A Retrospective Review of 1821 Cases of Eosinophilia in a Single‐Center Study in Eastern China

ABSTRACT

Objective:

Our study aims to explore the types of diseases that may cause eosinophilia and the changes in the disease spectrum of different degrees of eosinophilia.

Methods

We retrospectively analyzed patients who were admitted to the First Affiliated Hospital of Soochow University from July 1, 2023 to June 30, 2024, and had at least twice eosinophilia (absolute eosinophil count [ACE] > 0.5 × 109/L, the interval between the two tests was ≥ 3 days). Eosinophilia was classified as mild (1.5 × 109/L ≥ ACE > 0.5 × 109/L), moderate (5 × 109/L ≥ AEC > 1.5 × 109/L), or severe (AEC > 5 × 109/L).

Results

A total of 1821 patients were included, with a median age of 56 years (2–96 years), including 1252 males (68.8%) and a median ACE of 1.02 × 109/L (0.53–41.35). Here, 1679 cases were diagnosed with diseases that may be related to eosinophilia, allergic diseases were the most common (21.7%, 364/1679). Eight cases of clonal eosinophilia were diagnosed as myeloid/lymphoid neoplasms with eosinophilia and tyrosine kinase gene fusions (MLN‐TK). Around 91 patients were diagnosed as idiopathic eosinophilia after a series of screening. Here, 71.7% (1305/1821) of the patients showed mild eosinophilia, allergic diseases (21.7%, 283/1305) were still the main secondary factors, 23.6% (430/1821) of the patients showed moderate eosinophilia, hematological malignancies became the most common secondary factors (20.5%, 88/430), and 86 patients had severe eosinophilia, nearly half (40/86) of the patients were secondary to hematological malignancies.

Conclusion

Eosinophilia has a wide range of age, and is more common in males. Allergic diseases are the most common secondary factors. Different levels of eosinophilia have different causes.

Trial Registration: The authors have confirmed clinical trial registration is not needed for this submission.

1. Introduction

In 1879, Paul Ehrlich first described a group of cells that were strongly stained with eosin dye and named them eosinophils [1]. These cells originate from CD34 + hematopoietic stem cells, and their differentiation and proliferation are regulated by a variety of transcription factors (such as GATA‐1, PU.1, and c/EBP) and cytokines (including IL‐3, IL‐5, and granulocyte‐macrophage colony‐stimulating factor [GM‐CSF]). Then, under the regulation of IL‐5, they migrate from the bone marrow to the peripheral circulation [2, 3, 4]. Eosinophils stay in the peripheral circulation for only 18 h and then enter the tissue through the endothelium under the action of various chemokines and adhesion molecules, and accumulate in the mucosal tissues of the respiratory system, digestive system, and genitourinary tract [5, 6]. They play a key role in infection, allergy and inflammation by storing and secreting cytokines, chemokines and other mediators, and are involved in the initiation and progression of various diseases [7].

Normal human peripheral blood eosinophils account for about 3%–5% of the total number of white blood cells, equivalent to 0.35 − 0.5 × 10⁹/L [8]. Eosinophilia is defined as AEC > 0.5 × 10⁹/L. The incidence of eosinophilia varies depending on the study population and region [9, 10, 11, 12]. A large sample study has shown that about 4% of the 356,196 patients had detected eosinophilia [12]. Eosinophilia in peripheral blood is common in clinical practice. However, insufficient attention is paid to the eosinophil count in routine blood tests. Peju et al. found that nearly half of the patients with eosinophilia were neglected, and even the symptoms were not mentioned in the medical records [13]; in the study of doctors' cognition and attention to eosinophilia in China's large third‐grade Class‐A centers, it was found that only 19.4% of the 207 physicians who had diagnosed eosinophilia had conducted detailed examinations on patients [14]. Therefore, more research on the diagnosis and treatment of eosinophilia is needed to enhance clinicians' cognition, enabling them to make diagnosis and treatment decisions that are more conducive to patients' treatment efficacy. Therefore, the present study aimed to investigate the cause of eosinophilia using a large cohort of eosinophilia cases.

2. Methods

This is a retrospective cross‐sectional study, including patients with AEC > 0.5 × 10⁹/L detected at least twice in the First Affiliated Hospital of Soochow University from July 2023 to June 2024.

