Idiopathic thrombocytopenic purpura (ITP) is a quantitative disorder of platelets in which immune mechanisms are responsible for the premature destruction of circulating platelets. This article summarizes our current knowledge regarding its clinical manifestations, classification, pathophysiology, diagnostic evaluations and therapeutic strategies.
Clinical manifestations of ITP correlate with the degree of thrombocytopenia.1'2 They are listed in Table 1. Platelet counts of greater than 50,000 per mm3 are rarely associated with bleeding. Patients with platelet count of 20,000 to 50, 000 per mm3 have more risk of developing purpura with minor trauma, while those with platelet count of less than 20,000 per mm3 are at more risk of developing bruises spontaneously. Risk of serious mucosal bleeding, intracranial bleeding or other internal bleeding is high in patients with platelet counts of less than 20,000 per mm3.
The most serious and life-threatening complication of ITP is intracranial hemorrhage (ICH). Woerner et al reviewed 18 cases of childhood ITP and noted that the platelet counts less than 20,000 per mm3 increased the risk of ICH. 3 The incidence of ICH in childhood ITP is less than 2%. They emphasized the importance of avoiding antíplatelet drugs as well as the significance of location of ICH. Posterior fossa hemorrhages were observed to be more dangerous because of the possibility of rapid cerebellar herniation and brainstem compression. The ages of these children varied from 18 months to 16 years. History of trauma was absent in the majority of the patients. Interval between the diagnosis of ITP and ICH ranged from between 3 days to 5 years. Woerner et al recommended that splenectomy always be performed prior to any neurosurgical procedure. 3 All surgical procedures can be performed under the same general anesthesia. Eleven of the 18 patients had a favorable outcome.
The same authors further recommended that every effort should be made to avoid head trauma in children with ITP A theoretical question of whether stressful procedures such as bone marrow aspiration or frequent venipunctures increases the risk of ICH in children with ITP remains unanswered. However, it seems improbable since the onset of symptoms did not show any relationship to these procedures. The age of the patient and the duration of the onset of thrombocytopenia, previously considered to be important with reference to the likelihood of ICH, have no significance with reference to the frequency of ICH. However, a platelet count of less than 20,000 per mm3 is critically significant.
ITP is classified into acute and chronic forms. Chronic ITP is usually diagnosed if thrombocytopenia persists beyond 6 months after initial presentation.4 Differentiating features between acute and chronic childhood ITP are listed in Table 2. Acute ITP is far more common in childhood than chronic ITP. Acute ITP occurs in children below 10 years of age (with a peak incidence between 2 and 6 years) whereas chronic ITP in children above 10 years of age is common. Chronic ITP is observed more commonly in females than in males. Platelet counts in acute ITP are more often very low (ie, less than 20,000 per mm3) while in chronic ITP they are moderately low (ie, 20,000 to 75,000 per mm3). Serum IgA levels are usually low in chronic ITP. Platelet-associated IgG (PaIgG) is markedly elevated in acute ITP whereas it is unduly elevated in chronic ITP. In many patients with acute ITP, a history of antecedent viral illness occurring within several weeks of the onset of thrombocytopenia can be demonstrated whereas chronic ITP is more often associated with an insidious onset of bleeding manifestation. In children with chronic ITP, intercurrent viral infection can reduce platelet counts to dangerously low levels ( ie, less than 20,000 per mm3).
CLINICAL MANIFESTATION OF ITP
DIFFERENTIATING FEATURES BETWEEN ACUTE AND CHRONIC CHILDHOOD ITP4
Another interesting finding in children with ITP was reported by Matoth et al. 5 They noted that children with chronic ITP had a disproportionately high incidence of the syndrome of minimal cerebral dysfunction (short attention spans, learning difficulties, behavior problems, poor memory) and abnormalities of EEGs. The authors suggested that multiple minute intracraneal capillary bleeding could account for the above-mentioned abnormalities.
According to Lusher et al about 4% of children with ITP, thrombocytopema may recur after months or years from the remission.2 Such recurrences are often preceded by viral infections. In splenectomized patients, recrudescence of thrombocy topen ia could be due to exacerbation of dormant chronic ITP or to hypertrophy of accessory splenic tissue. Aspnes et al and later Verheyden et al emphasized the latter possibility in their case reports.6·7 A simple method to exclude such a diagnosis is an examination of blood smear for the presence of Howell-jolly bodies in red blood cells. If Howell-Jolly bodies are absent then further documentation of functioning accessory splenic tissues can be performed by radioisotopic scanning of the spleen.
