Concurrent Infections of Giardia duodenalis, Enterocytozoon bieneusi, and Clostridium difficile in Children during a Cryptosporidiosis Outbreak in a Pediatric Hospital in China: e2437

Abstract

Background

Over 200 cryptosporidiosis outbreaks have been reported, but little is known if other enteric pathogens were also involved in some of these outbreaks. Recently, an outbreak of cryptosporidiosis linked to poor hygiene by two Cryptosporidium hominis subtypes occurred in a pediatric hospital ward (Ward A) in China, lasting for more than 14 months. In this study, the concurrence during the outbreak of three other enteric pathogens with a similar transmission route, Giardia duodenalis, Enterocytozoon bieneusi, and Clostridium difficile, was assessed.

Methods/Principal Findings

The occurrence of G. duodenalis, E. bieneusi, and C. difficile in 78 inpatients from Ward A and 283 and 216 inpatients from two control wards (Wards C and D) in the same hospital was examined using molecular diagnostic tools. Significantly higher infection rates were found in children in Ward A for all study pathogens than in Wards C and D (P<0.01): 9.5% versus 1.4% and 0% for G. duodenalis, 10.8% versus 2.8% and 3.7% for E. bieneusi, and 60.8% versus 37.8% and 27.8% for C. difficile, respectively. These differences were mostly seen in children ≤12 months. Enteric pathogen-positive children in Ward A (31/58 or 53.4%) were more likely to have mixed infections than those in Ward C (4/119 or 3.4%) or D (5/68, 7.4%; P<0.01). Having cryptosporidiosis was a risk factor for G. duodenalis (OR = 4.3; P = 0.08), E. bieneusi (OR = 3.1; P = 0.04), and C. difficile (OR = 4.7; P<0.01) infection. In addition, a lower diversity of G. duodenalis, E. bieneusi, and C. difficile genotypes/subtypes was observed in Ward A.

Conclusions/Significance

Data from this study suggest that multiple pathogens were concurrently present during the previous cryptosporidiosis outbreak. Examination of multiple enteric pathogens should be conducted when poor hygiene is the likely cause of outbreaks of diarrhea.

Abstract

Background

Over 200 cryptosporidiosis outbreaks have been reported, but little is known if other enteric pathogens were also involved in some of these outbreaks. Recently, an outbreak of cryptosporidiosis linked to poor hygiene by two Cryptosporidium hominis subtypes occurred in a pediatric hospital ward (Ward A) in China, lasting for more than 14 months. In this study, the concurrence during the outbreak of three other enteric pathogens with a similar transmission route, Giardia duodenalis, Enterocytozoon bieneusi, and Clostridium difficile, was assessed.

Methods/Principal Findings

The occurrence of G. duodenalis, E. bieneusi, and C. difficile in 78 inpatients from Ward A and 283 and 216 inpatients from two control wards (Wards C and D) in the same hospital was examined using molecular diagnostic tools. Significantly higher infection rates were found in children in Ward A for all study pathogens than in Wards C and D (P<0.01): 9.5% versus 1.4% and 0% for G. duodenalis, 10.8% versus 2.8% and 3.7% for E. bieneusi, and 60.8% versus 37.8% and 27.8% for C. difficile, respectively. These differences were mostly seen in children ≤12 months. Enteric pathogen-positive children in Ward A (31/58 or 53.4%) were more likely to have mixed infections than those in Ward C (4/119 or 3.4%) or D (5/68, 7.4%; P<0.01). Having cryptosporidiosis was a risk factor for G. duodenalis (OR = 4.3; P = 0.08), E. bieneusi (OR = 3.1; P = 0.04), and C. difficile (OR = 4.7; P<0.01) infection. In addition, a lower diversity of G. duodenalis, E. bieneusi, and C. difficile genotypes/subtypes was observed in Ward A.

Conclusions/Significance

Data from this study suggest that multiple pathogens were concurrently present during the previous cryptosporidiosis outbreak. Examination of multiple enteric pathogens should be conducted when poor hygiene is the likely cause of outbreaks of diarrhea.

Author Summary

The transmission of Giardia duodenalis, Enterocytozoon bieneusi, and Clostridium difficile is poorly understood in developing countries despite their wide occurrence. Because they are transmitted by the same fecal-oral route as Cryptosporidium, in this study, we have examined the occurrence of these enteric pathogens in children during a cryptosporidiosis outbreak in a pediatric hospital in China. Using molecular diagnostic tools, we have detected significantly higher infection rates of these enteric pathogens in the outbreak ward than in two control wards in the same hospital. We have also shown a much higher occurrence of these pathogens in children having cryptosporidiosis than those having no cryptosporidiosis. We have demonstrated that the genetic diversity of enteric pathogens is much lower in the outbreak ward than in control wards. Therefore, other enteric pathogens are concurrently present during the cryptosporidiosis outbreak, and examinations for multiple enteric pathogens should be conducted when poor hygiene is considered the likely cause of outbreaks of diarrhea.

