Introduction

Clostridioides difficile (C. difficile) is a Gram-positive anaerobic bacillus that is a major pathogen causing healthcare-associated infections [1]. Since the early 21st century, Clostridioides difficile infection (CDI) has been a major global public health problem, and is considered an “urgent threat” to human health by the United States Centers for Disease Control and Prevention [2]. C. difficile pathogenicity is mediated by the protein toxins A and B (encoded by tcdA and tcdB, respectively), which cause clinical symptoms ranging from self-limited diarrhea to life-threatening pseudomembrous colitis, toxic megacolon, and even death [3, 4]. The burden of this disease has increased over the past few decades, especially outbreaks of the hypervirulent strain RT027/NAP1/BI, increasing the morbidity and mortality of hospital-acquired infections worldwide [5]. It is estimated that CDI is responsible for over 500,000 enteric infections, 29,000 deaths and over $4.8 billion in healthcare costs each year in the United States [6]. In addition, CDI is no longer restricted to the hospital setting, and higher rates have been reported in traditionally low-risk populations, including children who have not been exposed to hospital settings [7].

A CDI treatment plan depends on whether it is an initial or recurrent episode, as well as its severity [8]. The first episode of non-severe or severe CDI can be treated with vancomycin or fidaxomicin. For fulminant CDI, vancomycin is the treatment of choice. In the event of ileus, vancomycin may also be administered through the rectum. Especially if an ileus is present, it is recommended to administer oral or rectal vancomycin with intravenous metronidazole. The first recurrence of CDI can be treated with oral vancomycin or fidaxomicin for 10 days, or a prolonged taper of pulse oral vancomycin. CDI patients with a second or subsequent recurrence can be treated with oral vancomycin therapy using a tapered and pulsed regimen, fidaxomicin, and vancomycin followed by Rifaximin or fecal microbiota transplantation (FMT).

It is thought that CDI develops because of an imbalance in the host intestinal microbiota, which can be caused by a variety of factors. Broad-spectrum antimicrobials are considered to be the most important risk factor for CDI in adults and children by destroying the host intestinal microbiota and reducing colonization resistance to C. difficile and other enteric pathogens [9]. Earlier studies have shown a marked reduction in bacterial diversity among subjects with CDI. CDI patients were enriched with Enterococcus, Enterobacteriaceae, Erysipelotrichaceae, and Gammaproteobacteria class, but there was a decrease in Ruminococcaceae, Lachnospiraceae, Bacteroidetes, and Clostridial clusters IV and XIVa [10].

The epidemiology of C. difficile in children is characterized by asymptomatic colonization in many infants, with the highest colonization rates (which can exceed 40%) particularly among infants younger than 12 months of age [11]. Thus, the consensus guidelines recommend testing for C. difficile only if infants (< 12 months of age) present with pseudomembranous colitis or toxic megacolon or if they have symptoms of clinically significant diarrhea in which other causes of diarrhea have been ruled out [8]. However, a study based on children with diarrhea in Kenya showed that children with diarrhea were more susceptible to C. difficile infection than children with rotavirus or Cryptosporidium infection [12]. Here, we describe the clinical manifestations, diagnosis, treatment, changes in the intestinal microbiota of a pediatric patient with CDI, and reviewed the literature to improve clinicians’ awareness of CDI.

Case presentation

Clinical presentation

The pediatric patient was a 1-month-old female who came to the outpatient department for the first time due to diarrhea (7–8 loose yellow stools/day) more than 20 days ago. Diarrhea continued after treatment with cefixime and probiotics (Saccharomyces boulardii). She was admitted to our hospital on March 4, 2021. The infant was born by caesarean section at full term and was diagnosed with hypoglycemia and necrotizing enterocolitis (NEC). She was treated with broad-spectrum antibiotics during hospitalization, including cefoperazone, meropenem, piperacillin, tazobactam, sulbactam, fluconazole, vancomycin, and azithromycin.

Clinical findings

Clinical examination on admission revealed: temperature, 36.8℃; heart rate, 136 beats/min; and respiratory rate, 35 breaths/min. The infant was conscious and presented pharyngeal hyperemia, mildly swollen tonsils, and coarse breath sounds. The patient did not have fine wet rales and lower limb edema.

Diagnostic focus and assessment

Laboratory tests performed on the day of admission revealed the following results: white blood cell count, 26.3 × 109/L; neutrophil count, 11.47 × 109/L; lymphocyte count, 6.87 × 109/L; red blood cell count, 3.36 × 1012/L; hemoglobin, 104 g/L; platelet count, 599 × 109/L; C-reactive protein (CRP), 157.2 mg/L; albumin, 31.4 g/L; lactate dehydrogenase, 355.0 U/L; and procalcitonin, 0.20 ng/ml. Stool routine examination showed a fecal white blood cell count of 99 − 120 per high power field (HPF) and fecal red blood cell count of 15 − 25/HPF (Supplementary Table S1). Bone marrow aspiration revealed granulocytosis and thrombocytosis and a granulocytic-to-erythroid ratio of 23%. Stool cultures revealed Enterococci as the main bacteria, with intestinal parasites, rotavirus, and common bacterial intestinal pathogens (Shigella, Salmonella, Pathogenic Escherichia coli, etc.) remaining undetected.

After admission, she was initially diagnosed with neonatal acute diarrhea according to the Textbook of Pediatrics (9th edition, People’s Health Publishing House, 2018) [13]. In this case, a laboratory test revealed elevated white blood cell count and CRP in peripheral blood, bloody purulent stool, and a routine stool examination revealed increased white and red blood cells. After excluding the above intestinal pathogens, the pathogen of this case was considered to be invasive bacteria, and empirical treatment was performed with cefmenoxime and norvancomycin (an analogue of vancomycin). After 3 days of treatment, the frequency of diarrhea decreased to about 4 times/day.

