Immune thrombocytopenia (ITP) is one of the most common immune-mediated diseases in children and adults. ITP was previously called idiopathic thrombocytopenic purpura, as platelet destruction is caused by an immune response. ITP is defined as isolated thrombocytopenia without other thrombocytopenia-associated causes [1]. The symptoms of ITP include petechiae, ecchymosis, internal cranial hemorrhage, and hematochezia. In general, ITP is a pathogenesis in which platelets bound to autoantibodies are phagocytosed by macrophages in the spleen. Autoantibodies are then mass-produced by b cells, thereby accelerating platelet destruction [2]. Impairment of regulatory T cells and decreased platelet production due to inhibition of megakaryocyte differentiation and maturation are important causes [3-5].

Treatment may include corticosteroids, intravenous immunoglobulin (IVIG), or anti-Rhesus-D immunoglobulin, either alone or in combination. IVIG is the most common treatment [6]. Most children experience an acute clinical course of the disease. Approximately 70% of cases resolve within 6 months [7]. However, approximately 20% of pediatric patients progress to chronic ITP [1]. There are no sex differences in children with an abrupt onset [8]. ITP is often preceded by an infection in children. By contrast, adults have an insidious onset, and it is twice as common in women as it is in men. Antecedent infections are also common. Approximately 50% of cases progress to chronic disease [9]. This study aimed to identify chronicity predictors of pediatric primary ITP.

1) Patients

This study was approved by the Institutional Review Board (IRB) of Yeungnam University Medical Center (IRB No: YUMC 2022-01-048) on February 4, 2022. All procedures performed in studies involving human participants were in accordance with the ethical standards of the institutional and/or national research committee and the 1964 Helsinki Declaration and its later amendments or comparable ethical standards. Formal consent was not required for this type of study.

This study was conducted retrospectively with patients diagnosed with between January 1, 2010, and September 30, 2021. The patient population was defined as those aged <18 years, with a platelet count of <20,000/mL at the time of diagnosis, and who were treated with IVIG ±steroids (methylprednisolone, prednisolone, and dexamethasone). The IVIG dose administered was 2 g/kg over 2 days; however, the dose was reduced in some patients who experienced infusion-related adverse reactions, such as fever, headache, and allergic reactions. Patients with insufficient medical records and those diagnosed with hematologic diseases other than ITP were excluded from this study.

As for the policy of ITP treatment in our institution, IVIG was used when there was clinical bleeding with platelets less than 30,000/mL or platelets less than 20,000/mL. Steroid was used according to the decision of the phy-sician.

2) Definition

The phase and response criteria for ITP used in this study were previously defined by an international working group [1]. ITP patients diagnosed within 3 months were defined as in the newly diagnosed (ND) phase. Patients who had been diagnosed for more than 3 months and less than 12 months, and did not reach spontaneous remission or did not maintain a complete response off therapy were defined as in the persistent (P) phase. Patients with ITP lasting >12 months after diagnosis were defined as in the chronic (C) phase.

Treatment response was classified into three categories: complete response (CR), response (R), and no response (NR), considering both platelet counts and bleeding episodes. If the platelet count was >100,000/mL and there were no bleeding episodes, the treatment response was defined as CR. If the platelet count was ≥30,000/mL and bleeding was absent, but the baseline count increased by at least two-fold, the treatment response was defined as R. If the platelet count was less than 30,000/mL or increased by less than twice the baseline count, the treatment response was defined as NR.

Respiratory and gastroenteritis that occurred within 8 weeks of ITP diagnosis was defined as recent infection. Bleeding episodes include any bleeding that appears to be associated with thrombocytopenia, intracranial hemorrhage, epistaxis, melena and petechiae, bruise, and ecchymosis.

3) Statistical analysis

IBM SPSS version 25.0 (IBM Corp., Armonk, NY, USA) was used for all data analysis. All values are expressed as means and standard deviation. Patients were divided into three groups according to ITP phase, and the Kruskal-Wallis test was performed to compare continuous variables between the three groups. A linear-by-linear association was used to compare the distribution of proportional data among the three groups. The Mann-Whitney test was used to compare continuous variables between the 3-month and 12-month groups. The chi-square test was used to compare the distribution of proportional data between the two groups. Logistic regression analysis was used to analyze the risk factor for chronic ITP. Statistical significance was set at P<0.05.

