Effect of Chinese Herbal Medicine Therapy on Risks of Overall, Diabetes-Related, and Cardiovascular Diseases-Related Mortalities in Taiwanese Patients With Hereditary Hemolytic Anemias

Hereditary Hemolytic Anemias (HHAs) are a rare but heterogeneous group of erythrocytic diseases, characterized by intrinsic cellular defects due to inherited genetic mutations. We investigated the efficacy of Chinese herbal medicine (CHM) in reducing the overall, diabetes-related, and cardiovascular diseases (CVDs)-related mortalities among patients with HHAs using a nationwide population database. In total, we identified 33,278 patients with HHAs and included 9,222 non-CHM and 9,222 CHM matched pairs after matching. The Cox proportional hazards model was used to compare the risk of mortality between non-CHM and CHM users. The Kaplan-Meier method and log-rank test were used to compare the cumulative incidence mortality between non-CHM and CHM users. The CHM prescription patterns were presented by the association rules and network analyses, respectively. The CHM prescription patterns were presented by the association rules and network analyses, respectively. CHM users showed significant reduced risks for of overall (adjusted hazard ratio [aHR]: 0.67, 95% confidence interval [CI]: 0.61–0.73, p < 0.001), diabetes-related (aHR: 0.57, 95% CI: 0.40–0.82, p < 0.001), and CVDs-related (aHR: 0.59, 95% CI: 0.49–0.72, p < 0.001) mortalities compared with non-CHM users. Two CHM clusters are frequently used to treat Taiwanese patients with HHAs. Cluster 1 is composed of six CHMs: Bei-Mu (BM; Fritillaria cirrhosa D.Don), Gan-Cao (GC; Glycyrrhiza uralensis Fisch.), Hai-Piao-Xiao (HPX; Endoconcha Sepiae), Jie-Geng (JG; Platycodon grandiflorus (Jacq.) A.DC.), Yu-Xing-Cao (YXC; Houttuynia cordata Thunb.), and Xin-Yi-Qing-Fei-Tang (XYQFT). Cluster 2 is composed of two CHMs, Dang-Gui (DG; Angelica sinensis (Oliv.) Diels) and Huang-Qi (HQi; Astragalus membranaceus (Fisch.) Bunge). Further randomized clinical trials are essential to evaluate the safety and effectiveness of above CHM products and to eliminate potential biases in the current retrospective study.

HHAs are chronic and lifelong diseases, and usually require repeated blood transfusions, which may result in iron overload and require iron chelation therapy (Sayani and Kwiatkowski, 2015;Ballas et al., 2018). In Taiwan, thalassemia is responsible for approximately 85% of the HHAs patients and requires frequent blood transfusion and iron chelation therapy. The long-term complications of frequent blood transfusion in patients include hypogonadism (23.1%), diabetes (21.2%), congestive heart failure (18.9%), arrhythmia (17.6%), osteoporosis (17.4%), and liver cirrhosis (16.5%) . Furthermore, iron overload may lead to serious complications, such as CVDs, diabetes, liver diseases, and infertility (Zaninoni et al., 2020). Among these serious complications, CVDs are major causes of morbidity and mortality in patients with iron overload (Lekawanvijit and Chattipakorn, 2009).
Chinese herbal medicine (CHM) exhibits effective, safe, less toxic, and have few side effects and has been used to treat anemia and blood-related diseases (Liu et al., 2006;Wu et al., 2006;Wang et al., 2008;Gao and Chong, 2012;Chu et al., 2014;Fleischer et al., 2016a;Fleischer et al., 2016b;Fleischer et al., 2017;Chen et al., 2018;Fleischer et al., 2018;Sun et al., 2018;Zhu et al., 2018). Furthermore, due to many complications occurred from the primary treatments for HHAs patients, HHAs patients in Taiwan may also seek for adjunctive therapies to alleviate their anemia symptoms and reduce these side effects. In Taiwan, Chinese herbal medicine (CHM) has been one of the important health care system and has also been widely used in patients with anemia, chronic myeloid leukemia, chronic lymphocytic leukemia, acute myeloid leukemia, and patients with the HSCT treatment (Fleischer et al., 2016a;Fleischer et al., 2016b;Fleischer et al., 2017;Chen et al., 2018;Fleischer et al., 2018). However, there are still limited studies in the CHM effect on patients with HHAs.
To evaluate the prolonged CHM effect on patients with HHAs, we conducted a retrospective cohort study using a nationwide population-based database in Taiwan. We explored the efficacy of CHM usage and the risks of overall, diabetes-related, and CVDsrelated mortalities among patients with HHAs.