The observation that the development of acute ITP is often preceded by a viral illness suggests that on interaction between viral antigen, viral antibody, and immune complex formation may be responsible for platelet destruction. Platelet removal is probably facilitated through absorption of viral antigen onto the platelet surface followed by antiviral antibody attachment to virus incorporated onto the platelet membrane. Fc portion of this immune complex then may bind to the Fc receptor on a macrophage which then engulfs the platelet virus antibody complex by phagocytosis causing platelet destruction.
Experimental proof for the above hypothesis is lacking. However, this postulate is generally considered to be correct.
Chronic ITP is in most cases due to an antibody against a platelet-associated antigen, although the specific nature of this "autoantigen" remains obscure. In a study by Beardsley et al antibodies which bind to a 100,000-mol wt protein were found in many children with chronic ITP.8-9 Mason and McMillan's studies in this regard showed patients with chronic ITP produce antibody reacting with antigens associated with human platelet proteins and the antigenic pattern differs between patients. I0
McMillan et al demonstrated that the spleen in chronic ITP serves as an important site for production of antiplatelet antibody.11 McMillan also demonstrated that bone marrow cells from patients with chronic ITP produce antiplatelet antibodies and suggested that in patients who do not respond to splenectomy, the bone marrow is the major site of antibody production.12 The antiplatelet antibody is a "panantibody," ie, it binds to autologous platelets, hemologous platelets and also to megakaryocytes.
The reasons for the development of antiplatelet antibodies in chronic ITP remains unclear.
A defect in the immune regulation seems to play a role in production of both acute and chronic ITP. Shannon et al investigated lymphocyte subsets in children with acute and chronic ITP. 13 They found that the mean percentages of B lymphocytes and total number of T lymphocytes, suppressor T cells and helper T cells were similar in patients with acute and chronic ITP and in controls. However, the values for many children with ITP fell outside of the normal range. These results suggested a heterogenous nature of immunemediated thrombocytopenia in childhood.
According to McMillan, destruction of platelets in ITP results from antibody binding to antigen, with or without complement activation, followed by phagocytosis.1 Phagocytosis may be triggered by either the Fc fragment of the bound IgG molecule (with IgG1 and IgG-3 being most active) or perhaps in some patients by fixation of C3B to cell surface resulting from complement activation. The Fc receptor binding mechanism plays a most important role in destruction of platelets. Inhibition of this mechanism occurs when high-dose intravenous infusion of IgG (IVIg) is given. The spleen is undoubtedly the most important site for destruction of platelets in ITP, but under certain circumstances the bone marrow and liver may be involved.
PATHOGENESIS OF ITP IN CHILDREN
Basic Mechanism: Formation of IgG with affinity for platelet membrane (IMgG). Triggering cause for formation of PaIgG unknown.
A) Orthogenesis of acute ITP in children:
a) Immune complex theory:
Viral infection - absorption of viral antigen on to the platelet surface - binding of antibody against virus to virus-platelet complex f destruction of platelets by immune system: primarily by splenic macrophages via the Fc receptors on their surfaces.
b) "Autoantigen" theory: Viral infection - alteration in anti-genecity of platelet membrane ("autoantigen") - production of platelet specific antibodies - destruction of antibody coated platelets by immune system: primarily by splenic macrophages, via the Fc receptors on their surfaces.
B) Pathogenesis of chronic ITP in children:
Autoimmune disease process leading to production of an antibody (FaIgG) specifically directed against a plateletassociated antigen - destruction of platelets by reticuloendothelial system, mostly in spleen.
Platelet Kinetics (in ITP)
Harker and Finch showed that in normal subjects the following results of thrombokinetics were obtained14:
A) a mean platelet count of 250,000/ul ± 35,000
B) immediate recovery in the systemic circulation of injected 51 cr labeled platelets of 64.6% ± 4.1, indicating sequestration of about 25% to 30% of platelets in the normal spleen
C) a mean platelet survival of 9.9 days ± 0.6
D) the platelet turnover rate of 35,000 platelets/ul ±4,300
E) daily platelet production per nuclear unit of megakaryocyte averaged 49 ± 5
Platelet turnover reflects platelet removal from the circulation. In the stable state, the platelet turnover is also a measure of platelet release from the marrow into the circulating blood and it is referred to as effective platelet production. Since megakaryocytes may reflect the platelet-producing capacity of the marrow, this measurement is referred to by them as total platelet production.