According to the degree of elevated blood AEC, patients were divided into the mild eosinophilia group (1.5 × 10⁹/L ≥ AEC > 0.5 × 10⁹/L), the moderate eosinophilia group (5 × 10⁹/L ≥ AEC > 1.5 × 10⁹/L), and the severe eosinophilia group (AEC > 5 × 10⁹/L) [15]. When multiple eosinophil values were recorded, the maximum eosinophil value was used to define the eosinophilia category.

According to the different causes of eosinophilia, it is divided into: Familial (hereditary) eosinophilia, secondary (reactive) eosinophilia, clonal (primary) eosinophilia, and idiopathic (undetermined) eosinophilia [8]. Causes of eosinophilia were obtained from the patients' medical records. Secondary (reactive) causes of eosinophilia included allergic diseases (such as asthma, allergic rhinitis, atopic dermatitis, allergic bronchopulmonary aspergillosis, urticaria, and eczema), infections (including parasites, viruses, bacteria, and fungi), solid tumors (lung, renal, and colon tumors), autoimmune disorders (e.g., inflammatory bowel disease, celiac disease, eosinophilic granulomatosis with polyangiitis, rheumatoid arthritis, sarcoidosis, systemic sclerosis, Sjogren's syndrome, bullous pemphigoid, IgG4‐related disease, eosinophilic fasciitis, primary biliary cirrhosis, and autoimmune thyroiditis), medications (such as Aspirin, NSAIDs, antimicrobials, and DRESS syndrome), hematologic malignancies (such as lymphoma, acute leukemia, and chronic leukemia), and so on [16]. When the cause is not easy to identify, it will be classified as idiopathic eosinophilia. Hypereosinophilic syndrome (HES) is defined as eosinophilia greater than 1.5 × 10⁹/L on at least two occasions with a minimum time interval of 4 weeks, along with organ damage and/or dysfunction attributable to tissue hypereosinophilia, and the exclusion of other disorders or conditions as the major reason for organ damage.

According to the guidelines, reverse transcription polymerase chain reaction (RT‐PCR) or RNA sequencing was used to detect MLN‐TK related fusion genes [17]. Imaging examinations were used to assess organ involvement, including cardiac evaluation by echocardiography, cardiac magnetic resonance imaging (MRI), and bronchoalveolar lavage (BAL), endoscopic evaluations were conducted for the gastrointestinal system. Some patients underwent percutaneous biopsy (skin, lung, or liver) combined with immunohistochemical detection to determine whether the tissue was infiltrated by eosinophils.

Each case was categorized by two independent diagnostic physicians (both senior attending physicians or above) based on the original medical records, with neither physician informed of the other's assessment. In our pilot study (comprising 100 randomly selected cases), the initial diagnostic concordance rate between the two physicians was 88.2% (κ > 0.81). For cases with inconsistent diagnoses, the reasons for discrepancies were documented. In cases of diagnostic discrepancy between the two attending physicians, a joint discussion shall be conducted to facilitate consensus‐building. If consensus remains unattainable, or if the case involves complex/rare etiologies, the case shall be referred to a third senior expert (chief physician) for independent review and final diagnosis. For cases with highly complex or atypical clinical manifestations, a multidisciplinary team (MDT) discussion shall be initiated, involving specialists from hematology, rheumatology and immunology, infectious diseases, respiratory medicine, dermatology, and other relevant disciplines, to determine the definitive etiology based on comprehensive clinical data and follow‐up outcomes.

Statistical analyses were performed in SPSS Statistics 29, and graphs were made using Prism GraphPad 10. All clinical and laboratory parameters are presented as medians and ranges for continuous variables and as frequencies or percentages for categorical variables. Comparisons between groups were carried out using the Kruskal–Wallis test for continuous variables and the χ ² or Fisher's exact tests for categorical data. Statistical significance was defined at the p < 0.05 level.

