Incidence and cost of haemolytic uraemic syndrome in urban China: a national population-based analysis

Haemolytic uraemic syndrome (HUS), which was first described in the 1950s, is a severe disease. All patients with HUS present with microangiopathic haemolytic anaemia, thrombocytopenia, and acute kidney injury. The presence of lesions that are consistently restricted to the kidneys is key to distinguishing HUS from thrombotic thrombocytopenic purpura (TTP), which is another form of thrombotic microangiopathy (TMA) in which brain lesions prevail and are caused by a severe deficiency (< 10%) in the expression of ADAMTS13 (A Disintegrin And Metalloproteinase with a ThromboSpondin type 1 motif, member 13) [1]. Most HUS cases are caused by Shiga toxin-producing Escherichia coli (STEC) infections. STEC accounts for 70 to 90% of all incident cases of paediatric HUS [2]. The remaining HUS cases are originally diagnosed as “atypical HUS (aHUS)”. Currently, aHUS is classified as Streptococcus pneumoniae-HUS, influenza-HUS, alternative complement pathway dysregulation-HUS, complement-independent HUS (Cobalamin C, DGKE or INF2 mutation), HUS with coexisting disease (transplantation, autoimmune disorders, drugs, malignant hypertension, malignancy/cancer chemotherapy), and idiopathic HUS (unclear aetiology) [3, 4]. HUS is the leading cause of acute kidney injury in children. Although the HUS-related mortality rate in children in industrialized countries has decreased, 3 to 5% of patients still die during the acute phase of STEC-HUS, and approximately 12% of these patients progress to end-stage renal disease (ESRD). Most patients with aHUS have a poor outcome; up to 50% ~ 60% of these patients progress to ESRD or develop irreversible brain damage, and 25% die during the acute phase of the disease [5]. In addition, patients with HUS incur high medical expenses, and the total direct cost was reported to be 17,553.39 US dollars per patient in 2005 in Argentina [6]. Since 2011, the treatment of HUS has been revolutionised by the introduction of the anti-C5 monoclonal antibody eculizumab. More than 80% of aHUS patients achieved a TMA event-free status after receiving eculizumab, which is considered a first-line treatment for complement-mediated HUS in many other countries but is still not available in China [7]. 

The annual incidence of STEC-HUS varies by year and region. Prior to 2000, the overall incidence of STEC-HUS was estimated to be 2.1 cases per 100,000 persons/yr, with a peak incidence in children who were younger than 5 yr (6.1 per 100,000/yr) and the lowest incidence in adults who were between 50 and 59 yr of age (0.5 per 100,000/yr) [8]. The overall annual rate of notified HUS in Australia between 2000 and 2010 was 0.07 cases per 100,000 per year [9]. Among patients of all ages, the annual incidence of aHUS ranges from 0.23 to 1.9 per million people [10].

Only a few single-centre case series about children with HUS have been conducted in China, and relevant data across Asia are also limited. HUS was included in the Target Rare Diseases List (TRDL) in 2017 and was one of the top 10 rare diseases with the highest rates of readmission in China [11]. Because the epidemiology of HUS in China is unknown, we conducted a retrospective analysis of the epidemiological characteristics and the associated medical costs based on medical insurance database records of patients with HUS in China.

A total of 0.37 billion residents in 16 provinces were included in this study (Supplementary table 2). In addition, 1,060 patients received a confirmed diagnosis of HUS from 2012 to 2016, with a male to female ratio of 1.36:1. Most of the patients were Han Chinese (1001, 94.43%), with a disease onset in autumn and winter (606, 57.17%). In addition, the mean age (SD) of the total patients was 49.65 (15.82) and for male and female patients with HUS were 49.98 (15.66) and 49.21 (16.03) years, respectively (Table 1). Approximately 100 patients were diagnosed with TMA during the 5-year period.

In this large population-based study, we used a nationally representative database to calculate the incidence of HUS in China for the first time. The incidence rate of HUS was 0.66 cases per 100,000 person-years, with a peak incidence in children younger than 1 year old (5.08 cases per 100,000 person-years), and the incidence in males was slightly higher than that in females. The season with the highest incidence of HUS was autumn (0.9 cases per 100,000 person-years). It is also worth noting that the average total cost per patient was 2.15 thousand US dollars, and the large population base resulted in a large total cost.