Citation: Wang L, Xiao L, Duan L, Ye J, Guo Y, et al. (2013) Concurrent Infections of Giardia duodenalis, Enterocytozoon bieneusi, and Clostridium difficile in Children during a Cryptosporidiosis Outbreak in a Pediatric Hospital in China. PLoS Negl Trop Dis 7(9): e2437. doi:10.1371/journal.pntd.0002437

Editor: Gagandeep Kang, Christian Medical College, India

Received: March 14, 2013; Accepted: August 9, 2013; Published: September 12, 2013

This is an open-access article, free of all copyright, and may be freely reproduced, distributed, transmitted, modified, built upon, or otherwise used by anyone for any lawful purpose. The work is made available under the Creative Commons CC0 public domain dedication.

Funding: This work was supported by the National Natural Science Foundation of China (No. 31110103901, 31229005, 41001316); the National Basic Research Program of China (973 Project No. 2011CB200903); and Fundamental Research Funds for the Central Universities, China. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

Competing interests: The authors have declared that no competing interests exist.

Introduction

Cryptosporidium is a significant cause of diarrhea in humans worldwide [1]. Humans can acquire Cryptosporidium infections through the fecal-oral route via direct person-to-person or animal-to-person contact, or ingestion of contaminated water or food [2]. Thus far, over 200 waterborne, foodborne, person-to-person, and zoonotic cryptosporidiosis outbreaks have been reported [3], [4]. However, whether other co-pathogens were involved in some of these outbreaks remains largely unexamined.

Similar to Cryptosporidium, pathogens like Giardia duodenalis, Enterocytozoon bieneusi, and Clostridium difficile are also significant causes of diarrhea in humans worldwide and can be transmitted from persons to persons by the same fecal-oral route involved in cryptosporidiosis occurrence [1], [5], [6]. All of these pathogens are major causes of healthcare-associated infections, especially Clostridium difficile [7]-[9]. Despite their wide occurrence, the epidemiology of these enteric pathogens is largely unclear in developing countries. Only limited data exist on the molecular epidemiology of these pathogens in China [10]-[16].

In one recent molecular epidemiologic study on Cryptosporidium in in-patients from three pediatric hospitals, P. R. China, we identified an extended outbreak of cryptosporidiosis in a pediatric hospital ward (Ward A, Hospital I), with more than 50% (38/74) children affected by two C. hominis subtypes (IaA14R4 and IdA19) during a 14-month period (Sep. 2007-Oct. 2009) [17]. The infection rate in Ward A was significantly higher than the overall rates in Hospitals I (2.8%), II (0.6%) and III (0.4%). The diversity of Cryptosporidium species and C. hominis subtypes were significantly lower in Ward A than in other wards/hospitals, with only one species (C. hominis) and two C. hominis subtypes (IaA14R4 and IdA19) being found in 38 patients in Ward A while four species of Cryptosporidium and six C. hominis subtypes being found in 62 patients in other wards and hospitals [17].

Because concurrent infections of multiple pathogens are sometimes involved in gastroenteritis in hospitalized children [18], [19], in the present study, we retrospectively compared the infection rates and subtype distribution of G. duodenalis, E. bieneusi, and C. difficile in hospitalized children in Ward A with those in two control wards in the same hospital: Ward C for patients having hemophilia, anemia, and neurological diseases, and Ward D for patients having general surgeries. This was the first study to use genotyping and subtyping tools to investigate the transmission of multiple enteric pathogens during a cryptosporidiosis outbreak.

Methods

Ethics statement

Written informed consent was obtained from the parents or guardians of the children. This study was approved by the Ethics Committee of the East China University of Science and Technology.

Clinical specimens and study design

All specimens for this study were collected from in-hospital children during September 2007-October 2009 as described [17]. These children were hospitalized mostly due to non-gastrointestinal illness: Ward A for patients with various congenital or inherited diseases from a local welfare institute; Ward C for children attending the Department of Endocrinology, Hematology and Neurology; and Ward D for children attending the Department of General Surgery.

