Immune thrombocytopenia in infants: a retrospective study with comparison to toddlers

Abstract

Introduction:

Immune thrombocytopenia (ITP) is the most common cause of acquired bleeding in childhood, but little is known about the clinical course and outcomes in infants with ITP.

Methods:

This is a retrospective study of all infants (1–12 months of age) and toddlers (13–47 months of age) diagnosed with ITP from a single centre during a 13-year period. The following data were compared between the two patients groups: demographics, severity of bleeding, platelet counts, duration of illness, development of chronic ITP, treatment and association with recent vaccination.

Results:

Twenty-two infants and 30 toddlers were diagnosed and followed up for ITP during the study period. Infants with ITP generally had minor or mild bleeding (19, 86.4%) and seldom required treatment (7, 31.8%), and their thrombocytopenia resolved at a mean of 1.90 months after diagnosis. Besides age, the sex ratio, severity of bleeding, platelet counts and proportion that required treatment were comparable between infants and toddlers. Fewer infants developed chronic ITP (1/22 vs. 9/30, P = 0.032), but more infants had a history of vaccination in the preceding 6 weeks prior to diagnosis of ITP (13/22 vs. 1/30, P < 0.001).

Conclusion:

ITP in infants is almost always a self-limiting and transient illness, and the majority of cases do not require treatment.

INTRODUCTION

Immune thrombocytopenia (ITP), previously called idiopathic thrombocytopenic purpura, is the most common form of acquired bleeding disorder in childhood.[1] Annually, the incidence of ITP is 1.9–6.4 per 100,000 children under the age of 15 years.[2,3] For a long time, ITP was defined by a platelet count <150 × 10⁹/L and the disease was classified into acute and chronic forms based on illness duration of <6 months and >6 months, respectively.[4] In 2009, an international working group redefined ITP as platelet count <100 × 10⁹/L.[5] Furthermore, the disease was recategorised into three classes. Acute and persistent ITP referred to illnesses that lasted for 0–3 and 4–12 months, respectively. Chronic ITP was defined as illnesses that persisted beyond 12 months after diagnosis. This renewed classification is now widely adopted and is more relevant to clinical prognostication and management.

Compared to adult patients, children with ITP have much better prognosis in terms of spontaneous remission and serious complications. By the end of the first year, 76% of children diagnosed with ITP will recover from the illness.[6] Life-threatening haemorrhages occur in less than 1% of children with ITP.[6] Even within the paediatric age ranges, there is considerable heterogeneity in the clinical manifestations of ITP. For instance, children over the age of 10 years generally have milder illnesses, but a higher propensity to develop chronic ITP.[7] Infancy beyond the neonatal period is another time period that deserves special attention in paediatric haematology and oncology. Yet, there have been very few observations dedicated to ITP in infancy. The following report summarises our experience.

METHODS

This was a retrospective study from a single tertiary referral and treatment centre of a private paediatric haematology and oncology practice in Singapore. The study period was from March 2007 to February 2021, inclusively. The study subjects included all children aged 1–47 months with a diagnosis of ITP. Their hospital and clinic charts were reviewed, and the following data were collected: sex, age at diagnosis, platelet counts at diagnosis, lowest platelet counts recorded during follow-up, treatment received, duration of illness upon resolution of ITP, severity of bleeding and exposure to vaccination in the preceding 6 weeks before the initial diagnosis of ITP.

The diagnosis of ITP was based on clinical signs and symptoms of bleeding and/or petechiae and the presence of isolated thrombocytopenia (platelet count <100 × 10⁹/L) on full blood count. Examination of the peripheral blood film was mandatory to exclude pseudothrombocytopenia, malaria, microangiopathic changes, leukaemia and myelodysplasia, Döhle bodies, Pelger–Huet anomaly and microthrombocytopenia. ITP was classified as acute if spontaneous resolution of thrombocytopenia (platelet count >100 × 10⁹/L) occurred at 3 months or less after diagnosis, as persistent if spontaneous resolution occurred from 4 to 12 months after diagnosis and as chronic if the disease persisted after 12 months from diagnosis. Bone marrow aspiration was not required for diagnosis, but it might be indicated if infiltrative marrow disease or amegakaryopoiesis was suspected. Subjects whose thrombocytopenia was determined to be congenital or secondary to another autoimmune disease, malignancy or bone marrow failure syndromes were excluded. Thrombocytopenia diagnosed within the first month of life was also excluded.