3. Result

3.1. Baseline Characteristics

A total of 1821 patients (1252 males, 68.8%; median age 56 years, range: 2–96 years) were included and grouped according to blood AEC: mild eosinophilia, 1.5 × 10⁹/L ≥ AEC > 0.5 × 10⁹/L (n = 1305, 71.7%); moderate eosinophilia, 5 × 10⁹/L ≥ AEC > 1.5 × 10⁹/L (n = 430, 23.6%); severe eosinophilia, AEC > 5 × 10⁹/L (n = 86, 4.7%). The patients in the severe group were significantly younger than those in the mild and moderate groups at the time of eosinophilia (p < 0.001, Table 1).

TABLE 1.

Demographic features of the study group.

	Total n = 1821	Mild eosinophilia (0.5 < AEC ≤ 1.5 × 109/L) (n = 1305)	Moderate eosinophilia (1.5 < AEC ≤ 5.0 × 109/L) (n = 430)	Severe eosinophilia (AEC > 5.0 × 109/L) (n = 86)
Male sex, n (%)	1253 (68.81%)	879 (68.73%)	303 (70.47%)	53 (61.63%)
Age at diagnosis median (year)	56 (2–96)	56 (2–96)	58 (4–90)	42.5 (8–86)*
Eosinophil count median (× 109/L)	1.02 (0.53–41.35)	0.88 (0.53–1.5)	2.04 (1.51–5)	7.88 (5.01–41.35)

*p < 0.05, with statistical difference.

3.2. The Etiology of Eosinophilia

Of the 1821 total cases reviewed, 1687 (92.6%) had a specific diagnosis that could be linked to the peripheral blood eosinophilia. However, no reason for the eosinophilia was determined in 91 patients, these patients were classified as idiopathic eosinophilia (5%). In addition, the remaining 43 patients did not complete the relevant etiological examination. Almost all underlying causes of eosinophilia were secondary (1679/1821, 92.2%). Allergic diseases were the most common secondary cause (364/1679, 21.7%), followed by solid tumors (321/1679, 19.1%), infectious diseases (294/1679, 17.5%), hematological malignancies (280/1679, 16.7%), autoimmune diseases (133/1679, 7.9%), organ rejection (including acute/chronic graft‐versus‐host disease) (132/1679, 7.9%), medications (69/1679, 4.1%), renal failure (67/1679, 4%), four cases of Kimura disease (0.2%), four cases of splenectomy (0.2%), and 11 cases of multiple factors (0.7%). Only eight cases of patients presented with MLN‐TK.

Among the 364 patients with allergic diseases, total IgE levels were measured in 207 cases, of which 174 cases (84.06%) showed varying degrees of elevation in IgE levels. Infectious diseases are divided into parasitic infections and non‐parasitic infections, only two cases of parasitic infections were found in our cohort. Other notable findings included lung cancer (34% of solid tumors) as the most common solid tumor associated with peripheral blood eosinophilia, 70% of which were adenocarcinomas of the lung. For hematological malignancies, 48.2% (135/280) of cases were caused by chronic myeloid leukemia (CML), followed by Philadelphia (Ph) chromosome negative myeloproliferative neoplasms (Ph‐MPN) (n = 63), acute myeloid leukemia (AML) (n = 38), lymphoma (n = 31), acute lymphoblastic leukemia (ALL) (n = 7), and plasma cell diseases (n = 6).

Compared with the overall 1821 cases, the proportion of male in the autoimmune disease group was lower (1253/1821, 68.8% vs. 77/133, 57.9%, p = 0.0092), and the proportion of men in the renal failure group was higher (1253/1821, 68.8% vs. 54/67, 80.6%, p = 0.043). The allergic disease group (mean age 42 vs. 56, p < 0.001) and the organ rejection group were younger (mean age 37.5 vs. 56, p < 0.001), while the solid tumor group (mean age 67 vs. 56, p < 0.001) and the infection group (mean age 66 vs. 56, p < 0.001) were older. The ACE of the solid tumor group and the infection group was lower, and the ACE of the hematological tumor group was higher (Table 2).

TABLE 1.

Clinical characteristics of secondary eosinophilia.