In this study, Ward A (Cryptosporidium infection rate = 51.4%), where the cryptosporidiosis outbreak occurred, was regarded as the case ward, while two other wards (Wards C and D; Cryptosporidium infection rates = 1.8% and 2.3%, respectively) in the same hospital (Hospital I in Shanghai, China) without cryptosporidiosis outbreak were regarded as the control wards. Overall, 573 children, including 74 from Ward A (age range: 1-192 months; mean age: 20.7 months), 283 from Ward C (age range: 1-168 month; mean age: 41.3 months), and 216 from Ward D (age range: 1-216 months; mean age: 43.8 months), were examined for the occurrence and genotype/subtype distribution of G. duodenalis, E. bieneusi, and C. difficile. In addition, 2,672 children from other known or unknown wards in Hospital I (age range: 0-228 months; mean age: 46.9 months), 489 children from Hospital II (age range: 0-192 months; mean age: 37.2 months age), and 311 children from Hospital III (age range: 1-159 months; mean age: 40.4 months) in the same city, were also examined for G. duodenalis. Information on age, gender, and the occurrence of diarrhea as defined by the attending physicians was collected for each patient as previously described [17].

Molecular diagnosis of enteric pathogens

Genomic DNA was extracted from 0.2 ml of fecal materials using a FastDNA SPIN Kit for Soil (BIO 101, Carlsbad, CA). To detect G. duodenalis, a 532-bp fragment of the triosephosphate isomerase (tpi) gene was amplified by nested PCR [20]. A 511-bp fragment of the β-Giardia (bg) and a 530-bp fragment of the glutamate dehydrogenase (gdh) gene were further amplified from DNA of the tpi-positive specimens [21], [22]. Giardia duodenalis genotypes and subtypes were determined using the established nomenclature system based on multilocus sequence data [22].

A ~392-bp fragment of the rRNA gene containing the entire internal transcribed spacer (ITS) was amplified and sequenced to detect and identify E. bieneusi genotypes [23]. Genotypes of E. bieneusi were named according to established nomenclature [23], [24]. A PCR based on the tcdB gene was used to detect C. difficile [25]. Clostridium difficile in tcdB-positive specimens was subtyped by sequence analysis of the slpA gene as previously described [9].

Sequence analysis

All positive PCR products generated in the study were directly sequenced using Big Dye Terminator v3.1 Cycle Sequencing Kits (Applied Biosystems, Foster City, CA) and an ABI 3130 Genetic Analyzer (Applied Biosystems). Sequences were assembled using ChromasPro (version 1.5) software (http://technelysium.com.au/?page_id=27). The accuracy of the sequencing reads was confirmed by bidirectional sequencing. The nucleotide sequences of G. duodenalis, E. bieneusi, and C. difficile genotypes/subtypes obtained were aligned with reference sequences of each genetic locus

[Figure omitted, see PDF]

Figure 1. Infection rates of study pathogens in case and control wards.

There were 74 children in case ward (Ward A), 283 and 216 children in control wards (Wards C and D).

Table 1. Distribution of Giardia duodenalis, Enterocytozoon bieneusi, and Clostridium difficile infections in pediatric inpatients by ward and Cryptosporidium infection status.

Among the 573 specimens examined, 24 (4.2%) were positive for E. bieneusi at the ITS locus, with eight positives in each ward. The infection rate of E. bieneusi in Ward A (10.8%) was significantly higher than those in Ward C (2.8%) and D (3.7%) (P = 0.01; Figure 1; Table 1). Altogether, 212 of the 573 specimens were positive for C. difficile at the tcdB locus, with 45/74 (60.8%), 107/283 (37.8%), and 60/216 (27.8%) in Wards A, C, and D being positive, respectively (Figure 1). The infection rate of C. difficile was significantly higher in Ward A than in Wards C and D (P<0.01; Table 1).

Concurrent infections with multiple pathogens in case and control wards

Concurrent infections of multiple pathogens, including Cryptosporidium, G. duodenalis, E. bieneusi, and C. difficile, were detected in both the case and control wards. Comparing with the control Wards C and D, Ward A had a significantly higher overall infection rate of enteric pathogens (58/74 or 78.4% versus 119/283 or 42.0% and 68/216 or 31.5%; P<0.01). Over half of children with enteric pathogens in Ward A (31/58 or 53.4%) were concurrently infected with two or more pathogens, while only a small number of children with enteric pathogens in Ward C (4/119 or 3.4%) or D (5/68, 7.4%) were infected with multiple pathogens (P<0.01).

In this study, children who had cryptosporidiosis during the outbreak were more likely to be infected with other enteric pathogens (Table 1). Among the 573 children examined for all four pathogens, 48 were previously diagnosed as having cryptosporidiosis. These Cryptosporidium-positive children had higher infection rates of G. duodenalis (6.3% versus 1.5%; P = 0.08), E. bieneusi (10.4% versus 3.6%; P = 0.04), and C. difficile (70.8% versus 33.9%; P<0.01) than Cryptosporidium-negative children (Table 1).