Measurement of thrombokinetics performed by Harker in patients with ITP showed that megakaryopoiesis (total thrombopoiesis) and platelet turnover (effective thrombopoiesis) were increased in parallel to as much as eight times normal.15
Branehog et al also concluded their study with similar results performed on patients with ITR 16 Additionally, they observed that platelet size was significantly increased in ITP and the mean platelet diameter was 1.6 times the normal (normal diameter 1.70 ± 0.01 U and for patients with ITP, 2.65 ± 0.07 U). There was a significant relationship between platelet size and platelet production rate and an inverse relationship between platelet size and platelet mean life span. There was also a significant correlation between platelet size and the proportion of young megakaryocytes.
In a recent study by Siegel et al, it was concluded that patients with autoimmune thrombocytopenia with markedly shortened platelet survival time and a high platelet turnover rate had favorable response to splenectomy, whereas the patients with nearly normal survival times and low platelet turnover rate had poor response to splenectomy.17 Neither the pattern of platelet destruction nor the amount of FaIgG were valuable in predicting response to splenectomy. This study underscores the importance of thrombokinetics in predicting response to splenectomy in patients with autoimmune thrombocytopenia.
LABORATORY FEATURES OF ITP AND LABORATORY EVALUATION OF PATIENTS WITH ITP
A. Laboratory Features of ITP:
* Thrombocytopenia with increased number of large platelets (> 2.5U) on the blood smear or by electronic sizing device
* Increased number and size of megakaryocytes
* Reduced intravascular platelet survival and high turnover rate
* Elevated levels of platelet associated IgG (PaIgG)
B. Laboratory Evaluation of Patients with ITP;
After the diagnosis of ITP is made the following blood tests are performed to exclude an underlying cause of ITP:
* Coombs test - Direct and Indirect (Evans syndrome)
* Antinuclear antibody test
* Rheumatoid factor test
* Monospot test
Laboratory Features of ITP
The diagnosis of ITP is made on the basis of thrombocytopenia with normal marrow containing increased or normal quantities of megakaryocytes in patients who do not show evidence of splenomegaly or disseminated intravascular coagulopathy. Garg et al observed an excellent correlation between percentage of large platelets (> 2.5 U diameter) and the number of megakaryocytes per tow power field in patients with ITP.18 In normal controls percentage of large size platelets amount to approximately 10 with the presence of 2 megakaryocytes per low power field in bone marrow. In patients with ITlP there is a fourfold increase in both these parameters. As reported in the discussion of thrombokinetics in ITP, several authorities have shown that ITP is characterized by increased number and size of megakaryocytes, reduced intravascular platelet survival and high turnover rates.
During the past 10 years, various techniques have been developed to measure the presence of IgG on washed platelets from patients with ITP. Kelton has reviewed these techniques recently in systematic fashion.19 The IgG measured on the platelet surface is known as FaIgG (platelet associated IgG), since it is not known whether it represents specific (Fab binding), immune complex (Fc binding), or non-specifically absorbed IgG.
The following difficulties are encountered in the performance and interpretation of the test results of FaIgG: 1) difficulty is encountered in isolating sufficient number of platelets in thrombocytopenic patients with ITP; 2) nonspecific absorption of other plasma proteins by the platelets, and 3) a relatively narrow margin of difference between the amount of PaIgG on normal platelets and on ITP platelets. The amount of IgG in platelets from patients with ITP can be merely two to three times the normal value. Non-specific absorption of plasma proteins on the platelets has caused a major problem in developing direct-binding assays of PaIgG. These assays measure the binding of labeled antiserum to the platelets. Only a small proportion of the antiserum contains the antibody of interest. However, all proteins of antiserum are labeled. Therefore, the platelet binding of labeled but "non-immune" proteins must be differentiated from the binding of the specific antibody.
There are three general types of direct assays used to measure PaIgG:
1) Direct binding of a labeled probe: These assays measure the platelet-binding of labeled antiserum, usually amiIgG. A variety of techniques used to minimize the nonspecific absorption of labeled antiserum proteins to the platelets include;
A. Dilutation of the labeled antiserum in non-radiolabeled non-immunized serum from the same species results in the blocking of the non-specific uptake of the labeled nonimmune proteins.
B. Chemical modification of the target platelet membrane by the addition of a fixative such as paraformaldehyde reduce non-specific protein binding.
Direct-binding assays of PaIgG consist of the use of one of the following methods:
* use of radiolabeled anti-serum IgG
* use of immunofluorescent conjugated antiserum
* use of enzyme-linked antiserum for immunosorbent assays (ELISA)
* use of staphylococcus aureus protein A)1 an extract from the cell wall of s. aureus binds to the Fc region of IgG subclasses IgGl, IgGZ and IgG4
2) Two-Stage Assays:
Dixon et al reported the first study describing the clinical applicability of a technique that determines antiplatelet antibody directly on the platelet surface.20 The technique is a quantitative complement lysis- inhibition assay.