	Total n = 1821	Allergies n = 364	Solid tumors n = 321	Infections n = 294	Hematologic malignancy n = 280	Autoimmune n = 133	Organ rejection n = 132	Medications n = 69	Renal n = 67	Mix n = 11	Splenectomy n = 4	Kimura's disease n = 4
Male sex, n (%)	1253 (68.81%)	241 (66.21%)	223 (69.47%)	215 (73.13%)	193 (68.93%)	77* (57.89%)	85 (64.39%)	45 (65.22%)	54* (80.6%)	9 (81.82%)	3 (75%)	4 (100%)
Age (year)	56 (2–96)	42* (4–94)	67* (29–88)	66* (2–96)	53 (11–85)	49 (17–86)	37.5* (9–70)	64 (19–90)	54 (26–89)	71 (19–85)	55 (48–69)	41 (24–59)
Eosinophil count median (×109/L)	1.02 (0.53–41.35)	1.02 (0.53–15.93)	0.95* (0.53–11.25)	0.89* (0.53–13.53)	1.34* (0.53–41.35)	1 (0.56–30.61)	0.97 (0.54–7.54)	0.92 (0.58–4.17)	1.11 (0.54–8.25)	2 (0.76–7.16)	1.26 (0.73–4.53)	1.9 (1.54–6.84)

Note: Mix refers to the presence of two or more secondary factors.

*p < 0.05, with statistical difference.

Among the 43 patients without complete etiological evaluation, none showed symptoms of organ involvement, and the majority presented with mild eosinophilia; hence, eosinophil‐specific etiological testing was not conducted.

3.3. Underlying Etiology According to the Severity of Eosinophilia

In the mild group, the most common causes of eosinophilia were allergic diseases (n = 283, 22.2%). This was followed by solid tumors (n = 262, 20.6%), infections (n = 247, 19.4%), hematological malignancies (n = 152, 11.9%), autoimmune diseases (n = 92, 7.2%), organ rejection (n = 99, 7.8%), medications (n = 45, 3.5%), renal failure (n = 55, 4.3%), splenectomy (n = 3), multiple factors (n = 4), and MLN‐TK (n = 1). Among the patients in the mild group, the underlying etiology was not identified in 30 cases (2.4%).

In the moderate group, allergic diseases (n = 75, 17.9%) were not the most frequently identified causes. Hematological malignancies (n = 88, 21%) have become the most common secondary factor. Solid tumors (n = 55, 13.1%), infections (n = 44, 10.5%), autoimmune diseases (n = 38, 9.1%), organ rejection (n = 31, 7.4%), medications (n = 22, 5.3%), renal failure (n = 12, 2.9%), splenectomy (n = 1), multiple factors (n = 6), Kimura's disease (n = 3), and MLN‐TK (n = 3) were also diagnosed in this group. Unlike the mild group, there were more patients in whom the cause of eosinophilia could not be determined (n = 41, 9.8% vs. n = 30, 2.4%, p < 0.001).

In addition, the most common etiological causes of severe eosinophilia were hematological malignancies (n = 40, 46.5%). This was followed by allergic diseases (n = 6, 7%), solid tumors (n = 4, 4.7%), MLN‐TK (n = 4, 4.7%), autoimmune diseases (n = 3, 3.5%), infections (n = 3, 3.5%), organ rejection (n = 2, 2.3%), medications (n = 2, 2.3%), Kimura's disease (n = 1, 1.2%), and multiple factors (n = 1, 1.2%). Compared with other groups, the proportion of idiopathic eosinophilia in the severe group was significantly higher (n = 20, 23.3%) and the frequency of patients with hematological malignancies was the highest in the severe group (46.5% vs. 21% vs. 11.9%, p < 0.001) (Figure 1).

FIGURE 1.

FIGURE 1 The ratio of diseases in patients with different degrees of eosinophilia. As eosinophilia increases, the proportion of allergic etiology decreases gradually, while the proportion of hematological neoplasms increases progressively. EU = Idiopathic (undetermined) eosinophilia; MLN‐TK = myeloid/lymphoid neoplasms with eosinophilia and tyrosine kinase gene fusions.