A wide range of incidences (ranging from 0.07 to 10.5 cases per 100,000 person-years) of HUS have been reported in many countries, depending on the diagnostic criteria and populations studied [9, 14]. The incidence of HUS is positively correlated with the incidence of STEC infections, as STEC infection was reported to be the most common cause of HUS [8]. For instance, Argentina has one of the highest incidence rates of STEC and HUS in the world [14], while the incidence of HUS is much lower in Australia, where STEC outbreaks appear to less common than in many other countries [9]. The incidence rate in China is within the reported range and lower than that previously reported in Western countries [8]. The possible reasons for this observation are also related to the low incidence of STEC. First, our estimation of the incidence of HUS was extrapolated from the Urban Medical Insurance databases. There are some studies indicating that urban residents have a lower prevalence of STEC-HUS than rural residents [2, 15]. Cows, which are a reservoir of STEC, are much rarer in urban areas [16]. Second, with improvements in environmental hygiene, the incidence rate of STEC infection and STEC-HUS have decreased [17, 18]. Third, some authors have also reported that a higher socioeconomic status is associated with a higher risk of developing STEC-related disease and HUS, but the mechanism remains unclear [19, 20]. In addition, the genetic background plays a role in the susceptibility and severity of HUS. Some studies have shown differences in the genetic predisposition for the development HUS between black and white populations [2, 21]. However, the genetic predisposition for developing HUS in Asian populations is unknown, as there have been few large-scale epidemiological studies. Nevertheless, the mutations in completement factor H (CFH) and factor B (CFB) observed in aHUS patients are different between Asian and European populations, 0 %~ 17% vs. 20% ~ 30% and 3.4% ~ 18% vs. 1% ~ 4%, respectively [22, 23].

Consistent with previous studies, there was no significant sex difference [24, 25]. The 2016 incidence among children was 0.29 per 100,000, which is close to that in the United States in the same year (0.51 per 100,000) [18], while the peak age of onset in China is different from that in Western countries. In our study, the incidence rate was highest in children < 1 year of age (5.08 per 100,000 person-years). However, some previous reports from Europe and North America showed that the incidence rate was highest in children < 5 years of age (1.57 to 3 per 100,000 person-years, which is similar to the age-specific incidence pattern of STEC [24, 26]. We hypothesized that the higher incidence among Chinese infants may be attributable to genetic susceptibility, as hereditary aHUS must be considered in patients with a very young age of onset [22, 27‐29]. In addition, some studies have shown that Streptococcus pneumoniae is the main cause of HUS in Chinese children (accounting for 60% ~ 100% of the cases), rather than STEC [30, 31]. This may be due to the lower pneumonia vaccination rates, and higher population density in China. However, these hypotheses require future investigation.

The peak incidence of HUS in our study occurred in autumn (0.9 cases per 100,000 person-years), followed by winter (0.7 cases per 100,000 person-years). However, the peak incidence in Western countries occurs in summer, which is also the peak period of STEC infections [2]. This significant seasonal difference also suggests that the proportion of STEC-HUS in urban China may be lower than that previously reported by studies conducted in Western countries, possibly owing to the relatively lower STEC infection rate in urban areas [15], genetic susceptibility, and regional background [21]. The difference in high-incidence seasons also needs to be further studied.

The total cost per patient per year was 2.15 thousand US dollars, which was higher than the national average annual medical care cost for urban residents (250 US dollars per patient). This may be mainly due to the high cost of some essential treatments for HUS. For example, the cost of a single plasma exchange or continuous renal replacement therapy (CRRT) session in China is approximately 1.50 thousand US dollars, and some HUS patients may need to receive multiple treatments to recover [1]. Furthermore, eculizumab, an anti-C5 antibody that is now considered a first-line treatment of HUS, is more expensive and may impose a significant economic burden on the China’s medical insurance system after its introduction in the near future. Therefore, it is critical to promote guideline-recommended therapy to improve prognosis and minimize the economic burden.

This study used a large nationally-representative sample of the Chinese mainland population, providing good estimates of the incidence and costs of HUS. However, the use of a medical insurance database still results in certain limitations. First, our data were extracted from urban populations. As STEC infections are more common in rural areas [2, 15], we may have underestimated the annual incidence of STEC-HUS in China. Second, HUS is a rare disease. The incidence rates of HUS in hospitals lacking diagnostic capabilities are probably underestimated. Some HUS patients may be diagnosed with TMA because of the inability to test for ADAMTS13. However, only approximately 100 patients were diagnosed with TMA during the 5-year period of this study. Compared with the 1060 patients who had a confirmed diagnosis of HUS, these 100 patients would not have had a substantial impact on the results. Last, the exact causes of HUS in China are still not known due to the limited availability of the relevant information. In particular, due to the unavailability of stool test results, we were unable to determine the exact proportion of patients with STEC-HUS infection in our study.