Occurrence of enteric pathogens by age and gender

The age distribution of G. duodenalis, E. bieneusi, and C. difficile infections in 573 children from Wards A, C, and D is shown in Table 2. Infection rates of G. duodenalis were similar among all age groups (P>0.05). In contrast, children ≤6 months were more likely infected with E. bieneusi (11/99 or 11.1% versus 12/473 or 2.5% for other age groups, P<0.01), and children ≤12 months were more likely infected with C. difficile (124/277 or 44.8% versus 88/295 or 29.8% for other age groups, P<0.01; Table 2). Among children under 12 months, infection rates of all three study pathogens were significantly higher in Ward A than in control wards (5/59 or 8.5% versus 1/218 or 0.5%, P<0.01 for G. duodenalis; 7/59 or 11.9% versus 8/218 or 3.7%, P = 0.03 for E. bieneusi; 38/59 or 64.4% versus 86/218 or 39.4%, P<0.01 for C. difficile). In contrast, in children older than 12 months, only G. duodenalis was significantly more prevalent in Ward A than in the controls (2/15 or 13.3% versus 3/280 or 1.1%, P = 0.01 for G. duodenalis; 1/15 or 6.7% versus 7/280 or 2.5%, P = 0.88 for E. bieneusi; 7/15 or 46.7% versus 81/280 or 28.9%, P = 0.14 for C. difficile; Table 2). No gender difference was seen in the occurrence of G. duodenalis, E. bieneusi, and C. difficile infections in this study (P>0.05; Table 2).

Table 2. Distribution of study pathogen infections in case (A) and control (C and D) wards by age, gender, and diarrhea status.

Distributions of G. duodenalis, E. bieneusi, and C. difficile genotypes/subtypes

The distribution of G. duodenalis multilocus subtypes was different between case and control wards. In Ward A, six of the seven specimens positive for G. duodenalis at the tpi locus were also positive at the bg and gdh loci, and all of them belonged to the multilocus subtype AII (Table 3; Figure 2). In contrast, both multilocus subtype AII and subtypes belonging to the assemblage B (2 cases each) were found in Ward C. Similarly, both AII and B were detected in other known or unknown wards in Hospital I, and in Hospitals II and III (Table 3).

[Figure omitted, see PDF]

Figure 2. Phylogenetic relationship of multilocus sequence subtypes of Giardia duodenalis.

Sequences from this and a previous study [22] are included in the analysis. The relationship of multilocus sequence subtypes was inferred by a neighbor-joining analysis of concatenated sequences of the gdh, tpi, and bg fragments, based on the p-distance model. Bootstrap values >50% are shown. Novel and known genotypes identified in this study are indicated by black and white triangles, respectively.

Table 3. Distribution of Giardia duodenalis, Enterocytozoon bieneusi, and Clostridium difficile genotypes or subtypes.

Four known genotypes of E. bieneusi were found in this study, with Peru 11 as the dominant one (in 6 cases). The other three includes EbpC (1 case), EbpA (2 cases), and D (1 case). Twelve novel genotypes (SH1-12) were found in this study, with SH2 in three cases and all other genotypes in one case each (Table 3). All 16 E. bieneusi genotypes except SH5 belonged to Group 1 phylogenetically, while genotype SH5 belonged to Group 2 (Figure 3). Higher occurrence of E. bieneusi genotype Peru 11 was seen in Ward A (4/8 genotyped) than in Wards C and D (1/8 genotyped each; Table 3).

[Figure omitted, see PDF]

Figure 3. Phylogenetic relationship of Enterocytozoon bieneusi genotypes.

The relationship of genotypes identified in this study and some other genotypes in GenBank was inferred by a neighbor-joining analysis of ITS sequences, based on the p-distance model. Bootstrap values >50% are shown. Novel and known genotypes identified in this study are indicated by black and white triangles, respectively.

Among the 212 C. difficile-positive specimens based on PCR analysis of the tcdB gene, 160 specimens were subtyped at the slpA locus successfully. In total, 20 slpA subtypes were obtained, including 8 novel ones (Table 3). Most of the novel subtypes were genetically close to subtypes previously reported, although two of them, sh-01 and sh-02, had very different sequences and formed an independent clade in the phylogenetic tree (Figure 4). The most common subtype in Ward A was fr-01 (15/40 slpA-positive cases), compared to kr-03 in control Wards C (23/74 slp-A positive cases) and D (18/46 slp-A positive cases; Table 3).

[Figure omitted, see PDF]

Figure 4. Phylogenetic relationship of Clostridium difficile subtypes.

The relationship of subtypes identified in this study and other subtypes in a previous study [9] was inferred by a neighbor-joining analysis of slpA sequences, based on the p-distance model. Bootstrap values >50% are shown. Novel and known subtypes identified in this study are indicated by black and white triangles, respectively.