The general principle of this and all other two-stage assays is the same, ie, antiserum in excess is incubated with varying dilutions of washed test platelets. The unbound antiserum is then separated from the platelets and assayed by measuring its biologic activity by complement lysis assay (ie, sheep erythrocytes labeled with human IgG + anti-human IgG serum (or unbound serum) + complement lysis of sheep erythrocytes) or its binding to immobilized IgG (immunoradiometric assay and enzyme linked assay).
3) Measurement of IgG after Solubilization of the Platelets:
In this assay system, platelet lysis allows measurement of both surface IgG as well as internal IgG (ie, total PaIgG). Antiserum probes are used to quantify the PaIgG. Morse et al described a nephelometric (nephelometry = measurement of turbidity or cloudiness in a suspension of solution) technique for measuring PaIgG. 21 The technique uses solubilized platelets to which anti-human IgG is added. This allows formation of an immune complex. The rare of immune complex formation is measured and then related to a standard curve. The assay has an acceptable coefficient of variation and just like other direct assays is positive in 90% of patients with ITP.
After the first report by Harrington et al that a thrombocytopenic factor could be demonstrated in plasma of patients with ITP, investigators developed several in vitro techniques to measure antiplatelet antibody in plasma of patients with ITP.22 However, these techniques proved to be insensitive and non-specific and hence they did not gain popularity.
Indirect techniques measure the binding of antibody to platelets by changes in a platelet -dependent end point such as platelet release (leading to platelet aggregation reaction, or release of 51 cr from chromium linked platelets, or serotonin release from platelets) or platelet interaction with other cells (ie, interaction of antibody platelets causing lymphocyte transformation, leukocyte phagocytosis or complement formation) or inhibition of platelet function (eg, inhibition of clot retraction).
In patients with recurrent or chronic thrombocytopenia, a diagnosis of pseudo-von Willebrand's (platelet type von Willebrand's) disease is worthy of consideration.
TREATMENT OF ACUTE ITP OF CHILDHOOD
The use of corticosteroide in acute ÍTP of childhood continues to be controversial. Lusher et al have favored management of this disease in children under 13 years of age without the use of corticosteroids.2 In their experience, the analysis of platelet data indicated no advantage in terms of rate of recovery when corticosteroids were used. They also argue that if the major purpose of corticosteroid treatment for this disease is to reduce the risk of intracranial hemorrhage, then they have not seen a single case of ICH among 465 consecutive cases of acute ITP in children, the majority of whom did not receive corticosteroids. Dunn et al on the basis of their experience and review of other evidences make a case favoring the use of corticosteroids in children with acute ITP. 2^ They recognize that spontaneous recovery occurs in 80% to 95% of children with ITP within 4 months, however, the risk of hemorrhagic complications does exist during the period of thrombocytopenia. In a prospective controlled study of prednisone treatment of acute ITP in childhood, they concluded that median time to normalization of the platelet count was 21 days in the steroid-treated group as compared with 60 days in the controlled group with p value of 0.03.
However, the latest information in this regard comes from a cooperative randomized double-blind study conducted by a group of pediatrie centers from Switzerland and Southern Germany and reported by Sartorius. 24 In this study, a total of 93 patients between the ages of 6 months and 16 years with ITP were randomized to receive prednisone placebo tablets (Regimen A) for 21 days in a double-blind system. Platelet counts and tourniquet tests were performed at prescribed intervals. In order to obtain optimum information, the following three variables were analyzed: 1) the time each patient required to attain a platelet count of 30,000/mm3, ie, the time to overcome the increased risk of bleeding; 2) the time a patient took to reach a platelet count of 100,000/mm3; a value which may reflect the onset of remission; and, 3) the time when the Rumpel-Leede test became negative. The end results of the study showed that 90% of steroid-treated patients had reached a platelet count of 30,000/mm3 within the first 10 days of treatment, as compared with only 45% of patients who achieved a platelet count of 30,000/mm3 in a group of untreated patients. Similarly, a platelet count of 100,000/mm3 was attained in the 80% of steroid treated patients at day 15 as compared with 30% of untreated patients. All of the patients treated with steroids attained Rumpel-Leede test negativity by day 18 whereas approximately only 45% of untreated patients attained negativity by day 18. In all evaluated parameters the p value was highly significant (p < 0.01). Thus, it is important not to withhold corticosteroid therapy from children with ITP who present with severe thrombocytopenia, since platelet counts under 30,000/mm3 are known to be associated with an increased risk of bleeding and shortening of this potentially hazardous period is clearly accomplished when corticosteroids are used. Also to be noted in favor of its use, is the observation that serious complications or side effects were not encountered in the above-mentioned trial. Thus benefits of corticosteroids clearly outweigh their risks.
