HLH is a critical disease of the blood system with rapid progress and high mortality [6]. With the deepening of understanding, the diagnosis technology of the primary cause of HLH (including the underlying disease) is daily increasingly improved, and definite progression has been made on the prognosis of patients [7]. Therefore, to present the overall epidemiological status of HLH in China, and provide Chinese data for the international HLH epidemiological investigation, this study was the first one to collect data on all cases registered in China's HLH registry network in 2019 and analyzed their incidence, diagnosis rate, and characteristics of the etiological distribution.
According to foreign reports, adult cases account for 40% of the total. The sex ratio of children with HLH is close to 1:1, but adults are more common in males [8]. In this study, the ratio of children to adult patients was about 1.9:1. The number of male and female patients was similar. The ratio of male to female (M/F) in children was about 1.05:1, while in adults it was about 1.3:1, which is consistent with previous reports. As it is known, 90% of patients with primary HLH are younger than 2 years old, and patients over 8 years old are rare. When common triggers such as infections and tumors stimulate the silent status with atypical mutations (e.g., subtype mutations) of HLH-associated genes, the patient may manifest as late-onset primary HLH [9]. In this study, there were 128 cases of primary HLH, including 55 children aged 2 years and younger, 43 children between 2 and 8 years old, 30 cases over 8 years old, and the median age was 3 years old (P25 was 2 and P75 was 8). The maximum age was 44, suggesting that with the maturity and popularization of genetic testing technology, the ability to identify late-onset primary HLH has improved.
It was reported in South Korea that the most common cause of secondary HLH was hematological malignancies, followed by EBV infection [10]. It was also reported in China that adult HLH malignancies were dominant, especially non-Hodgkin’s lymphoma [11]. These results may be related to the age distribution of the patients in both studies with the median age close to 50 years old. In our study, EBV as the primary cause accounted for nearly 45% of the total. It suggests that the most common primary cause of HLH in China is the infection of EBV, which may also be related to the prevalence of EBV in our country. The most common subtype of HLH in Japan is EBV-HLH (approximately 40% of the total), suggesting that EBV-HLH may have an ethnic genetic background [12]. However, it is worth noting that cases of unknown cause in our study accounted for 15%, exceeding the proportion of lymphoma and primary HLH. On the one hand, it may be due to the complex manifestations of HLH and the uneven diagnosis level in different regions. The diagnosis level of the primary cause of HLH needs to be improved; on the other hand, it may be related to the course of HLH because HLH progresses rapidly, making it more difficult to diagnose the cause.
It has been reported in China that the proportion of EBV infection in adult HLH patients is lower than that in children, while the incidence of fungal infection and NK/T cell lymphoma in adults is higher [12]. In this study, the overall median age of patients is 8 (P25 is 3, P75 is 30) years old, so 8 and 18 are used as age stratification. Pertaining to EBV-HLH, there was no statistically significant difference between children under 8 (including 8) and 9 to 18 years old (P > 0.05), while there was statistically significant difference between children 9 to 18 and adults over the age of 18 (P < 0.001). The proportion of EBV infection in adults was 33.4%, while in children it was 49.6%, which was consistent with the previous report. Among patients with lymphoma-associated HLH (LAHS), there was a statistically significant difference between children under 8 (including 8) and 9 to 18 years old (P < 0.01) as well as children 9 to 18 and adults over 18 (P < 0.001), suggesting that the incidence of LAHS is positively correlated with the age of onset.
Previous studies have found that the frequency of gene mutations is inversely proportional to the age of onset of HLH [3]. In this study, we found that among patients with primary HLH, there was no significant difference between children under 8 years old (including 8) and 9 to 18 years old (P > 0.05) while there was between children 9 to 18 and adults over 18 years old (P < 0.001), which is inconsistent with previous studies. However, it still reflects that genetic mutations are more common in children.
The epidemiological survey of lymphoma in Japan shows that the overall sex ratio of lymphoma (M/F) is 1.17. Some subtypes of lymphoma patients are mainly males (M/F > 3:1), but there is no statistical difference in the sex of patients with the main subtypes of lymphoma such as diffuse large B cells lymphoma (DLBCL) [13]. In this study, LAHS was significantly different between males and females (P < 0.05). The sex ratio (M/F) of total case number was 1.1, while for the LAHS, the ratio was 1.5, suggesting that patients of LAHS are mainly males; Rheumatic immune diseases are complex, such as systemic lupus erythematosus (SLE), osteoarthropathy (OA), and Sjogren’s syndrome (SS) are more common in females, while ankylosing spondylitis and gout are more common in males [14]. Among them, the most common diseases related to HLH are systemic juvenile idiopathic arthritis (sJIA) and adult still's disease (AOSD). It has been reported that sJIA complicated by macrophage activation syndrome (MAS) is more common in women [15]. Similarly, AOSD is more common in women than men. Reports of male patients with AOSD-HLH are rare [16]. In this study, there were 78 cases of rheumatic immune-associated HLH, including 20 males and 58 females. The difference of sex was statistically significant (P < 0.001), suggesting patients of rheumatic immune-associated HLH are mainly women, which is consistent with previous reports.