Pathogen occurrence and diarrhea

A significantly higher diarrhea rate was observed in Ward A than in control Wards C and D (43/74 or 58.1% versus 180/499 or 36.1%, P<0.01). Infection with Cryptosporidium was significantly associated with the occurrence of diarrhea (OR = 1.95, p = 0.002). However, a large number of asymptomatic G. duodenalis, E. bieneusi, and C. difficile infections were observed in both case and control wards in this study(Table 2). None of the three pathogens were significantly associated with the occurrence of diarrhea in the pediatric inpatients (P>0.05; Table 2). None of the dominant genotypes/subtypes of the three study pathogens were significantly associated with the occurrence of diarrhea (P>0.05; data not shown). In addition, in Ward A, the difference in diarrhea rates between children with multiple infections and children with single infection was not significant (17/31 or 54.8% versus 17/27 or 63.0%, P = 0.51). This was also the case in the control Wards C and D (3/9 or 33.3% versus 56/178 or 31.5%, P = 1.0).

Discussion

Molecular epidemiological investigations have improved our understanding of the transmission of enteric pathogens, including those examined in the present study [2], [7]-[9], [14]. They are especially useful in identifying the occurrence of outbreaks, linking seemingly un-associated cases, and tracking infection sources. In the present study, using genotype and subtype tools, we retrospectively identified concurrent transmission of G. duodenalis, E. bieneusi, and C. difficile during a cryptosporidiosis outbreak previously identified in Ward A of Hospital I in Shanghai, China. This is reflected by higher infection rates and lower genetic diversity of these enteric pathogens in Ward A than in control wards.

The low infection rates of G. duodenalis in wards other than Ward A in Hospital I (0-1.4%), Hospitals II (0.6%), and III (1.6%) are similar to those reported in out-patients and inpatients (0.2-0.6%) and the general population (2.5%) in China [11], [26]-[29]. Likewise, infection rates of E. bieneusi in hospitalized children in Wards C (2.8%) and D (3.7%) are also very low. In contrast, significantly higher infection rates of G. duodenalis (9.5%; P<0.01) and E. bieneusi (10.8%; P = 0.01) were seen in Ward A, indicating that these pathogens were transmitted frequently within this ward during the cryptosporidiosis outbreak. Although a high carriage of C. difficile was found in control Wards C (37.8%) and D (27.8%), this is similar to the 16-35% carriage of C. difficile in hospital inpatients in other countries [30], and significantly lower than the infection rate of C. difficile in Ward A (60.8%; P<0.01). Previously, no data existed on the prevalence of C. difficile in hospitalized children in China, although an infection rate of 9.5% was reported in adults in Shanghai [14].

The low genetic diversity of G. duodenalis, E. bieneusi, and C. difficile found in Ward A versus in control wards also supports the hypothesis of concurrent transmission of these enteric pathogens during the cryptosporidiosis outbreak. For G. duodenalis, AII was the only subtype seen in Ward A, although assemblage B and both AI and AII subtypes of assemblage A are commonly found in humans around the world (7), and they were all found in other wards and hospitals in the present study (Table 3). For E. bieneusi, Peru 11 was the dominant genotype in Ward A, being found in half of the genotyped specimens, while it was only found in one of eight specimens each genotyped in Wards C and D (Table 3). For C. difficile, the fr-01 subtype was dominant in Ward A and accounted for 1/3 of all C. difficile infections in this ward, whereas the most prevalent genotype kr-03 in control Wards C and D accounted for less than 1/5 of all C. difficile infections (Table 3).

Outbreaks involving multiple enteric pathogens have been infrequently reported and, in the investigations of the few such outbreaks, sewage contamination of water or food was often the main cause for concurrent transmission of multiple enteric pathogens. For example, a waterborne outbreak of Shigella sonnei, Giardia, and Cryptosporidium infections on a Lake Michigan dinner cruise was caused by contamination of potable water with diluted sewage as the result of storm runoff in the cruise ship [31]. Another waterborne outbreak of gastroenteritis with multiple etiologies in resort island visitors and residents in Ohio in 2004 was caused by sewage contamination of groundwater [32]. Likewise, a national multi-pathogen outbreak of diarrheal illness in Botswana in 2006 was caused by sewage contamination of the environment during heavy rains in late 2005 and early 2006 [33]. Similarly, contact with manure from calves was responsible for two multi-pathogen outbreaks at a farm day camp in Minnesota [34]. In the present study, over half of the patients with enteric pathogens (31/58) in Ward A were infected with more than one pathogen, compared to a very limited number of cases (9/187) in control wards (P<0.01). Of note is the significant association of the three enteric pathogens examined in this study and occurrences of cryptosporidiosis in these children (Table 1).