1) Patient characteristics

A total of 45 patients participated in this study (Table 1). The mean age of the patients was 4.07±3.98 years. Among them, 36 (80%) were under 6 years of age, and 9 (20%) were aged 6 years or older. There were 25 male patients (55.6%) and 20 female patients (44.4%). Eighteen patients (40%) had a history of infection within 8 weeks of ITP diagnosis. The mean dose of prescribed IVIG was 2.00±0.07 g/kg. IVIG was administered at <2 g/kg to 12 patients (26.7%). Fourteen (31.1%) patients were prescribed IVIG alone, and 31 (68.9%) patients received additional steroids.

Table 1 . Characteristics of patients.

Variables	n (%)
No. of patients	45 (100)
Age (yr)	4.07±3.98
Age group (yr)
<6	36 (80.0)
≥6	9 (20.0)
Sex
Male	25 (55.6)
Female	20 (44.4)
Recent infection history with 8 weeks
Yes	18 (40.0)
No	27 (60.0)
IVIG dose (g/kg)	2.00±0.07
IVIG dose (g/kg)
<2	12 (26.7)
≥2	33 (73.3)
Medication
IVIG only	14 (31.3)
IVIG+steroid	31 (68.9)
Laboratory finding
WBC (/mL)	9,350±3,450
Neutrophil (%)	40.3±12.7
Lymphocyte (%)	49.1±12.5
Platelet (/mL)	7,000±5,000
ESR (mm/hr)	10±12
CRP (mg/dL)	1.95±11.41
LDH (IU/L)	593±131

IVIG, intravenous immunoglobulin; WBC, white blood cell; ESR, erythrocyte sedimentation rate (normal range: 0-20 mm/hr); CRP, C-reactive protein (normal range: <0.5 mg/dL); LDH, lactate dehydrogenase (normal range: 208-450 IU/L)..

The mean age of the patients seems to be higher in the group with a longer ITP duration; however, the difference was not statistically significant (P=0.084) (Table 2). In the ND group, the proportion of patients aged under 6 years was relatively high (89.3%, 25/28), whereas in the chronic group, the proportion of patients aged under 6 years was relatively low (54.5%, 6/11) (P=0.028) (Fig. 1). The prescribed IVIG dose was significantly lower in the chronic group (P=0.024) (Fig. 2). In the ND group, the proportion of patients with an IVIG dose of ≥2 g/kg was relatively high (85.7%), whereas in the chronic group, the proportion of patients with an IVIG dose of ≥2 g/kg was relatively low (45.5%, P=0.011). There was no statistically significant difference between sexes in the ITP phase. There were no statistically significant differences in the use of additional steroids, laboratory findings, or bleeding episodes between groups.

Table 2 . Characteristics of patients according to disease phase.

Variables	Newly diagnosed	Persistent	Chronic	P-value
No. of patients	28	6	11
Age (yr)	3.0±2.6	4.1±2.1	6.9±6.1	0.084
Age group (yr)				0.028
<6	25 (89.3)	5 (83.3)	6 (54.5)
≥6	3 (10.7)	1 (16.7)	5 (45.5)
Sex				0.864
Male	15 (53.6)	5 (83.3)	5 (45.5)
Female	13 (46.4)	1 (16.7)	6 (54.5)
Infection history				0.293
Yes	15 (53.6)	4 (66.7)	8 (72.7)
No	13 (46.4)	2 (33.3)	3 (27.3)
IVIG dose (g/kg)	2.02±0.06	1.99±0.03	1.96±0.09	0.024
IVIG dose (g/kg)				0.011
<2	4 (14.3)	2 (33.3)	6 (54.5)
≥2	24 (85.7)	4 (66.7)	5 (45.5)
Medication				>0.999
IVIG only	9 (32.1)	1 (16.7)	4 (36.4)
IVIG+steroid	19 (67.9)	5 (83.3)	7 (63.6)
Laboratory finding
WBC (/mL)	9,800±3,220	7,850±2,930	9,010±4,240	0.273
ALC (/mL)	5,075±2,243	3,795±2,013	4,203±2,657	0.189
Neutrophil (%)	38.7±11.4	40.9±12.6	44.4±16.2	0.215
Lymphocyte (%)	51.5±11.0	46.9±10.1	44.3±16.1	0.105
PLT (/mL)	7,000±4,000	6,000±5,000	9,000±6,000	0.265
ESR (mm/hr)	9±11	5±4	14±16	0.499
CRP (mg/dL)	0.21±0.31	0.31±0.35	6.99±22.56	0.950
AST (IU/L)	46±37	33±6	31±10	0.301
ALT (IU/L)	33±53	15±3	16±6	0.665
LDH (IU/L)	609±113	601±74	553±187	0.570
Bleeding episode
Epistaxis	3 (10.7)	1 (16.7)	1 (9.1)	>0.999
Petechiae, bruise	22 (78.6)	6 (100)	9 (81.8)	0.829
GI bleeding	4 (14.3)	0 (0)	0 (0)	0.246
No bleeding	5 (17.9)	0 (0)	2 (18.2)	>0.999