Buchanan et al also conducted a randomized clinical trial using prednisone vs. placebo treatments in children with acute ITP.25 They'failed to find statistically significant differences between the two treatment groups except on day 7 of therapy when children receiving prednisone had higher platelet counts and lower bleeding scores and bleeding times than those taking placebo. Bleeding times correlated inversely with the platelet count in both groups. Prednisone did not appear to influence bleeding time independent of its effect on platelet count. They concluded that prednisone did not clearly improve hemostasis in childhood acute ITP except transiently at the end of 1 week of treatment.
Many investigators have attempted to explain the mechanism of action of corticosteroids in ITP. The prolongation of platelet life span after treatment with corticosteroids in ITP is best explained by postulating that corticosteroids inhibit removal of sensitized platelets by the reticuloendothelial system. Similarly, Handin et al showed that granulocytes from steroid-treated patients with ITP had a decreased ability to phagocytize antibody-coated platelets. 26 This inability of granulocytes may be related to defective granulocyte adherence in steroid treated patients. A decrease in plateletassociated IgG has been observed in ITP patients responding to corticosteroids. The platelet count often rises before PaIgG falls, reflecting the decreased ability of splenic leukocytes to phagocytize platelets in a steroid responsive patient as demonstrated by McMilían et al.27 As demonstrated by Sartorius corticosteroids also correct capillary fragility in children with ITP.24
In our clinic, we use the following regimen to treat acute ITP of childhood. 28
For children over 2 years of age:
60 mg prednisone/day po in divided doses: week #1
40 mg prednisone/day po in divided doses: week #2
20 mg prednisone/day po in divided doses: week #3
10 mg prednisone/day po in divided doses: week #4
For children under 2 years of age:
40 mg prednisone/day po in divided doses: week #1
30 mg prednisone/day po in divided doses: week #2
20 mg prednisone/day po in divided doses: week #3
10 mg prednisone/day po in divided doses: week #4
The doses of prednisone are tapered irrespective of the platelet counts. Prednisone is discontinued at the end of 4 weeks regardless of response. If there is no response or the platelet count falls again to less than 30,000 per mmr3, a single repeat 4-week course of prednisone is given.
In the event of conventional-dose prednisone failure, we have used intravenous high-dose solumedrol.29 Short-term increases in the platelet counts occur in almost all of the patients during administration of IV high dose solumedrol therapy. A 7-day course of solumedrol, 500 mg/m2 IV in 3 divided doses was given. The dose was tapered over the next 7 days to discontinue totally. The side effects of this therapy include hyperglycemia, glycosuria, hypertension, flushing and mood fluctuations.
INTRAVENOUS GAMMA GLOBULIN
Imbach et al wete the first to report the efficacy of highdose intravenous gammaglobulin (IvIg) in treatment of ITP in childhood. 30 In this report, they documented a rise in the platelet counts within 5 days following infusion of IVIg in children with chronic or intermittent ITP. They made a serendipitous observation of the disappearance of a severe thrombocytopenia in two patients with congenital agammagldbulinemia who were given high dose IVIg.
The mechanism of action of immunogiobulin (IVIg) effect in treatment of ITP is yet unclear. However, several possibilities have been postulated. These include:
A) Reticuloendothelial system (RES) Fc receptor blockade: Fehr et al showed that following treatment with high-dose IVIg the clearance time of anti-D sensitized autologous red blood cells by RES was prolonged.31 This led them to conclude that there was a blockade of Fc receptors of RES resulting in a decrease in the rate of destruction of antibodycoated platelets. Salama et al elaborated further on the mechanism of blockade at the RES Fc receptor site and suggested that it was due to coating of autologous red blood cells by the administered IgG (ie, coating of red blood cells with autoimmune antibodies such as anti-A anti-B). 32 Since red blood cells coated with IgG far outnumbered the other blood cells, they preferentially were sequestered in RES culminating into saturation of macrophages and their inability to phagocytose antibody-coated platelets. Even though this hypothesis is attractive, its validity has been in question. According to Bussel et al, RES blockade may be due to two separate effects: first, a decrease in the Fc receptor affinity for the IgG coating platelet from patients with ITP independent of the serum IgG level and, second, competition of Fc receptors by the increased serum IgG.33
B) Elimination of circulating microbial antigens: for example clearance of persistent viral infection by infusion of specific antibody may remove the underlying cause for ITP. This attractive hypothesis lacks evidence at the present time.
C) Decrease in autoantibody synthesis:33 This hypothesis also lacks proof and hence at the present time is under investigation. It is thought that IVIg activates T-cells while suppressing B cells.