3.4. Myeloid/Lymphoid Tumor With Eosinophilia and Tyrosine Kinase Gene Fusion

Of the 1821 patients with eosinophilia, only eight cases were diagnosed as MLN‐TK. All eight patients were male, with a median age of 46 years (32–69). All patients had ACE > 0.5 × 10⁹/L. Among them, seven had a value greater than 1.5 × 10⁹/L, with a median ACE of 6.18 × 10⁹/L (range: 0.54–21.7). The fusion gene contained five cases of FIP1L1::PDGFRA, one case of ETV6::PDGFRB, one case of ETV6::ABL1, and one case of FNBP1::JAK2. The clinical phenotype showed that two cases were in the blast phase (one case of mixed‐phenotype sarcoma, one case of T‐lymphoblastic lymphoma), and the remaining six cases were in the chronic phase. Patients in the chronic phase exhibited sensitivity to tyrosine kinase inhibitors (TKIs), with all patients achieving complete molecular response (CMR) at the 3‐month follow‐up assessment (Table 3).

TABLE 3.

Baseline characteristics of MLN‐TK.

No.	Gender	Age (year)	ACE (× 109/L)	Fusion gene	CP/BP	Organ involved	Treatment	3‐month evaluation
1	Male	32	0.54	FNBP1::JAK2	MPAL	Splenomegaly, mesenteric structures, lymph nodes	VA+half‐HAAG	CCyR(FNBP1::JAK2 0.11%)
2	Male	50	2.01	FIP1L1::PDGFRA	CP	Splenomegaly	Imatinib	CMR
3	Male	60	3.01	FIP1L1::PDGFRA	CP	N/A	Imatinib	CMR
4	Male	36	3.8	ETV6::ABL1	CP	Skin	Hydroxyurea	N/A
5	Male	69	8.56	FIP1L1::PDGFRA	CP	N/A	Imatinib	CMR
6	Male	42	9.34	ETV6::PDGFRB	T‐LBL	Splenomegaly, hepatomegaly, lung	Hyper‐CVAD+Imatinib	CMR
7	Male	33	19.73	FIP1L1::PDGFRA	CP	Splenomegaly	Imatinib	CMR
8	Male	61	21.7	FIP1L1::PDGFRA	CP	N/A	Imatinib	CMR

Abbreviations: ACE = absolute eosinophil count; BP = blast phase; CCyR = complete cytogenetic remission; CMR = complete molecular remission; CP = chronic phase; MPAL = mixed phenotype acute leukemia; T‐LBL = T‐cell lymphoblastic lymphoma.

3.5. Idiopathic (Undetermined) Eosinophilia

A total of 91 patients did not find the cause of eosinophilia, so they were classified as idiopathic eosinophilia. This group of patients were mainly male (61/91, 67%), with a median age of 51 years (range: 14–90 years) and a median ACE 2 × 10⁹/L (range: 0.7–9.13). Compared with secondary eosinophils, these patients had a higher proportion of hypereosinophilia (67% vs. 26%, p < 0.001).

3.6. Hypereosinophilic Syndrome

Of the 1821 patients with eosinophilia, 84 (4.6%) had end‐organ damage, and 10 had two or more organ damage. The median age of HES patients was 50 (range: 4–90) years. The male‐to‐female ratio of HES was 1.9:1. Here, 88.1% of HES patients had hypereosinophilia (> 1.5 × 10⁹/L). Around 26.2% of HES patients had severe eosinophilia (> 5 × 10⁹/L). In our cohort, 32 patients underwent biopsy of target organs, which confirmed eosinophil infiltration. This included 10 patients with an eosinophil count of less than 1.5 × 10⁹/L. The rest were diagnosed based on clinical clues. The skin (40/84, 47.6%) is the most commonly involved organ, followed by the gastrointestinal tract (18/84, 21.4%), nose (11/84, 13.13%), lung (8/84, 9.5%), lymph nodes (7/84, 8.3%), heart (2/84, 2.4%), vessels (2/84, 2.4%), nervous system (2/84, 2.4%), and other rare organs (Table 4).

TABLE 4.

Patterns of organ involvements.

