggesting an inherited hematologic defect such as Fanconi's anemia. Our study suggests that age over 10 years and/or presenting platelet count above 50 10^sup 9^/L are relative indicators of chronic disease and may warrant referral to a pediatric hematologist. Our results do not find presentation as an infant an independent risk variable12 requiring further evaluation or unique therapy.

The most commonly used additional laboratory test to confirm the diagnosis of presumptive ITP is a bone marrow aspirate to rule out acute leukemia. Our study confirms the findings of other researchers that a routine bone marrow aspirate is unnecessary in typical cases.13 In none of our 296 cases with bone marrow evaluadon was the diagnosis changed by the procedure. Bone marrow evaluation is thus not routinely needed and can be reserved for patients with atypical features.

A child presenting with classic clinical presentation and no worrisome historical, physical, or laboratory findings docs not require referral to a pediatric hematologist. A pediatrician should feel comfortable making treatment decisions, perhaps with phone consultation, if necessary. In contrast, children with atypical presenting features may require further hematologic evaluation. Table 1 lists a broad range of clinical problems that may superficially mimic ITP but require different diagnostic and management approaches.14,15

The most appropriate initial therapy for children with apparent ITP remains controversial. Basically, there are 2 opinions expressed by pediatric hematologists: (1) Watch and wait with the expectation of a mild clinical course and spontaneous resolution in most patients or (2) initiate treatment to lessen the risk of significant bleeding. At a panel discussion held at the 2002 Meeting of the American Society of Pediatric Hematology/Oncology, treatment approaches were evenly divided among participating members.16 Pediatric hematologists can agree that children presenting with significant hemorrhage deserve intervention, but care of the child with severe thrombocytopenia with little or no bleeding remains uncertain. Our data support both arguments. For example, clinically apparent bleeding was infrequent, occurring in only 6% of patients, and in the vast majority of our patients (99%), long-term follow-up documented clinical resolution or lack of bleeding. Our 1 case of intracranial hemorrhage was a presenting complaint and thus therapeutic intervention had. no impact on the occurrence of this most feared complication. However, most of our patients (83%) received some form of therapy, reflecting the biases of the pediatric hematologists in our group. The treatment was well tolerated with a low incidence of recognized side effects. Our group continues to treat most patients with platelet counts less than 20 10^sup 9^/L.

If the decision is made to treat patients with childhood ITP, the next question is what available therapy to recommend. Our results demonstrate 90% response rates with all of the common therapies, including corticosteroids, intravenous immunoglobulin, or anti-D. Thus, the treating physician can prescribe any of the 3 options with confidence based on individual patient characteristics. IVIG and anti-D require intravenous access and expose children to blood product components. Corticosteroids require parental and child compliance. In our practice, oral prednisone at a starting dose of 2 mg/kg/day given for 10-14 days followed by a taper is the most common prescribed initial therapy. However, the use of IVIG or anti- D is increasing. Whether therapy is prescribed or not, patient and parent education and ongoing follow up are crucial.

The most feared complication of ITP is intracranial hemorrhage.17 The incidence of this complication is 0.1% to 0.5% or 1 to 5 in 1,000 cases.18 Although there is no definitive means to define in advance those patients who will experience intracranial hemorrhage, concurrent or recent head trauma is reported in up to one fourth of cases.16 Thus, warning patients and parents to avoid head trauma and to seek immediate evaluation in the event of head trauma is a key component of our management algorithm. We provide written instruction sheets to the family to further emphasize these points. Common sense also mandates informing families to avoid exposure of their thromhocytopenic children to agents that further alter platelet function, specifically aspirin and the nonsteroidal antiinflammatory drugs. Our single patient with intracranial hemorrhage had a history of aspirin use at the time of presentation. We typically follow platelet counts response to therapy or watchful waiting within a week of initial diagnosis. If treatment is prescribed and no platelet increase is noted by 1 week, the provider must consider whether to modify or discontinue treatment. Once the illness is resolved, family and providers should remain vigilant for signs of recurrence, especially following another viral illness or antigen exposure. Management of recurrent episodes can follow the same guidelines as initial cases. Persistence of symptoms beyond 6 months defines the child as having "chronic ITP" and warrants further investigation and usually referral to a pediatric hematologist. However, it is important to recognize that even children with chronic symptoms have an excellent prognosis for long term disease resolution or control.

Before I summarize our findings and recommendations, it is proper to ask whether or not our population reflects experience in the "real world" or is tainted by its retrospective nature and academic medical center referral bias resulting in a skewed view of ITP. For the purposes of this report, I believe the data are sound and serve as evidence to guide practice decisions. Alabama is predominantly a rural state. The 2000 US Census documented a total population of 4.4 million for the state of Alabama.19 For ages birth to 18 years the state population was 1.1 million. The Children's Hospital of Alabama, located in Birmingham, is the primary referral site for pediatric subspecialty care for more than 75% of the state. Thus, the referral experience of the Division of Pediatric Hematology/ Oncology at University of Alabama at Birmingham housed at the Children's Hospital mimics most of the state of Alabama. On average, 41 cases of ITP were referred to or diagnosed by the Division from 1994 to 2003 (range: 22 to 51). This yearly incidence is comparable to the previously reported incidence of ITP in the United States, 2- 8 per 100,000 population per year.4-6 For example, the average 41 cases in a population of 1 million children aged 0-18 years is equivalent to an annual incidence of 4 per 100,000 per year, right in the middle of the estimated incidence range. Undeniably, milder, classic cases were treated in the local communities by experienced pediatricians with or without telephone consultation with our group. However, it is unlikely that many or any severe or chronic cases escaped referral or consultation. This fact makes the excellent outcome of our group even more remarkable since it is likely that our population and results overestimate the rate of chronic disease in pediatric patients and do not fully represent the milder spectrum of this disease.