D) Protection of platelets by high IVIg from platelet antibody:33 This mechanism does not seem to play a role because normal platelets with or without previous in vitro incubation with IVIg showed no difference in survival when given to a patient with ITP.
Recent indications for the use of IVIg in treatment of childhood ITP are still not well-defined. However, they have been found to be of benefit in:
A) Chronic ITP: Studies from several centers have shown that splenectomy could be postponed or avoided in children with chronic ITP with the infusion of IVIg, 3*-36 Intravenous gamma globulin (Sandoglobulin, Pharmaceutical Division, Sandoz, Inc., East Hanover, NJ, or Gamimune, Cutter Biological, Berkeley, CA) in a dose of 400 mg/kg per day for 5 consecutive days is administered according to the manufacturers' prescriptions. Some children improved slowly to the point of requiring no further treatment with IVIg. Thus, in chronic ITP, IVIg can be used as an alternative to splenectomy. Interestingly, it has also been found to be a useful mode of treatment in children with chronic ITP who previously failed to respond to splenectomy.
In chronic ITP, the use of IVIg has also enabled children to come off immunosuppressive drugs.
Following treatment with IVIg, platelet function has been shown to be normal. Use of IVIg is also indicated to increase platelet counts of patients with ITP prior to surgery.
B) Acute ITP: Imbach et al recently concluded that children with acute ITP respond more rapidly to IVIg than to prednisone and that the extent of initial response to IVIg has prognostic significance.36 However, these results are preliminary and the study lacks controls. Results of controlled trial in the future are being eagerly awaited to reach a final conclusion.
In summary, IVIg is a useful alternative to splenectomy. It can also be used as an adjunctive therapy to raise platelet counts rapidly in the management of serious bleeding resulting from acute ITP.
Although adverse reactions to IVIg occurs in less than 1% of patients who are not immunodeficient, they deserve mention. As with all blood products containing IgA, IVIg is contraindicared in patients with selective IgA deficiency. Headaches have been reported in patients receiving IVIg. Recently, Lever et al have reported non-?, non-B hepatitis occurring in agammaglobulinemic patients after treatment with IVIg.'7 Bone marrow aplasia developed in one patient following the occurrence of non-?, non-B hepatitis.
Approximately 10% to 15% of children with ITP fail to achieve a permanent remission and continue to remain thrombocytopenic. In this group of children with ITP, optimal management is difficult. The benefits of splenectomy in a given patient have to be weighed against its risks. The following general guidelines outlined are helpful in making that decision:
* The procedure is always indicated for control of lifethreatening hemorrhages such as intracranial hemorrhage, gastrointestinal hemorrhage, acute menorrhagia unresponsive to hormonal therapy.
* Splenectomy is indicated for patients who persistently have platelet counts of less than 50,000/mm3 for 12 months after the diagnosis of ITP is made despite corticosteroid and IVIg therapy.
* Patients with high levéis of physical activity are also candidates for splenectomy even if they are asymptomatic.
Because of the risk of post-splenectomy overwhelming sepsis, a patient is given pneumococcal vaccine prior to splenectomy and then maintained indefinitely on prophylactic penicillin by mouth. Any febrile illness or infection is attended with great caution and vigorous attempts are made to identify the type of underlying infection. Prompt treatment is instituted routinely with antibiotics such as ampicillin and tobramycin. Splenectomy is successful in treatment of chronic ITP in about 75% of patients. The rationale for splenectomy is based on the fact that the spleen is the most important site of destruction of platelets in ITP and is also one of the major sites responsible for the production of antiplatelet antibody.
Every attempt should be made to avoid splenectomy in with ITP under 2 years of age. However, this prois inevitable if a life-threatening hemorrhage complithe course of these infants.
NON-STEROIDAL IMMUNOSUPPRESSIVE AGENTS
The use of non-steroidal immunosuppressive agents is only under exceptional circumstances in treatof ITP of childhood. Such an indication exists when in splenectomized patient a bleeding tendency continues platelet counts of less than 20,000 to 50,000 per mm3, steroid and IVIg treatment.
In the past, cyclophosphamide,38 vincristine,39 aiahave been used in treatment of patients with ITP. Another interesting approach was reported by Ahn et al who attempted to selecrively enhance delivery of vinblastine by administration of vinblastine- loaded platelets to macrophages, the cells responsible for platelet destruction in ITP.41 This therapy led to complete remission in 50% of the adults with ITP. The use of such therapies in pediatrie patients should be undertaken with great caution and reluctance. Toxicities of these drugs and long-term side effects as oncogenesis should be weighed against their benefits. In short, they should be avoided by all means. Always a trial with IVIg is warranted before their use.