	Number (%)	Diagnose (n)
Skin	40 (47.6%)	Atopic dermatitis (8) Eosinophilic dermatitis (9) AML (3) Lymphoma (3) MLN‐TK (3) Gingival carcinoma (1) Tuberculous serositis (1) Arthritis deformans (1) EGPA (3) Urinaemia (1) Bullous pemphigus (2) Idiopathic eosinophilia (5)
Gastrointestinal tract	18 (21.4%)	Eosinophilic gastroenteritis (15) MLN‐TK (1) Asthma (1) Idiopathic eosinophilia (1)
Nose	11 (13.1%)	EGPA (3) Nasosinusitis (7) Asthma (1)
Lung	8 (9.5%)	EGPA (2) Pulmonary infection(bacterial) (1) Asthma (1) Idiopathic eosinophilia (2) Eosinophilic pleuritis (1) MLN‐TK (1)
Lymph nodes	7 (8.3%)	Idiopathic eosinophilia (3) Kimura's disease (2) MLN‐TK (1) Lymphoma (1)
Splenomegaly	4 (4.8%)	MLN‐TK (2) Idiopathic eosinophilia (2)
Nervous system	2 (2.4%)	EGPA (2)
Heart	2 (2.4%)	Idiopathic eosinophilia (2)
Vessels	2 (2.4%)	Idiopathic eosinophilia (2)
Submandibular gland	1 (1.2%)	Kimura's disease (1)
Mandible	1 (1.2%)	Kimura's disease (1)
Muscle	1 (1.2%)	EGPA (1)
Hepatomegaly	1 (1.2%)	MLN‐TK (1)

Abbreviations: AML = acute myelocytic leukemia, EGPA = eosinophilic granulomatosis with polyangiitis, MLN‐TK = myeloid/lymphoid neoplasms with eosinophilia and tyrosine kinase gene fusions.

4. Discussion

The incidence of eosinophilia is not well characterized and varies across different regions or populations. In a large cross‐sectional study of routine medical samples, the incidence of eosinophilia was estimated to be approximately 4% [12], whereas it has been found in 4.8%–50% of returning travelers or arriving refugees [13, 14, 18]. In China, the proportion of eosinophilia in inpatients was 6.0% (7835/131,566), with a wide age distribution (range: 0–101 years), and it was more common in males [19]. Mild eosinophilia accounted for 92.1%, while severe eosinophilia accounted for only 1% [19]. Our cohort did not show the incidence of eosinophilia, but similar population characteristics were observed. Eosinophilia is widely distributed in age (4–96 years old), more common in males, and most of cases show mild eosinophilia.

The etiology of eosinophilia also varies according to the population and region of the study. However, the majority of patients had a specific diagnosis associated with peripheral blood eosinophilia, including inflammation, infection, allergies and so on [15, 20]. Similarly, 92.2% of patients in our cohort had possible secondary factors. Previous studies have suggested that the most common secondary factor in developed countries is allergic diseases [15, 21], while in developing countries, it is parasitic infections [22]. But currently, however, parasitic diseases seldom occur in China [23]. Our hospital is a general hospital, and many patients with various diseases visit it. But only 2 of the 1821 patients had eosinophilia related to parasitic diseases. The parasite infection rate was similar to that reported by Chen et al. in patients with moderate to severe eosinophilia, which was about 0.3% [19]. In addition, similar to prior reports in Italy, Israel, and the United States [20, 24, 25], our study also shows that allergic diseases are the most common secondary factors.

The etiology of eosinophilia is also related to the degree of eosinophilia [26]. Allergic diseases, such as asthma and allergic rhinitis, often cause mild eosinophilia, while helminth infection and neoplasms are associated with severe eosinophilia [27, 28, 29]. In children, with the aggravation of eosinophilia, the possibility of allergic causes decreases, while the proportion of eosinophilic diseases, infectious causes and malignant tumors gradually increases [25]. Hematological malignancies are the most common cause of severe eosinophilia in adults, with 51% of cases being secondary to CML [29]. Similar to the pediatric cohort, in our study, as eosinophilia increases, allergic causes become less probable, and the proportion of hematological tumors gradually rises. In the severe eosinophilia group, 40 cases (46.5%) are attributed to a hematologic malignancy, and 35 (87.5%) are secondary to CML. This is higher than the proportion reported by Lam et al., probably because our cohort includes outpatient cases, and CML patients are more likely to receive treatment in the outpatient clinic.