In summary, this decade-long natural history review of children with ITP treated at a single subspecialty referral center offers evidence for the following recommendations for practicing pediatricians.

1. Newly diagnosed children with ITP meeting clinical and laboratory criteria are appropriate for management by the general pediatrician without bone marrow aspirate and without referral to a pediatric hematologist. Cases that do not meet the classic picture should be reviewed with a pediatric hematologist before therapy is initiated.

2. Whether to provide pharmacologic therapy to increase the platelet count is an individual decision between the pediatrician and the patient/patie\nt's family. In large populations, watchful waiting, corticosteroids, intravenous immunoglobulin, or anti-D are equally effective treatments. Each patient requires an individual decision on the best therapy.

3. Regardless of treatment chosen, parent and patient education is critical and may further lessen the risk of serious bleeding complications.

4. Patients with significant bleeding at diagnosis should receive pharmacologic treatment.

5. Patients who experience chronic ITP deserve further evaluation, often in cooperation with a pediatric hematologist. However, the prognosis for children with chronic ITP remains excellent.

6. Age and platelet count at initial presentation can help identify a subset of patients with higher or lower risk of subsequent chronic disease.

7. Splenectomy is an effective therapy for ITP in both the acute and chronic phase, but careful patient selection for this procedure is necessary.

8. Most children with ITP are expected to subsequently resolve their illness without long-term sequelae.

REFERENCES

1. Wilson DB. Acquired platelet defects. In: Nathan DG, Orkin SH, Ginsburg D, Look TA, eds. Nathan and Oslo's Hematology of Infancy and Childhood, ed. 6. Philadelphia: WB Saunders; 2003:1500-1609.

2. Chu YW, Korb J, Sakamolo KM. Idiopathic thrombocytopenic purpura. Pediatrics in Review. 2000;21:95-103.

3. Cines DB, Blanchetle VS. Immune thrombocytopenic: purpurn. N Engl J Med. 2002;346:995-1008.

4. Law C, Marcaccio M, Tain P, et al. High-dose intravenous immune globulin and the response to splenectomy in patients with idiopathic thrombocytopenic purpura. N Engl J Med. 1997; 336:1494- 1498.

5. Swinscow TDV. Statistics at Square One. London: British Medical Association; 1983.

6. Butros LJ, Bussel JB. Intracranial hemorrhage in immune thrombocytopenic purpura: a retrospective analysis. J Pediatr Hematol Oncol. 2003; 25:660-664.

7. Medeiros D, Buchanan GR. Current controversies in the management of immune thrombocytopenic purpura during childhood. Pediatr Hematol. 1996;43:757-772.

8. Benesch M, Kerbl R, Lackner H, et al. Low dose versus high- dose immunoglobulin for primary treatment of acute immune thrombocytopenic purpura in children: results of a prospective, randomized single-center trial. J Pediatr Hematol Oncol. 2003; 25:797-800.

9. 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.

10. Kuhne T, Buchanan GR, Zimmermann S, et al. A prospective comparative study of 2,450 infants and children with newly diagnosed idiopathic thrombocytopenic purpura (ITP) from the intercontinental childhood ITP study group. J Pediatr. 2003; 143: 605-608.

11. Rothoj S, Hedlund-Treutiger I, Rajantie J, et al. Duration and morbidity of newly diagnosed idiopathic thrombocytopenic purpura in children: a prospective Nordic study of an unsclecled cohort. J Pediatr. 2003; 143:302-307.

12. Sandoval C, Visintainer P, Ozkaynak F, et al. Clinical features and treatment outcomes of 79 infants with Immune thrombocytopenic purpura. Pediatr Blood Cancer. 2004;42:109-112.

13. 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;84:1068-1071.

14. Bader-Meunier B, Proulle V, Triebet C, et al. Misdiagnosis of chronic thrombocytopcnia in childhood. J Pediatr Hematol Oncol. 2003;25:548-552.

15. Drachman JG. Inherited thrombocytopenia: when a low platelet count does not mean ITP. Blood. 2004; 103:390-398.

16. Bolton-Maggs P, Tarantino MD, Buchanan GR, et al. The child with immune thrombocytopenic purpura: is pharmacotherapy or watchful waiting the best initial management? J Pediatr Hematol Oncol. 2004;26:146-151.

17. Lilleyman JS. Intracranial hemorrhage in idiopathic thrombocytopenic purpura. Archi Dis Child. 1994; 71:251-25:5.

18. Bolton-Maggs P. Severe bleeding in idiopathic thrombocytopenic purpura. J Pediatr Hematol Oncol. 2003;25:s47-51.

19. US Census Bureau. 2000 Data Set. Profile of General Demographic Characteristics. State of Alabama.

Raymond G. Watts, MD, FAAP

Associate Professor of Pediatrics, Division of Hematology- Oncology, University of Alabama at Birmingham and Children's Hospital of Alabama.

Reprint requests and correspondence to: Raymond G. Watts, MD, Associate Professor of Pediatrics, Division of Hematology-Oncology, 1600 7th Avenue South, ACC 512, Birmingham, AL 35233.

2004 Westminster Publications, Inc., 708 Glen Cove Avenue, Glen Head, NY 11545, U.S.A.

Copyright Westminster Publications, Inc. Oct 2004

Source: Clinical Pediatrics

More News in this Category