Study Subjects
In this study, we identified 28,867,331 anonymized beneficiaries from the National Health Insurance Research Database (NHIRD) of Taiwan during Jan. 1, 2000 and Dec. 31, 2016. The International Classification of Disease, 9 th Revision, Clinical Modification (ICD-9-CM) code 282 was used to identify patients with hereditary hemolytic anemias (HHAs) during the period between Jan. 1, 2003 andDec. 31, 2013 (Figure 1).
Patients with HHAs were enrolled with the ICD-9-CM-code 282 recorded in the outpatient department (OPD) ≥ 3 times or the HHAs patients recorded in the inpatient department (IPD) at least one time. After removing the repeated patients, there were resulting in 83,840 patients with HHAs (the ICD-9-CM-code 282). For patients with HHAs, there were 3 main etiologies: hemoglobinopathy, RBC membranopathy, and RBC enzymopathy (Kim et al., 2017;Shim et al., 2020). Patients were excluded (N = 50,562) with the following criteria: 1) patients with less than 14 cumulative days of CHM prescriptions within 1 year following the diagnosis of HHAs (N = 50,153); and 2) patients who underwent splenectomy during the study period (N = 409). Patients who had more than 14 CHM cumulative days within 1 year after the diagnosis of HHAs were designated as CHM users (N = 15,195) (Figure 1, Figure 2). Patients who had no CHM prescriptions during the study period were designated as non-CHM users (N = 18,083). The non-CHM users and CHM users were matched by age, sex, iron chelation therapy, and Charlson comorbidity index (CCI) at a 1:1 ratio using the propensity score method; and they were also matched by blood transfusion therapy and thalassemia at a 1:1 ratio using the exact matching method. Finally, 9,222 non-CHM users and 9,222 CHM users were identified after matching (Table 1) ( Figure 1). The index date is the date with a completion of 14 CHM cumulative days within 1 year after the diagnosis of HHAs ( Figure 2). The CHM-users continued to use CHMs during the follow-up period (Supplementary Table S2). The basic characteristics included age, sex, CCI, blood transfusion therapy, iron chelation therapy, and patients with HHAs and thalassemia (Table 1). Comorbidities, blood transfusion therapy, and iron chelation therapy were used within 1 year before or after

Chinese Herbal Medicine
In this study, the licensed Chinese medicine doctors prescribed Chinese herbal medicine for HHAs patients in Taiwan  (Supplementary Table S1) (Supplementary Figures  S1-S8). The pharmaceutical companies with Good Manufacturing Practice (GMP) certification in Taiwan produced the CHM products, including single herbs and herbal formulas (Li et al., 2018;Cheng et al., 2019a;Tsai et al., 2019a). A single herb includes minerals, the organs of insects or animals, and any part of the plant (e.g. seeds, roots, stems, leaves, flowers, and fruits). The herbal formula is composed of more than two herbs.

Association Rule
The pattern of CHM prescriptions for patients with HHAs was shown by the association rule Tsai et al., 2019a;Cheng et al., 2019b;Tsai et al., 2019b) with SAS version 9.4 (SAS Institute, Cary, NC, United States). The association strength between CHM co-prescriptions (X and Y) were presented by the confidence value (CHM_X→ CHM_Y; %), support value (X) (%), and lift value ( Table 5).

Network Analysis
In this study, the CHM clusters were investigated by network analysis as previously described Cheng et al., 2019b;Tsai et al., 2019b;Chen CJ. et al., 2020;Tsai et al., 2020) using version 3.7.0 of Cytoscape (https://cytoscape.org/). The red circle, the green circle, and the circle size indicate the herbal formula, a single herb, and the CHM prescription frequency, respectively. The line color and the line size indicate the lift value and the support value between the CHM pairs, respectively. The darker and thicker connection line indicate a higher connection between the CHM pairs.