IVIG, intravenous immunoglobulin; WBC, white blood cell; ESR, erythrocyte sedimentation rate (normal range: 0-20 mm/hr); CRP, C-reactive protein (normal range: <0.5 mg/dL); LDH, lactate dehydrogenase (normal range: 208-450 IU/L); ALC, absolute lymphocyte count; AST, aspartate aminotransferase (normal range: 10-35 IU/L); ALT, alanine aminotransferase (normal range: 0-40 IU/L); GI, gastrointestinal; PLT, platelet..

Figure 1. Relationship between age and disease phase. (A) Number of patients by disease phase. (B) Number of patients with disease phase by age group.

Figure 2. Immunoglobulin dose according to disease phase.

In the ND group (n=28), the mean age was 3.0±2.6 years, while in the P+C group (n=17), the mean age was 5.9±5.2 (P=0.026), respectively (Supplement Table 1). Of the patients aged under 6 years, there were 25 (89.3%) and 11 (64.7%) in the ND and P+C groups, respectively (P=0.063). In the ND group, the IVIG dose was 2.02±0.06 g/kg, while in the P+C group, the dose was 1.97±0.07 g/kg (P=0.010). Four patients in the ND group (14.3%) and eight patients in the P+C group (47.1%) received an IVIG dose <2 g/kg (P=0.034, OR=0.188). There were no significant differences between the ND and P+C groups in terms of sex, infection history, steroid combination therapy, laboratory findings, or bleeding episodes.

2) Response to treatment

As shown in Table 3, we analyzed the prognosis according to platelet response on specific days after IVIG infusion.

Table 3 . Prognosis according to treatment response by period.

Time	Response	Newly diagnosed (n=28)	Persistent+chronic (n=17)	P-value	OR	95% confidence interval
D+3	CR vs. non-CR
	CR	15 (53.6)	7 (41.2)	0.542	1.648	0.487-5.574
	Non-CR	13 (46.4)	10 (58.8)
	R vs. NR
	R	23 (82.1)	16 (94.1)	0.385	0.288	0.031-0.700
	NR	5 (17.9)	1 (5.9)
D+14	CR vs. non-CR
	CR	25 (89.3)	4 (23.5)	<0.001	27.083	5.253-139.636
	Non-CR	3 (10.7)	13 (76.5)
	R vs. NR
	R	27 (96.4)	11 (64.7)	0.008	14.727	1.583-136.971
	NR	1 (3.6)	6 (35.6)
D+30	CR vs. non-CR
	CR	26 (92.9)	6 (35.3)	<0.001	23.833	4.147-136.973
	Non-CR	2 (7.1)	11 (64.7)
	R vs. NR
	R	27 (96.4)	14 (82.4)	0.144	5.786	0.550-60.875
	NR	1 (3.6)	3 (17.6)

Time indicates the number of days after IVIG infusion (D)..

CR, complete response; R, response; NR, no response; OR, odds ratio..

Three days after IVIG, 15 (53.6%) patients in the ND group and seven (41.2%) patients in the P+C groups showed CR (P=0.542). Twenty-three (82.1%) and 16 (94.1%) patients showed R in the ND and P+C groups, respec-tively (P=0.385).

Fourteen days after IVIG, 25 (89.3%) patients in the ND group and 4 (23.5%) patients in the P+C group showed CR (P<0.001, OR=27.083). Twenty-seven (96.4%) and 11 (64.7%) patients showed R in the ND and P+C groups, respectively (P=0.008; OR=14.727).

At 30 days after IVIG, 26 (92.9%) patients in the ND group and four (35.3%) patients in the P+C group showed CR (P<0.001, OR=28.833). Twenty-seven (96.4%) and 14 (82.4%) patients showed R in the ND and P+C groups, respectively (P=0.144).

3) Risk factor for chronic ITP

In univariate analysis of risk factor for chronic ITP, age 6 years or older and IVIG dose <2 g/kg were found to be risk factors for chronic ITP, and these two factors were also confirmed as significant risk factors in multivariate analysis (Table 4, 5).

Table 4 . Univariate analysis of risk factor for chronic ITP.

Variables	P-value	Odds	95% confidence interval
Age ≥6 yr	0.023	6.250	1.287-30.349
Female	0.44	1.714	0.4366-6.742
Infection history within 8 weeks	0.327	2.105	0.475-9.338
IVIG dose <2 g/kg	0.022	0.179	0.041-0.783
IVIG only	0.666	0.729	0.174-3.056

Table 5 . Multivariate analysis of risk factor for chronic ITP.