As suggested in our previous investigation of cryptosporidiosis in these children [17], poor diaper changing and hand washing practices by caregivers were probably responsible for this multi-pathogen outbreak among pediatric inpatients in Ward A, Hospital I. Children in Ward A were orphans from a welfare institute. They were taken care of by hired caregivers. In contrast, children in other wards were primarily from the general community and cared for by their parents [17]. Considering the fact that most infections in Ward A occurred in children younger than 12 months (Table 2), who mostly stayed in cribs and beds, hired caregivers in Ward A might have acted as vehicles for the disease transmission among pediatric inpatients. This is also supported by the finding that in children under 12 months, Ward A had significantly higher infection rates of all study pathogens than Wards C and D, but in children older than 12 month, Ward A had only significantly higher infection rates of G. duodenalis than Wards C and D.

Very few studies have been conducted on molecular epidemiology of G. duodenalis, E. bieneusi, and C. difficile in China [10]-[16]. The occurrence of both assemblages A and B of G. duodenalis in non-outbreak children is in accordance with previous findings of near equal distribution of the two genotypes in 18 Giardia-positive humans in Henan [11] and 8 in Anhui [10]. In contrast, the dominance of Group 1 E. bieneusi genotypes in children in this study is different from the dominance of Group 2 genotypes in children in Jilin [13], although we also detected a novel Group 2 genotype in a child from Ward C (Figure 3; Table 3). The high diversity of known and novel E. bieneusi genotypes reported in this study and previous studies [12], [13] suggests that there is a need for more studies to examine the characteristics of E. bieneusi transmission in humans in China.

Ribotypes 027 and 078 are recognized as leading causes of nosocomial outbreaks of C. difficile infection in the world [9], [15]. However, neither has been reported in China thus far [14]-[16]. Interestingly, the dominant C. difficile slpA subtypes fr-01 in Ward A was previously characterized as toxin A-negative and toxin B-positive (A-B+), whereas the dominant subtype kr-03 in control wards was toxin A-positive and toxin B-positive (A+B+) [9]. In a previous study, A-B+ strains were the dominant ones (24.0%) in patients in three hospitals in Beijing, Shandong and Guangzhou in China [16]. The high prevalence of A-B+ strains in China indicates that toxin B, rather than toxin A, is probably a key virulence determinant as previously suggested [35]. Nevertheless, in the present study, no significant association was found between any of the C. difficile subtypes and the occurrence of diarrhea, although we previously showed a link between cryptosporidiosis and diarrhea in these children [17]. A new group of slpA subtypes including sh-01 and sh-02 were found in many children from all three wards (Figure 3; Table 3). Further studies are needed to better understand the public health importance of this new group of subtypes.

The results of this study and our previous study [17] showed that although Cryptosporidium infection was associated with the occurrence of diarrhea, single-pathogen infection with G. duodenalis, E. bieneusi, or C. difficile was not. None of the dominant genotypes/subtypes of G. duodenalis, E. bieneusi, and C. difficile were significantly associated with the occurrence of diarrhea, and concurrent infections of multiple pathogens were not more associated with occurrence of diarrhea than infections with single pathogens. The lack of differences in occurrence of diarrhea between children with single-pathogen infection and children with mixed infections in this study was probably attributable to the already high diarrhea rates in Ward A (58.1%) and low occurrence of mixed infections in control Wards C and D (9 cases of mixed infections versus 178 cases of single-pathogen infection). In addition, with the exception of Cryptosporidium, none of the other pathogens examined in this study were among the recently identified major pathogens for moderate-to-severe diarrhea in the Global Enteric Multicenter Study [1]. This has probably also made it difficult to use attributable fraction calculation in estimating the role of mixed infections in the occurrence of diarrhea in a hospital study setting with high occurrence of diarrhea.

In conclusion, using genotyping and subtyping tools we retrospectively identified a multi-pathogen outbreak in a pediatric hospital ward. As reported previously [17], this outbreak lasted ≥14 months, with ~60 inpatient children affected by cryptosporidiosis. Most of the Cryptosporidium-positive children were co-infected with G. duodenalis, E. bieneusi, or C. difficile. The young age of affected children and concurrent infections with multiple enteric pathogens clearly implicated poor diaper changing and hand washing by hired caregivers as the cause of the outbreak. Thus, better training of caregivers on hygienic practices such as hand washing and proper use of disposable gloves and disinfectants is needed to reduce the risk of pathogen transmission in healthcare facilities. Results of this study also highlight the importance of molecular epidemiologic investigations in understanding the transmission of enteric pathogens in hospitals.