The severity of bleeding in each child was determined by the previously described Buchanan criteria [Table 1].[8] In essence, non-serious bleeding refers to either minor or mild grades in which bleeding is limited to petechiae and/or bruises. Serious disease includes moderate or severe grades where there is mucosal or internal bleeding, including life-threatening and intracranial bleeding.

Table 1.

Grading of bleeding severity in childhood immune thrombocytopenia.

Severity	Clinical features
Grade 1 (minor)	Few petechiae (≤100 total) and/or ≤5 small bruises (≤3 cm in diameter)
Grade 2 (mild)	Many petechiae (>100 total) and/or >5 large bruises (>3 cm in diameter)
Grade 3 (moderate)	Overt mucosal or gastrointestinal bleeding with/without petechiae or bruises
Grade 4 (severe)	Mucosal bleeding or internal bleeding that requires emergency attention and/or decrease in haemoglobin >2 g/dL

Adapted from Buchanan and Adix[8]

Subjects were categorised as infants (1–12 months old) and toddlers (13–47 months old) at the time of diagnosis. Demographic data, clinical and laboratory features were compared between the two subject groups. Parametric variables were compared by two-tailed Student’s t-test and non-parametric variables by Fisher’s exact test. Statistical significance was defined at a P value < 0.05.

RESULTS

Fifty-two children with ITP were identified during the study period [Table 2]. Twenty-two were infantile cases and 30 were toddlers. A slight male predominance (29/52, 55.8%) was seen, but the sex distribution did not differ between the two groups of children. Severity of bleeding, proportion of subjects who received treatment, platelet counts at diagnosis and lowest recorded platelet counts were comparable between infants and toddler patients. Among those whose disease remitted, the mean duration of thrombocytopenia was 1.90 months in infants and 4.92 months in toddlers, but no statistically significant difference was found. However, fewer infants developed chronic ITP compared to toddlers (1/22 vs. 9/30, P = 0.032).

Table 2.

Comparison of clinical and laboratory parameters between infants and toddlers diagnosed with ITP.

Parameter	Infants (n=22)	Toddlers (n=30)	P
Age (yr)	0.47±0.32	2.37±0.93	<0.001
Male	13	16	0.780
Severity of bleeding
Minor + mild	19 (9+10)	25 (17+8)	1
Moderate + severe	3 (3+0)	5 (4+1)
Platelet at diagnosis (×109/L)	19.64±22.20	26.77±27.00	0.316
Lowest platelet recorded (×109/L)	15.59±20.41	24.17±25.32	0.197
Chronic ITP	1	9	0.032
Duration of illness at resolution (mth)	1.90±1.57	4.92±10.06	0.180
Treatment given	7	11	0.775
Recent vaccination	13	1	<0.001

Data presented as n or mean±standard deviation. ITP: immune thrombocytopenia

A history of vaccination before the diagnosis of ITP was more often obtained among infants compared to toddlers (13/22 vs. 1/30, P < 0.001). The vaccinations included: Bacillus Calmette–Guérin (BCG) plus hepatitis B in three cases; the 5-in-1 vaccine with diphtheria, acellular pertussis, tetanus, inactivated poliomyelitis and Haemophilus influenzae B in three cases; the 5-in-1 vaccine, hepatitis B and pneumococcus vaccine in three cases; hepatitis B alone in two cases; the 5-in-1 vaccine, pneumococcus vaccine and rotavirus vaccine in one case; and mumps–measles–rubella vaccine in one case.

Bone marrow aspiration was done in three cases for exclusion of amegakaryopoiesis: an infant with pre-existing lymphatic malformation, a toddler with poor platelet response after treatment with intravenous immunoglobulin and corticosteroid, and a toddler with pre-existing central nervous system degeneration.

Among the 22 infantile cases of ITP, 15 (68.2%) did not require any treatment. Disease severity was not serious (minor or mild grade) in 19 (86.4%) cases, even though 14 (73.7%) of them had platelet counts ≤ 10 × 10⁹/L. Of the seven infants who received treatment, a single infusion of intravenous immunoglobulin (0.8–1.0 g/kg) in five infants and a single course of oral prednisolone (2 mg/kg/day for 1 week) in another child were all that was used during the first month of diagnosis. The only case of chronic ITP required sequential treatment with intravenous immunoglobulin and prednisolone. She was commenced on treatment with eltrombopag at 6 months after diagnosis. Treatment had been maintained for over 12 months to keep the platelet counts at 50–100 × 10⁹/L with excellent control of bleeding and quality of life.