Statistical Analysis
Categorical data, including age, sex, blood transfusion therapy, iron chelation therapy, and thalassemia are shown as numbers (percentages). Continuous data, such as age and CCI, are shown as mean ± standard deviation (SD). The χ 2 test and the Student's t-test were used to analyze the discrepancy between non-CHM users and CHM users for categorical and continuous data, respectively ( Table 1).
The crude and adjusted hazard ratios (HRs) were calculated for the risks of overall (Table 2), diabetes-related (Table 3), and CVDs-related mortalities ( Table 4) by the Cox proportional hazard model. The overall mortality was adjusted by age, sex, CHM use, CCI, blood transfusion therapy, and iron chelation therapy. The diabetes-related and CVDs-related mortalities were adjusted by age, sex, CHM use, CCI, and blood transfusion therapy. The cumulative incidences of the overall mortality between non-CHM users and CHM users among patients with HHAs and those with and without thalassemia were calculated using the log-rank test and the Kaplan-Meier method ( Figure 3). The composition, usage patterns, and frequency of CHM for patients with HHAs are shown (Supplementary Table S1). We used version 9.4 of SAS software (SAS Institute) to analyze all statistical data and p < 0.05 was considered as significant. Table 1 presented the demographics among patients with HHAs in Taiwan. In total, 18,083 non-CHM users and 15,195 CHM users were included in this study. There were significant discrepancies among age, sex, CCI score, blood transfusion therapy, and thalassemia between the two groups at baseline (p < 0.05;   Table 1). After matching, the differences between the 9,222 non-CHM and 9,222 CHM matched pairs were not significant (p > 0.05; Table 1). The CHM-users continued to use CHMs during the follow-up period (Supplementary Table S1).
Kaplan-Meier curves showed the differences in the cumulative incidence of overall mortality between CHM users and non-users among the patients with HHAs and those with and without thalassemia (Figure 3). The cumulative incidences of overall mortality of CHM users were significantly decreased in all three groups compared with non-CHM users (p ≤ 0.001, log-rank test).

Risk of Diabetes-Related Mortality in Taiwanese Patients With HHAs
CHM users had a 43% reduced risk of diabetes-related mortality in HHAs patients (aHR: 0.57, 95% CI: 0.40-0.82, p = 0.002) compared with non-users after adjustment (Table 3). An older age also showed an elevated risk of diabetes-related mortality in the patients with HHAs (aHR: 1.06, 95% CI: 1.05-1.07, p < 0.001) ( Table 3). However, the sex-based difference in patients with HHAs was not significant. Patients with higher CCI scores presented a 1.26-fold increase in risk of diabetes-related mortality in patients with HHAs (aHR, 1.26; 95% CI: 1.21-1.32, p < 0.001) after adjustment ( Table 3). Patients who received blood transfusion therapy presented a 2.38-fold higher risk of diabetes-related mortality (aHR: 2.38, 95% CI: 1.68-3.37, p < 0.001) than those who did not in patients with HHAs after adjustment ( Table 3).
The network analysis presented the CHM prescription patterns among patients with HHAs in Taiwan (Figure 4). Two clusters with eight CHMs are significant for these patients. In cluster 1, Gan-Cao (GC; Glycyrrhiza uralensis