In addition to the most common secondary eosinophilia, clonal eosinophilia is another rare type of eosinophilia. The mean annual incidence of clonal eosinophilia was less than 0.18 cases per 1,000,000 population [30]. Some scholars believe that genetic work‐up of eosinophilia has found clonal abnormalities in a maximum of 10% of patients [31]. In previous large sample studies of all eosinophilia populations, clonal eosinophilia was rarely reported [15, 21]. About 1% of patients with eosinophilia were detected as FIP1L1::PDGFRA positive in the referral tertiary centers [32, 33]. In our cohort, only eight cases (8/1821, 0.4%) of clonal eosinophilia were identified, which were confirmed as MLN‐TK through PCR or RNA‐seq testing. As a type of clonal eosinophilia, MLN‐TK is characterized by increased eosinophils in peripheral blood or bone marrow, carrying specific tyrosine kinase fusion genes, and being effective for specific TKIs [34]. If eosinophilia persists and traditional treatments prove ineffective, further evaluation for clonal eosinophilia should be considered, and early targeted therapy is necessary to benefit the patient [35].

Sustained eosinophilia may lead to end‐organ damage, which is defined as HES. Essentially, all organs may be involved [36]. Previous studies have shown that dermatologic, pulmonary, and gastrointestinal manifestations are the most common [37]. The incidence of HES is low, approximately 0.4 cases per 1 million people [38]. However, HES accounts for about 10% of the population with severe and severe eosinophilia [22, 29]. In our study, the proportion of HES (84/1821, 4.6%) was slightly lower because we included many cases of mild eosinophilia. Dermatologic involvement (47.6%) was the most common clinical manifestation. Interestingly, 10 of the 84 patients with HES had eosinophil counts less than 1.5 × 10⁹/L, but we have pathological evidence of eosinophil infiltration. Eosinophils are mostly distributed in tissues, and a mild elevation of eosinophils in the peripheral circulation may indicate a higher burden in tissues [39]. So some patients may present with significant tissue involvement and end‐organ damage despite a lower eosinophil count [17].

There are some limitations in this study. Our data are restricted to retrospective cases derived from a single clinical center. Considering the vast territorial expanse of China, the findings of this study may only reflect the actual clinical scenario in specific regions. Conducting a nationwide multicenter prospective registry study to perform stratified analyses of etiological differences across regions with distinct geographical and economic characteristics will be the central focus of our future research efforts.

Furthermore, although most cases had a specific diagnosis that could be linked to the peripheral blood eosinophilia, but it is difficult to evaluate the causal relationship between the detected peripheral blood eosinophilia and its related diagnosis. Despite these limitations, our study shows that eosinophilia has a wide range of onset ages, is more common in males, and mostly presents with mildly elevation. Most patients with eosinophilia have possible secondary factors, and allergic diseases are the most common. Clonal eosinophilia is rare, accounting for only 0.44%, but it cannot be ignored. In addition, with the different severity of eosinophilia, the disease spectrum behind it is also changing.

Author Contributions

Concept or design: Y.M. and S.C. Acquisition of data: Y.M. and L.W. Analysis or interpretation of data: Y.M., L.W., Y.C., and X.Y. Drafting of the manuscript: Y.M. and S.C. Critical revision of the manuscript for important intellectual content: Y.C., X.Y., and S.C. All authors had full access to the data, contributed to the study, approved the final version for publication, and take responsibility for its accuracy and integrity.

Funding

This study was supported by grant from the National Key R&D Program of China (2019YFA0111000, 2022YFC2502700), the National Natural Science Foundation of China (82470166, 82400188, 82200149, 82170158, 82100175, 82100170), the priority academic program development of Jiangsu Higher Education Institution, the Natural Science Foundation of Jiangsu Province (BK20231195, BK20210087), the Open Project of Jiangsu Biobank of Clinical Resources (SBK202003001, SBK202003003). The funder had no role in study design, data collection/analysis/interpretation or manuscript preparation.

Ethics Statement

This study was performed in line with the principles of the declaration of Helsinki. Approval was granted by the ethics committee (No. 730 of 2024 LSP). The need for informed consent was waived by the Institutional Review Board of the First Affiliated Hospital of Soochow University due to the retrospective nature of the study.

Conflicts of Interest

The authors declare no conflicts of interest.

Data Availability Statement

The data supporting this study's findings are available from the corresponding author upon reasonable request.