DISCUSSION

Childhood ITP has the highest rates of occurrence during the first 5 years of life.[3,4] With a known association with vaccination,[9] ITP occurring in infancy is of no surprise to paediatricians as this is the time when most of the primary immunisations are given. What remains uncertain is whether children in the first year of life would suffer more severely from thrombocytopenia compared to older children. Infancy does not present any burden to the haemostatic system. Ambulation occurs late and is characterised by low-speed cruising or walking. Epidemiological studies and the author’s previous experience in children with or without ITP and spontaneous intracranial haemorrhages also do not find young age to be a risk factor.[10,11,12] Indeed, our experience as presented in this study shows that infants with ITP often have only minor or mild bleeding despite very low platelet counts.

The medical literature contains very few observations about ITP in infancy. Of the five articles published in the last 20 years,[13,14,15,16,17] four defined infancy as age up to 24 months old. A total of 445 children were included, and the proportion of patients with serious bleeding varied from 8% to 41% among the different series [Table 3]. Of interest, bone marrow aspiration was considered mandatory in three series,[14,16,17] even though current guidelines state that the diagnostic procedure is largely unnecessary in childhood ITP.[18] Also, the great majority of the cases were treated with immunomodulatory agents (422/445, 95%). Most were treated with intravenous immunoglobulin and/or corticosteroids, but up to 8% of the patients were treated with anti-D immune globulin, azathioprine or cyclosporine.[16,17] Thus, the published data suggest that infants with ITP are managed very aggressively in terms of diagnostic workup and there is a reluctance to withhold pharmacotherapy, which are different from our experience.

Table 3.

Published series of ITP in infancy after 2000.

Study, country	Infant definition	Number of patients	Serious bleeding	Intracranial haemorrhage	No treatment received	Chronic ITP	Bone marrow aspiration
Sandoval et al.,[13] USA	≤24 months	79 (26 ≤12 months)	6 (8%)	0	7 (9%)	7 (9%)	6
Wang et al.,[14] Taiwan	≤12 months	17	5 (29%)	0	0	0	All cases
Ramyar and Kalantari,[15] Iran	≤24 months	96 (78 ≤12 months)	16 (17%)	0	1 (1%)	Not evaluated	38
Hashemi et al.,[16] Iran	≤24 months	66 (52 ≤12 months)	27 (41%)	2	0	16 (24%) (10≤12 months, 6>12 months)	All cases
Farhangi et al.,[17]Iran	≤24 months	187	43 (23%)	3	15 (8%)	12 (6%)	All cases
Current study, Singapore	≤12 months	22	3 (14%)	0	15 (68%)	1 (5%)	1

ITP: immune thrombocytopenia

Even though infants and toddlers with ITP share the most common clinical and laboratory features, there are subtle differences despite the small age gap. First, a recent exposure to vaccination is more common among infantile cases. Second, infants rarely develop chronic ITP. Wang et al.[14] did not observe any case of chronicity in their series. Sandoval et al.[13] reported seven chronic ITP out of 79 cases, with a median age of 18 months. Farhangi et al.[17] reported 15 chronic ITP out of 187 cases, with a mean age of 533.8 days (about 18 months). Although the exact numbers were not specified, the majority of chronic ITP in these two series occurred in the second year of life. All these observations are consistent with ours and indicate that chronic ITP is hardly a complication of infant ITP.

In developed countries like Singapore and the USA, individuals receive most of their lifetime-recommended vaccinations during the first year of life.[19,20] In Singapore, an average child can receive a maximum of 26 vaccines in either single shots or combination vaccines before the first birthday. In the USA, up to 29 vaccines can be given during the same period. When ITP develops during infancy, it is difficult to tell if a recent exposure to a vaccine or several vaccines is aetiologically related. Case–control studies suggest that measles–mumps–rubella vaccine, given within 6 weeks before the emergence of thrombocytopenia, is the only vaccine that is significantly associated with the occurrence of ITP.[9,21] Even so, the attributable risk of developing ITP after the vaccine is one in 25,000 vaccinations.[21]

The major limitation in this study is a potential referral bias. As a tertiary referral and treatment centre, our practice may have missed patients at the mild end of the spectrum. Case number is relatively small, and hence, extrapolation of the findings to the general population has to be cautioned.

In conclusion, our experience shows that ITP in infancy deserves unique attention. It is generally a mild disease. The great majority of patients recover without treatment. The chance of developing chronic ITP is remote. More studies are required to expand our understanding of the impact of ITP in infants for its optimal management.

Conflicts of interest

There are no conflicts of interest.

Funding Statement

Nil.