Hereditary spherocytosis (HS) is the most common form of inherited hemolytic anemia caused by abnormalities in red blood cell membrane proteins. The known causative genes of HS include ANK1 (OMIM: *612641), SPTB (OMIM: *182870), SLC4A1 (OMIM: *109270), SPTA1 (OMIM: *182860), and EPB42 (OMIM: *177070) and the majority of them exhibit autosomal dominant inheritance, while a minority demonstrate autosomal recessive inheritance. Clinical manifestations of HS show heterogeneity. Typical HS patients usually present with anemia and jaundice in childhood, and as they grow older, they develop splenomegaly, gallstones, and other complications. Diagnosis was previously made through peripheral blood smear examination and red blood cell fragility tests, but now genetic sequencing can provide a definitive diagnosis by identifying the underlying genetic mechanisms. Although all adult patients in our cohort exhibited typical symptoms such as anemia, jaundice, splenomegaly, and gallstones, two adult patients were initially misdiagnosed with autoimmune hemolytic anemia, which tested negative for direct antiglobulin test (DAT). These patients received treatment with intravenous immunoglobulin, corticosteroids, and cyclosporine, but due to poor response, the diagnosis was reassessed. Reasons for misdiagnosis of HS include the absence of a family history, the presence of spherocytes in autoimmune hemolytic anemia, a relatively low percentage of spherocytes in the peripheral blood, and the misconception that jaundice, splenomegaly, and gallstones are not specific to HS. Limited awareness of the disease among physicians also contributes to misdiagnosis. We recommend that genetic testing for children and adults with unexplained hemolytic anemia to establish a definitive diagnosis at the genetic level.

The incidence and spectrum of pathogenic gene mutations vary among different races and regions. Wang et al. [9]summarized the data of Chinese HS patients from 2000 to 2020, and the proportions of ANK1, SPTB, SLC4A1, and SPTA1 mutations were 46%, 42%, 11%, and 1%, respectively, which is similar to the research data from Korea [10]. Our study confirmed previous findings: ANK1 mutations accounted for 52.94%, SPTB mutations accounted for 44.12%, SLC4A1 mutations accounted for 2.94%, and no SPTA1 or EPB42 mutations were found. In addition, ANK1 variants account for 40–65% in the United States and Europe, 5–10% in Japan [11], and 27% in the Netherlands [12], suggesting that the pathogenicity rate of ANK1-HS in China is closer to that of Europe and the United States, while other East Asian regions, such as Japan, may be limited by sample size in research.

Through pedigree analysis, it was found that 32.0% (8/25) of mutations were inherited within families, while 68.0% (17/25) were de novo mutations. Additionally, In our current study, 9 mutations were previously unreported. Miraglia et al. [13] analyzed core pedigrees of 19 HS patients, of which 12 cases (63%) were de novo mutations. Peng et al. [14] conducted pedigree analysis of 16 HS cases, with 6 cases (37.5%) being inherited mutations and 10 (62.5%) being de novo. This indicates the importance of proactive and early molecular diagnosis for HS, as it has significant implications for genetic counseling. However, in early diagnosis, physicians may miss the opportunity for a definitive diagnosis due to a lack of genetic evidence and milder phenotypes. Furthermore, the incomplete penetrance observed in 3 families (Family 14, 29, and 21) in our study highlights the importance of proactive molecular diagnosis. Van Vuren et al. [12] suggested that phenotypic heterogeneity within affected individuals in a family may be influenced by factors such as functional compensation by normal alleles in the heterozygous state, the impact of modifier genes (such as a-LEPRA and a-LELY) [12, 15,16,17], and the regulatory role of epigenetic modifications on HS gene expression. Additionally, 20–30% of HS patients exhibit compensatory hemolysis that is characterized by elevated indirect bilirubin and reticulocytes but normal hemoglobin levels. These patients may develop anemia with age and require close monitoring of disease progression [12].