Acknowledgments

We thank the attending physicians for assistance in patient enrollment. The findings and conclusions in this report are those of the authors and do not necessarily represent the views of the Centers for Disease Control and Prevention.

Author Contributions

Conceived and designed the experiments: LX YF. Performed the experiments: LW LD JY YG. Analyzed the data: LW LX MG LL YF. Wrote the paper: LW LX MG LL YF.

References

Kotloff KL, Nataro JP, Blackwelder WC, Nasrin D, Farag TH, et al. (2013) Burden and aetiology of diarrhoeal disease in infants and young children in developing countries (the Global Enteric Multicenter Study, GEMS): a prospective, case-control study. Lancet 382: 209-222. doi: 10.1016/s0140-6736(13)60844-2

Xiao L (2010) Molecular epidemiology of cryptosporidiosis: an update. Exp Parasitol 124: 80-89. doi: 10.1016/j.exppara.2009.03.018

Karanis P, Kourenti C, Smith H (2007) Waterborne transmission of protozoan parasites: a worldwide review of outbreaks and lessons learnt. J Water Health 5: 1-38. doi: 10.2166/wh.2006.002

Yoder JS, Beach MJ (2010) Cryptosporidium surveillance and risk factors in the United States. Exp Parasitol 124: 31-39. doi: 10.1016/j.exppara.2009.09.020

Mare A, Man A, Toma F, Szekely E, Lorinczi L (2007) Re-emergent etiologies: Clostridium difficile. Bacteriol Virusol Parazitol Epidemiol 52: 181-188

Henriques-Gil N, Haro M, Izquierdo F, Fenoy S, del Aguila C (2010) Phylogenetic approach to the variability of the microsporidian Enterocytozoon bieneusi and its implications for inter- and intrahost transmission. Appl Environ Microbiol 76: 3333-3342. doi: 10.1128/aem.03026-09

Feng Y, Xiao L (2011) Zoonotic potential and molecular epidemiology of Giardia species and giardiasis. Clin Microbiol Rev 24: 110-140. doi: 10.1128/cmr.00033-10

Matos O, Lobo ML, Xiao L (2012) Epidemiology of Enterocytozoon bieneusi infection in Humans. J Parasitol Res 2012: 981424. doi: 10.1155/2012/981424

Kato H, Ito Y, Akahane T, Izumida S, Yokoyama T, et al. (2010) Typing of Clostridium difficile isolates endemic in Japan by sequencing of slpA and its application to direct typing. J Med Microbiol 59: 556-562. doi: 10.1099/jmm.0.016147-0

Yong TS, Park SJ, Hwang UW, Yang HW, Lee KW, et al. (2000) Genotyping of Giardia lamblia isolates from humans in China and Korea using ribosomal DNA Sequences. J Parasitol 86: 887-891

Wang R, Zhang X, Zhu H, Zhang L, Feng Y, et al. (2011) Genetic characterizations of Cryptosporidium spp. and Giardia duodenalis in humans in Henan, China. Exp Parasitol 127: 42-45. doi: 10.1016/j.exppara.2010.06.034

Wang L, Zhang H, Zhao X, Zhang L, Zhang G, et al. (2013) Zoonotic Cryptosporidium species and Enterocytozoon bieneusi genotypes in HIV-positive patients on antiretroviral therapy. J Clin Microbiol 51: 557-563. doi: 10.1128/jcm.02758-12

Zhang X, Wang Z, Su Y, Liang X, Sun X, et al. (2011) Identification and genotyping of Enterocytozoon bieneusi in China. J Clin Microbiol 49: 2006-2008. doi: 10.1128/jcm.00372-11

Huang H, Wu S, Wang M, Zhang Y, Fang H, et al. (2008) Molecular and clinical characteristics of Clostridium difficile infection in a university hospital in Shanghai, China. Clin Infect Dis 47: 1606-1608. doi: 10.1086/593365

Huang H, Fang H, Weintraub A, Nord CE (2009) Distinct ribotypes and rates of antimicrobial drug resistance in Clostridium difficile from Shanghai and Stockholm. Clin Microbiol Infect 15: 1170-1173. doi: 10.1111/j.1469-0691.2009.02992.x

Yan Q, Zhang J, Chen C, Zhou H, Du P, et al. (2013) Multilocus sequence typing (MLST) analysis of 104 Clostridium difficile strains isolated from China. Epidemiol Infect 141: 195-199. doi: 10.1017/s0950268812000453

Feng Y, Wang L, Duan L, Gomez-Puerta LA, Zhang L, et al. (2012) Extended outbreak of cryptosporidiosis in a pediatric hospital, China. Emerg Infect Dis 18: 312-314. doi: 10.3201/eid1802.110666