In conclusion, we recommend that:
* A trial of prednisone with standard dose be given to all children with ITP presenting with platelet counts of less than 50,000 per mm3 and/or bleeding manifestations.
* Failure of conventional dosage steroid may be followed by short-term high-dose Solumedrol.
* In patients refractory to prednisone, IVIg should be used. In case of poor or no response to IVIg1 the combination of both steroids and IVIg should be tried.
* In patients who are at risk of injuries due to increased physical activities and who have chronic ITP which is nonresponsive to steroid and/or IVIg, serious consideration should be given to splenectomy.
* In patients withserious life-threatening bleeding, eg, in treatment of ICH: high-dose solumedrol should be administered and splenectomy be performed. Administration of IVIg is also warranted. In these situations, eg, in life-threatening menstrual bleeding, we have found one dose of vincristine added to the above regime useful.
* In menstruating females, Depo-provera or any other long-acting progesterone should be used to suspend menstruation during the phase of severely low platelet count to prevent excessive menstrual bleeding.
1. McMillan R: Immune thrombocytopenia. Clin Haematol 1983; 12:69.
2. Luther JM. Emami A, Ravindranath Y, et al: ldiopathic chrombocytopenic purpura in children - The case fot management withoui corticosteroidi. Am J Pediatr Hermatol Oncol 1984; 6:149.
3. Woemer SJ, Abildgaard CT, French BN: Intracranial hemorrhage in children wirh idiopathic thrombocytopenic purpura, ftdianics 1981; 67:453.
4. Stuart MJ, McKenra R: Diseases of coagulation - The platelet and vastulatute in hemanJogy of infancy and childhood. Philadelphia, WB Saunders Co. 1981, p 1234.
5. Matoth Y, Zaiiov R. Frankel JJ: Minimal cerebral dysfunction in children with chronic thrombocytopenia. Pediatrics 1971; 47:698.
6. Aspnts GT, Kanon HA, Spencer RP, et al: Recurrent idiopathic thiombocytopenic purpura with "accessory" splenic tissue. Pediatrics 1975; 44:131.
7. Verheyden CN. Beart RW, Clifton MD. et al: Accessory splenectomy in the management of recurrent thiombocytopenic purpura. Mayo Clin Proc 1978; 53:442.
8. Beaidsley DJS, Spiegel JE, Jacobs MM. el al: Identification of antigen-bearing proteine - Localization of the target antigen in a subset of patients with idiopathic (or immune) thrombocytopenia purpura. Blood 1982-, 60Suppl I);l84a.
9. Beaidsley DJS, Spiegel JE, Jacobs MM, el al: Platelet membrane glycoprotein IHa contains target antigens that bind an tip la te I et antibodies in immune t h rom - bocytopeniai. J Clin Invest 1984:74:1701.
10. Mason D, McMilian R: Platelet antigens in chronic idiopathic thmmhocytopenic purpura. BrJ HaemotoJ 1934, 56:529.
11. McMillantt, Longmire RL, Yelenosky R, étal: Quantität ion uf plaie let -bind ing IgG produced in vino by spleens from patients with idiopathic thrombocy topen ic purpura. N Engl J Med 1974i 291 ;812.
12. McMillan R: Chronic idiopathic thrornboc y tópeme purpura. N EngíJ Med 1981; 304:1135, 1981.
13. Shannon KM, Buchanan GR, Fink CW, et al· Lymphocyte populations in childhood idiopathic thrombocytopenic puipura. AmJDuCWd 1984; 138:64.
14. Harker LA, finch CA: Thtombokinetics in man. J CIm Im*« 1969; 48;963.
15. Harker LA; Thrombokinetics in idiopathic thrombocytopenic purpura. BrJ Haanaial 1970; 19:95.
16. Branehogl, KuttiJ, Ridell B. et al.· The relation of thrombokinetics to bone mairow megakaryocytes in idiopathic thrombocytopenic purpura (GG?). Blood 1975; 45:551.
17. Siegel RS, Coleman RE, Kurland« R, étal: Platelet turnover - an important fac tor in predicting response to splenectomy in autoimmune thrombocytopenic purpura. Blood 1984; 64:5(Suppl 1) 241a. Abstract 873.
18. Garg SK, Amorosi EL. Karpadtin S: Use of the megathrombocyte as an index of megakaryocyte number. N Engt J Med 1971; 284:11.
19. Kelton JG: The measurement of platelet-bound immunoglobulins - an overview of the methods and the biological relevance of platelet-associated IgG, in Brown EB ed. Hematalog). Vol XIlI, New York. Gruñe and Stratum: p 163, 1983.