There is limited research on the diagnosis of HS patients in children and adults. In this study, a comparison between pediatric and adult diagnosis revealed that anemia was more severe in pediatric patients compared to adult patients. Among pediatric patients, RBC, HB, MCV, and MCH levels were significantly lower than in adult patients. In adult ANK1-HS patients, MCH levels were significantly higher than in pediatric patients, but there were no significant differences in RBC, HB, MCV, and MCHC levels. In SPTB-HS patients, adults had significantly higher levels of RBC, HB, and MCH compared to children (P < 0.05), while MCV and MCHC levels showed no significant differences. Although infants have slightly lower hemoglobin (HB) levels than adults, their bone marrow has a relatively enhanced compensatory capacity. Among this patient group, 17 out of 22 (77.27%) pediatric cases presented with moderate anemia, while 5 out of 22 (22.73%) exhibited severe anemia, and there were no instances of mild anemia. Among the adult patients, there were 5 cases of mild anemia (41.67%, 5/12), 6 cases of moderate anemia (50.0%, 6/12), and 1 case of severe anemia (8.33%, 1/12). It is clear that anemia presents with greater severity in paediatric patients. However, it should be noted that the sample size in this study is relatively limited, necessitating further confirmation through subsequent investigations. Additionally, there were no significant differences in HB and RBC levels between adult ANK1-HS and SPTB-HS patients. We found that the severity of HS in adult patients does not differ significantly between different gene mutations, which is consistent with some previous research findings. However, in patients under 14 years of age, ANK1-HS had more severe anemia compared to SPTB-HS, which may be related to the relatively small number of children in previous studies. Our study suggests significant phenotypic differences between pediatric and adult diagnosis of HS, which need to be further confirmed with larger sample sizes.

Among the 34 cases in this study, nonsense mutations accounted for 50%, frameshift mutations accounted for 26.47%, and splice site mutations accounted for 23.53%. This indicates that HS is mainly caused by nonsense, frameshift, and splice site mutations, which is consistent with most reports [10, 18,19,20], in which only a few studies have reported missense mutations in HS genes [19, 21]. The mutations in our cohort were scattered throughout the exons and introns of the ANK1 and SPTB genes, and no hotspot mutations or regions were identified, which is consistent with previous reports [22].

The relationship between gene variants and phenotype in HS patients is controversial. Aggarwal et al. [20] and Tole et al. [23] have found that the impact of ANK1 mutations on red blood cell indices is similar to that of SPTB mutations, and mutation type or location cannot predict the severity of the disease. On the other hand, Park et al. [10] found that SPTA1 mutations are associated with the most severe disease, while SLC4A1 mutations result in the mildest phenotype. Qin et al. [19] observed that MCV and MCH levels were significantly higher in ANK1-HS compared to SPTB-HS, and the percentage of spherocytes in the peripheral blood was significantly lower in ANK1 mutation patients. Additionally, the MCHC levels in the nonsense, frameshift, and splice site mutation groups were significantly higher than in the missense mutation group, and the severity of the disease did not differ significantly between different gene mutations. In our study, an analysis of genotype and phenotype revealed that RBC and HB levels in ANK1-HS were significantly lower than in SPTB-HS, but there were no significant differences in MCV, MCH, and MCHC levels, which differs from previous reports [18, 19]. Additionally, when classified by mutation type, there were no significant statistical differences in HB and RBC counts between the nonsense, frameshift, and splice site mutation groups (P > 0.05), which is consistent with the findings of Wang et al. [9].

The retrospective analysis spanned five years and encompassed both children and adults diagnosed in Jiangxi Province, China. It is worth noting that the limited sample size may not be fully representative of comprehensive data on HS patients in China. Some of the conclusions require further verification by increasing the sample size and extending the follow-up period. Disease progression indicators, such as blood count, splenomegaly, and gallstones, are also being actively monitored. Once enough data has been collected, we will conduct further analysis. This report only focuses on identifying new variants in known HS genes due to technical limitations and does not include other possible mutations. It may not cover all genetic factors related to HS, including potential new genes. In the next step, we will analyze other variables such as family background, age group, different environments, and lifestyle factors.