DISCUSSION
The effectiveness, safety, less toxicity, and few side effects of Chinese herbal medicine (CHM) in treating anemia and bloodrelated diseases are reported (Liu et al., 2006;Wu et al., 2006;Wang et al., 2008;Gao and Chong, 2012;Chu et al., 2014;Fleischer et al., 2016a;Fleischer et al., 2016b;Fleischer et al., 2017;Chen et al., 2018;Fleischer et al., 2018;Sun et al., 2018;Zhu et al., 2018). However, the long-term effect of CHM on patients with hereditary hemolytic anemias (HHAs) remain unclear. The aim of this retrospective cohort study was to investigate the effect of CHM on the risks of overall, diabetes-related, and CVDsrelated mortalities in patients with HHAs using a nationwide population database in Taiwan. In this study, at least 85% of the patients with HHAs were diagnosed with thalassemia. Patients with CHM usage had significantly reduced risks of overall, diabetes-related, and CVDs-related mortalities, respectively, when compared with non-users. We also identified that two CHM clusters with eight CHMs are important for patients with HHAs. Our findings suggested that the protective effect of CHMs against the risks of overall, diabetes-related, and CVDs-Frontiers in Pharmacology | www.frontiersin.org May 2022 | Volume 13 | Article 891729 related mortalities was observed in patients with HHAs and may implicate their clinical potential as an adjunctive therapy.
Our study results supported previous studies that showing that CHMs are effective, safe, less toxic, and have few side effects in treating anemia and blood-related diseases (Liu et al., 2006;Wu et al., 2006;Wang et al., 2008;Gao and Chong, 2012;Chu et al., 2014;Fleischer et al., 2016a;Fleischer et al., 2016b;Cheng et al., 2016;Fleischer et al., 2017;Chen et al., 2018;Fleischer et al., 2018;Sun et al., 2018;Zhu et al., 2018). Zhang et al. reported that a complex Chinese medicine, Yisui Shengxue Granule (YSSXG) may increase the relative mRNA expression level of the gamma-globin gene in the human K562 myelogenous cells in vitro . For patients with thalassemia, Wu et al., reported that the YSSXG may increase the blood levels of hemoglobin (Hb) and red blood cell (RBC; erythrocytes) in patients with thalassemia . Chu et al. reported that this YSSXG Chinese medicine may contribute to these patients with thalassemia via increasing the relative mRNA expression levels of globin genes and enhancing the antioxidant activities in erythrocytes (Chu et al., 2014). Cheng et al., reported that YSSXG Chinese medicine may also increase proliferation of hematopoietic stem cells (Cheng et al., 2016). While, Liu et al. suggested that YSSXG Chinese medicine may have a protective effect against thalassemia via inhibiting the formation of alpha-globincytotoxic precipitates in erythrocytes by upregulating GATA-1 transcription factor and alpha-hemoglobin stabilizing protein (AHSP) (Liu et al., 2006). For patients with hemoglobin H disease, Wang et al., reported that this YSSXG Chinese medicine may also increase the blood levels of Hb and RBC via enhancing the antioxidant activities in erythrocytes (Wang et al., 2008;Wang et al., 2012). Furthermore, natural compounds from Chinese herbs have also been reported for the benefit effects for the antioxidant Confidence (X →Y) (%) = Frequency of prescriptions of X and Y products/Frequency of prescriptions of X product x 100%. Lift = Confidence (X →Y) (%)/P (Y) (%). P (Y) (%) = Frequency of prescriptions of Y product/total prescriptions x 100%. Frontiers in Pharmacology | www.frontiersin.org May 2022 | Volume 13 | Article 891729 8 activity against iron overload, erythroid differentiation, hemoglobin expression, and the colony formation of hematopoietic stem cells (Thephinlap et al., 2007;Ma et al., 2013;Lin et al., 2015;Ren et al., 2016;Liu et al., 2020).
In cluster 2, patients with HHAs in Taiwan frequently used the combination of Huang-Qi (HQi; Astragalus membranaceus (Fisch.) Bunge) and Dang-Gui (DG; Angelica sinensis (Oliv.) Diels). This combination is often used to treat iron-deficiency anemia by improving hemoglobin levels and increasing iron levels through ferritin synthesis (Huang et al., 2016). The synergistic effect of Huang-Qi (HQi; Astragalus membranaceus (Fisch.) Bunge) and Dang-Gui (DG; Angelica sinensis (Oliv.) Diels) could also balance T lymphocytes, accelerate the recovery of hematopoietic stem cells, restore the balance of the T cell immune response, and inhibit the immune attack-induced apoptosis of bone marrow cells (Liu J. et al., 2019). Huang-Qi (HQi; Astragalus membranaceus (Fisch.) Bunge) alone may recover the function of megekaryocyte hematopoiesis and increase serum megakaryocyte colony-stimulating activity in anemic mice (Zhu and Zhu, 2001). Astragalus polyasccharides extracted from Huang-Qi (HQi; Astragalus membranaceus (Fisch.) Bunge) could stimulate the liver to secrete hepcidin to activate p38 mitogenactivated protein kinase and IL-6 to lower the iron load in mice (Ren et al., 2016). Angelica sinensis polysaccharide (ASP) from the Dang-Gui (DG; Angelica sinensis (Oliv.) Diels) could revive the function of hematopoietic stem cells and prevent mitochondrial apoptosis by inhibiting T cell immune abnormality (Chen Z. et al., 2020). ASP also inhibited the activation of NF-κB p65 through the IκB kinases-IκBα pathway, thereby decreasing the production of TNF-α and IL-6, which is known to suppress erythropoiesis (Wang et al., 2017).

DATA AVAILABILITY STATEMENT
The data analyzed in this study is subject to the following licenses/ restrictions: The datasets presented in this article are not readily available because Only a limited number of databases allowed access to raw data from the Taiwanese NHIRD database. Requests to access the datasets should be directed to Y-JL, yjlin.kath@gmail.com. Requests to access these datasets should be directed to Y-JL, yjlin.kath@gmail.com.

ETHICS STATEMENT
The studies involving human participants were reviewed and approved by The Institutional Review Board (IRB) issued the ethical approval to this study (CMUH107-REC3-074 (CR1)) at