20. Dixon R. Rosse W. Ebbert L: Quantitative determination of antibody in idiopathic thrombocytopenic purpura. N EnglJ Med 1975; 292:230.
21. Morse BX, Giuliani D, Nussbaum M: Arapidquantitatbonof plate let- associated IgG by nephelometry. AmJ Hematol 1902; 12:271.
22. Kaninglon W], Minnich V, Hollingsworth JW, et al: Demonstration of a thrombocytopenic factor in the blood of patients with thrombocytopenia purpura. J Lob Clin Md 1951; 38:1.
23. Dunn NL, Maurer HM: ftednisone treatment of acute idiopathic thrombocytopenic purpura of childhood - advantages. AmJ ftditur Hematol Oncol 1984; 6:159.
24. Sartori us JA: Steroid treatment of idiopathic thrombocytopemc purpura m children - preliminary results of a randomtteJ cooperative study. Am J físitofr Hemotoi Oncol 1984; 6:165.
25. Buchanan GR, Holtkamp CA: Prednisone therapy for children with newly diagnosed idiopathic thrombocytopenic purpura - a randomized trial. Am J ftdiorr HfmmoJ Oncol 1984; 6:355.
26. Handin Rl, Siussel TP: Effect of corticosteruid therapy on the phagocytosis of antibody-coated platelets by human leukocytes. Qbcd 1978; 51:771.
27. McMillan R, Longmire RL. Tavassoli M, et al: In vino phagocytosis by splenic leukocytes in idiopathk thrombocytopenic purpura. N Engl J Med 1974; 290:249.
28. Lanzkowilcy P: Disorders of Platelets in rWiatrK HemutoJcgyOncology-A Treanw ^n ttie amidon. New York, McGraw-Hill Book Co., 1980, p 250.
29. Lanikowiky P, Kapoor V, Shende A, et al: High dose steroid therapy (HDST) in the treatment of chronic idiopathic thrombocyiopenic purpura (ITP). ftdion Ra 1984; abstract 900, ? 2?7.
30. lmbach P, d'Apuuo V, Hilt A. et al: High dose intravenous gammaglobulin for idiopathic thrombocytopenic purpura in childhood. Lancet 1981; I 1228.
31. Fehl), Hofmann V, Kappeier U: Transient reversal of thrombocytopenia in idiopathic thrombocytopenic purpura by high-dose intravenous gamma globulin. N Engl J Med 1982; 306:1254.
32. Salaroa A, Mue. Her- Eckhardt C, Kictel V: Effect of intravenous immunoglobulin in immune thrombocytopenia - competitive inhibition of reticulo-endothelial system function by sequestration of autologous red blood cells. Lancet H 1933; 193.
33. Bussel J1 Hilgartner MW: The use and mechanism of action of intravenous immunogtobulin in the treatment of immune haematologic disease. BrJ HacmauA 1954; 56:1.
34- Bussel JB, Schulman I, Hilgartner M, et al: Intravenous use of gammaglobulin in the marment of chronic immune thiombocytopenic purpura as a means to defer splenectomy. J Miau 1983; 103:651.
35. flforrier I, Lusher JM: Intravenous gammaglobulin treatment for chronic idiopathic thrombocytopenic purpura in children. AmJ Med 1984; 77:19Î.
36. lmbach P, Barandun 5, Hirt A, et al: Intravenous immunoglobulm for idiopathic thrombocytopenic purpura (ITP) in childhood. Am J Rwdiarr HemaioJ OncoJ 1984; 6:171.
37. Lever AML, Brown D. Webster ADB, et al: Non-?. non-B heparitis occurring in agammaglobulinemic patients after intravenous immunoglobulm. Lancet iI 1984; 1062.
36. VerlinM, Laro* PK, ftmwtJ^Treatmentof refractory thrombocytopenic purpura with cyclophosphanude. AmJ Hematol 1976; 1:97.
39. Ahn YS, Hartington W], Seelman RC: Vincmrine therapy of idiopathic and secondary thrombocytopenias. NEngfJMed 1974; 291:376.
40. Hilgurner MW, Lanikowiky P: The use of aiathioprme t? reftactory idiopalhic thrombocytopenias purpura in children. Acta Pediatrics Scand i970: 59:409.
41. Ahn YS, Bymes JJ, Harrington W, et al: The treatment of idiopathic thrombocytopenia with vinblastine- loaded platelets. N EngíJ Med 1978; 298:1101.
CLINICAL MANIFESTATION OF ITP
DIFFERENTIATING FEATURES BETWEEN ACUTE AND CHRONIC CHILDHOOD ITP4