In addition to EBV, other infection factors include cytomegalovirus (CMV), human herpesvirus type 6 (HHV-6), influenza virus, Mycobacterium tuberculosis, parasites, fungi, and common infectious diseases consist of HIV, leishmaniasis, brucellosis, tuberculosis, etc. [17]. Leishmaniasis [18] is mainly distributed in Gansu, Shanxi, Shaanxi, Sichuan and Xinjiang, etc. In this study, Gansu has the highest incidence and is one of the main endemic areas of leishmaniasis. Therefore, there are more cases of HLH related to Leishmania infection than other infectious diseases. Brucellosis is mainly distributed in pastoral areas [19] such as Qinghai, and Lyme disease is mainly distributed in forest areas such as Northeast, Northwest and North of China [20]. Other infectious diseases have no obvious regionality. Therefore, when considering the primary cause of HLH, the local epidemiological situation should be considered, and the contact history of the affected area should be emphatically asked to avoid missing rare infections other than EBV. About other infectious diseases with no obvious geographical bias, it is important to note that malignant tumors and opportunistic infections are important triggers of HLH in HIV patients, and acute HIV infection itself can cause HLH. To make matters worse, the treatment of HIV-associated HLH is still challenging and the use of steroid therapy can not improve the prognosis of patients [21], which reminds us that we should enhance the prevention and education of HIV.
According to reports, the incidence of HLH in children has increased to (1 to 225)/300,000, and it is related to geographical factors [8]. In this study, the overall incidence of HLH in China in 2019 was about 1.04/1000,000 (excluding Taiwan Province, Hong Kong and Macau Special Administrative Region do not provide data). The incidence is highest in Gansu with 4.684/1000,000 followed by Shaanxi, Hubei, Jiangxi, etc. Compared with other areas, the incidence is lower in the Yangtze River Delta and lowest in Shanghai with 0.083/1000,000. The overall incidence has shown a downward trend from inland to coastal and border areas. It is worth noting that the incidence in Beijing and Tianjin is relatively high. Since the epidemiological investigation of HLH mainly relies on case reports, the incidence obtained by statistics depends on the diagnostic level of the local area. Therefore, this study further explored the diagnosis rate in various regions and its correlation with the local economic level. It found that the diagnosis rate had a significant correlation with the local GDP per capita (P < 0.05).
HLH is a syndrome of pathological immune activation. Common symptoms are persistent fever, splenomegaly, and pancytopenia, but these symptoms are not specific [1], which increases the difficulty in distinguishing other symptoms-overlapping inflammatory diseases with HLH. At present, HLH is mainly diagnosed based on HLH-2004 [23]. At present, it is known that primary HLH is a gene defect that causes the cytotoxicity of NK cells and cytotoxic T lymphocytes (CTL) to be weakened or even absent (mainly NK cells), which leads to the accumulation of antigen-presenting cells (APC). Then CD8 + CTL is continuously stimulated to release a large number of cytokines to trigger a "cytokine storm" [24]. NK cytotoxicity is determined by cytotoxicity test, and sCD25 concentration is related to T cell activation [22]. The above two detection methods are more sensitive than other indicators of HLH-2004 [25] but have higher requirements for laboratories. For example, the NK cytotoxicity test requires the use of radioactive 51Cr and the results are affected by the number of NK cells [26], so it has not been universal yet [3]. In this study, data from Taiwan, Hong Kong and Macao was not available. The incidence of the other 31 provinces, municipalities and autonomous regions has no statistical correlation with GDP indicators (P > 0.05). Qinghai and Tibet has a small number of cases and they all come from the local area, considering the geographical occlusion affects the medical habits of local people. In order to avoid making additional effect on the statistical results of the diagnosis rate, they were not included in the analysis. Excluding the above five regions, the diagnosis rate has no significant correlation with GDP and GDP increment but has a significant correlation with GDP per capita (P < 0.05), and the Pearson correlation coefficient is positive (value is 0.403), indicating that the local diagnosis level is positively correlated with the local economic level. Relatively economically developed areas have better medical resources, higher levels of diagnosis and treatment, and the ability to develop new diagnostic technologies. However, the medical level of economically underdeveloped areas is relatively backward. This study shows that areas with high incidence are concentrated in the northwest inland of China, which is also an economically underdeveloped area. Improving the diagnosis level in this area is of great significance to improve the prognosis of patients and enhance health of national people. The supply of medical resources to the underdeveloped region should be increased, medical talents and technology should be introduced, and the construction of laboratories as well as the use of Internet medical platforms should be strengthened. At the same time, research on new and easy-to-obtain detection methods should be carried out to make early diagnosis and early treatment of suspected cases in areas with high HLH incidence in China.