Variables	P-value	Odds	95% confidence interval
Age >6 yr	0.026	0.156	0.030-0.803
IVIG dose ≥2 g/kg	0.027	7.235	1.252-41.821

This study investigated predictors of chronic ITP by performing a retrospective study of patients from a single institution.

Regarding the association between the prognosis and age of chronic ITP, Higashide et al. reported that age ≥23 months and platelet count <9,000/mL were poor prognostic factors for short-term response [10]. Ahmed et al. found that age >8.5 years, WBC count <6,250/mL, and absolute lymphocyte count (ALC) <3,050/mL were associated with a significant risk of developing persistent ITP beyond 6 months [11] . Glanz et al. reported that children whose illness was diagnosed at ≥10 years of age and who had a platelet count ≥20,000/mL had an approximately five-fold risk of progressing to chronic disease compared with children who were under 2 years of age at diagnosis with a platelet count <20,000/mL [12]. A recent meta-analysis conducted by Heitink-Pollé et al. revealed that older age at presentation is one of the strongest risk factors for chronic ITP [13].

In our study, there were 6.25 times more patients with chronic ITP aged over 6 years than those aged under 6 years (P=0.028, OR=6.25). In addition, WBC count and ALC did not appear to affect persistent or chronic ITP in the present study. Although there are differences between the studies, it appears that the probability of progression to chronic ITP increases with age.

Several studies have proposed that acute ITP is more closely related to a history of recent infection than chronic ITP. The association between recent infection history and acute ITP has been controversial [14-16]. However, Choi et al. reported that a recent history of infection was not statistically related to the disease phase [15]. In our study, the proportion of patients with a history of infection within the last 8 weeks was higher in the chronic group than in the ND group. However, this difference was not statistically significant (72.7% vs. 53.6%, P=0.293).

In several ITP studies, IVIG was used as a protocol with a 1 g/kg/day infusion for 1-2 days [17]. The recent American Society of Hematology guidelines recommend a single dose of 0.8 g/kg [7].

Morimoto et al. and Higashide et al. conducted studies in which the IVIG treatment doses were 2 g/kg and 1.4 g/kg, respectively, and found no difference in IVIG response between the two studies [10,18]. Choi et al. compared IVIG doses of 1 g/kg/day and 2 g/kg/day, and reported that the treatment effect was similar even at a low dose with fewer side effects [19].

In a study by Kim et al., there was no significant difference in the dose of IVIG treatment for patients with ITP in the recovery of platelet counts [20]. Parodi et al. reported that IVIG had no effect on the natural outcome of the disease [21]. In a previous study conducted by Benesch et al. on patients with acute child ITP, the group administered with IVIG 2 g/kg had a higher platelet level on the third day than the group administered 0.6 g/kg [22].

In our study, the therapeutic dose of IVIG was set on the basis of a dose of 2 g/kg, where the IVIG dose was gradually decreased in patients with ND, persistent, and chronic diseases (P=0.024). In addition, the proportion of patients receiving IVIG ≥2 g/kg in the ND group was higher than that in the chronic group (P=0.011).

In a study of prognostic factors for pediatric ITP conducted by Jung et al., among ITP patients who received IVIG for treatment, the proportion of patients with platelet counts exceeding 45,000/mL after 1 month of treatment was significantly higher in the acute ITP group [14]. Moreover, in a study conducted by Choi et al., patients with ITP who received IVIG had a significantly lower probability of progressing to chronic disease when their platelet count was restored to 100,000/mL within 3 months, than those who did not restore their platelet count [15]. Kim et al. reported that the diagnosis of chronic ITP was high in the group with a slow response to IVIG treatment; therefore, it could be used as an early predictor of chronic ITP [20]. In a study by Elalfy et al., the response to platelet treatment was prolonged in patients with chronic ITP compared to that in patients with acute ITP [23].

In our study, the proportion of patients with CR on days 14 and 30 after IVIG infusion in the ND group was higher than those in the persistent and chronic groups.

A limitation of our study is that it was a retrospective study conducted with a small number of patients at a single center.

In conclusion, in our study, patients over 6 years of age, and those who received IVIG doses less than 2 g/kg were more likely to progress to chronic ITP.

This work was supported by a 2021 Yeungnam University research grant.

Supplementary materials can be found via https://doi.org/10.15264/cpho.2022.29.2.44.

The authors have no conflict of interest to declare.