El-Mahallawy HA, El-Din NH, Salah F, El-Arousy M, El-Naga SA (2004) Epidemiologic profile of symptomatic gastroenteritis in pediatric oncology patients receiving chemotherapy. Pediatr Blood Cancer 42: 338-342. doi: 10.1002/pbc.10394

Carraturo A, Catalani V, Tega L (2008) Microbiological and epidemiological aspects of rotavirus and enteric adenovirus infections in hospitalized children in Italy. New Microbiol 31: 329-336

Sulaiman IM, Fayer R, Bern C, Gilman RH, Trout JM, et al. (2003) Triosephosphate isomerase gene characterization and potential zoonotic transmission of Giardia duodenalis. Emerg Infect Dis 9: 1444-1452. doi: 10.3201/eid0911.030084

Lalle M, Pozio E, Capelli G, Bruschi F, Crotti D, et al. (2005) Genetic heterogeneity at the beta-giardin locus among human and animal isolates of Giardia duodenalis and identification of potentially zoonotic subgenotypes. Int J Parasitol 35: 207-213. doi: 10.1016/j.ijpara.2004.10.022

Caccio SM, Beck R, Lalle M, Marinculic A, Pozio E (2008) Multilocus genotyping of Giardia duodenalis reveals striking differences between assemblages A and B. Int J Parasitol 38: 1523-1531. doi: 10.1016/j.ijpara.2008.04.008

Sulaiman IM, Fayer R, Lal AA, Trout JM, Schaefer FW (2003) Molecular characterization of microsporidia indicates that wild mammals harbor host-adapted Enterocytozoon spp. as well as human-pathogenic Enterocytozoon bieneusi. Appl Environ Microbiol 69: 4495-4501. doi: 10.1128/aem.69.8.4495-4501.2003

Santin M, Fayer R (2009) Enterocytozoon bieneusi genotype nomenclature based on the internal transcribed spacer sequence: a consensus. J Eukaryot Microbiol 56: 34-38. doi: 10.1111/j.1550-7408.2008.00380.x

Kato H, Kato N, Watanabe K, Iwai N, Nakamura H, et al. (1998) Identification of toxin A-negative, toxin B-positive Clostridium difficile by PCR. J Clin Microbiol 36: 2178-2182

Wang Q, Zhou Y, Lv B, Fu KD, Chen L, et al. (2009) Prevalence of intestinal parasite infection in Kaifeng city, Henan Province (in Chinese). Chinese J Trop Med 9: 510-514

Sun YR, Xi JW, Yao HX, Xu JL, Wang HL, et al. (2010) The prevalence of intestinal parasites of inpatients in a hospital in Zhengzhou city, Henan Province (in Chinese). Chinese J Trop Med 10: 563-566

Xu LN, Wu GQ, Xi JW, Qi M, Yang N, et al. (2011) Prevalence of intestinal parasitic infection in children in a hospitalized children in Zhengzhou (in Chinese). Chinese J Trop Med 11: 17-22

Jiang ZX, Xu LQ, Yu SH, Li BS, Fang YY, et al. (1997) The epidemiology of giardiasis in China (in Chinese). Chinese J Publ Health 13: 407-408

Aslam S, Hamill RJ, Musher DM (2005) Treatment of Clostridium difficile-associated disease: old therapies and new strategies. Lancet Infect Dis 5: 549-557. doi: 10.1016/s1473-3099(05)70215-2

Serdarevic F, Jones RC, Weaver KN, Black SR, Ritger KA, et al. (2012) Multi-pathogen waterborne disease outbreak associated with a dinner cruise on Lake Michigan. Epidemiol Infect 140: 621-625. doi: 10.1017/s0950268811000896

O'Reilly CE, Bowen AB, Perez NE, Sarisky JP, Shepherd CA, et al. (2007) A waterborne outbreak of gastroenteritis with multiple etiologies among resort island visitors and residents: Ohio, 2004. Clin Infect Dis 44: 506-512. doi: 10.1086/511043

Mach O, Lu L, Creek T, Bowen A, Arvelo W, et al. (2009) Population-based study of a widespread outbreak of diarrhea associated with increased mortality and malnutrition in Botswana, January-March, 2006. Am J Trop Med Hyg 80: 812-818

Smith KE, Stenzel SA, Bender JB, Wagstrom E, Soderlund D, et al. (2004) Outbreaks of enteric infections caused by multiple pathogens associated with calves at a farm day camp. Pediatr Infect Dis J 23: 1098-1104

Lyras D, O'Connor JR, Howarth PM, Sambol SP, Carter GP, et al. (2009) Toxin B is essential for virulence of Clostridium difficile. Nature 458: 1176-1179. doi: 10.1038/nature07822