Given the patient’s long-term history of antibiotic use, C. difficile related tests were performed on the 3rd and 7th day of admission. As a result, enzyme immunoassay (EIA) for C. difficile glutamate dehydrogenase (GDH) and nucleic acid amplification tests (NAATs) for toxin genes were positive. Bacterial culture revealed colonies of C. difficile, which were confirmed to be sequence ty** 54 by multilocus sequence ty**. Based on the above test results, the patient was finally diagnosed with CDI, so cefmenoxime was discontinued and norvancomycin was continued.

Therapeutic focus and assessment

After we treated the patient with norvancomycin for 12 days, the infant stool frequency decreased, about 2–3 times/day, and the stool was sticky, with no fever, nausea, or abdominal pain, and was discharged. After discharge, intravenous antibiotics were discontinued and changed to a combination of oral norvancomycin and probiotics.

The patient was treated with three courses of antibiotics with oral norvancomycin after hospital discharge. After the first course (9 days) of treatment, the infant’s stools were viscous (Fig. 1a) with a frequency of about 3 times/day, and routine stool examination shows normal white and red blood cell counts (Supplementary Table S1). However, 7 days after treatment withdrawal, the patient began to have frequent loose stools (Fig. 1b), about 6 times/day. The routine stool examination showed elevated white blood cell count (Supplementary Table S1). C. difficile toxigenic culture (TC) and NAATs were positive. The second course of antibiotics was started based on symptoms and test results. After 25 days of treatment, the patient’s routine stool examination and frequency returned to normal (2–3 times/day) (Supplementary Table S1). The stool was mushy (Fig. 1c), and stool culture and NAATs were negative. Ten days after treatment discontinuation, the patient began to have frequent loose, mucus-containing stool (Fig. 1d), approximately 5 times/day. The routine stool examination showed fecal white blood cell count of 26–44/HPF (Supplementary Table S1), TC and NAATs were positive. The treatment period was extended based on clinical symptoms and the recurrence of diarrhea. In the third course, the dose of norvancomycin was adjusted five times sequentially according to clinical status as follows: 0.06 g twice a day for 10 days, 0.06 g four times a day for 29 days, 0.06 g twice a day for 17 days, 0.07 g once a day for 11 days, and 0.07 g once every 2 days for 31 days. Stool frequency returned to normal after the third course of treatment, about 2 times/day, the stool was normal (Supplementary Table S1), stool culture of C. difficile, GDH/toxin EIA, and NAAT were negative (Fig. 1e and f). The diagnosis and treatment details are shown in Fig. 2.

Fig. 1
figure 1

Stool characteristics during treatment. a Viscous stool after the first course of treatment; b Loose and yellow stool after treatment withdrawal; c Mushy stool after the second course of treatment; d Loose and yellow stool after treatment withdrawal; e Normal stool after the third course of treatment; f Normal stool during the 3-month follow-up

Full size image
Fig. 2
figure 2

Diagnosis and treatment of Clostridioides difficile infection. Samples were numbered according to the date of collection. The treatment cycle was divided into six stages (S1 to S6) according to sampling time and treatment dose. S1 to S5 correspond to different treatment doses, and S6 is the recovery stage

Full size image

16 S rRNA gene-based fecal microbiota profiling

To assess changes in the fecal microbiota during treatment, the treatment course was divided into six stages (S1 to S6) (≥ 3 stool samples per stage) according to sampling time and treatment dose (Figs. 2 and 3). S1 to S5 corresponded to different treatment doses, and S6 was the recovery stage. Samples were collected in a sterile collection tube, and stored at -80℃ until analysis.

Fig. 3
figure 3

Composition, abundance, diversity, and predominance of intestinal microorganisms in our patient. a Phylum-level taxonomic distribution of the microbial community; b Family-level taxonomic distribution of the microbial community; c Genue-level taxonomic distribution of the microbial community; d-e Alpha diversity analysis (Chao and Shannon index) of metagenomic sequencing data (≥ 3 samples per stage). Data are mean ± SEM. Differences between data were assessed using the Kruskal-Wallis test and Dunn’s test. *P < 0.05; f Principal coordinate analysis of metagenomic sequencing data at different stages (≥ 3 samples per stage); g-i Relative abundance of bacterial phyla (g), families (h) and genus (i) at different stages; j-k Predominant microorganisms across stages (≥ 3 samples per stage). (j) Hierarchical tree diagram based on linear discriminant analysis effect size; (k) Distribution histogram based on linear discriminant analysis

Full size image

The OMEGA Soil DNA Kit (M5635-02) (OMEGA Bio-Tek, Norcross, GA, USA) was used to extract total genomic DNA from stool samples. The fecal microbiota characteristics were determined by sequencing the V3-V4 region of the 16 S rRNA gene using the Illlumina NovaSeq platform with NovaSeq 6000 SP Reagent Kit (500 cycles) at Shanghai Personal Biotechnology Co., Ltd (Shanghai, China).

QIIME2 2019.4 was used for microbiome bioinformatics with slight modifications in accordance with the official tutorials (https://docs.qiime2.org/2019.4/tutorials/). Sequence data were demultiplexed using the demux plugin, followed by primers being cut with the cutadapt plugin. The DADA2 plugin then filtered, denoised, merged, and removed chimera from sequences [14]. An amplicon sequence variant (ASV)’s taxonomy was assigned using the classify-sklearn Naive Bayes classifier (Greengenes Database) in the feature-classifier plugin [15]. In order to obtain the sharing information between groups, the relative abundance of ASVs and Venn diagram was analyzed. Each group’s Alpha diversity level was then calculated based on the distribution of ASVs. In addition, to measure the difference in beta diversity between each group, the distance matrix for each sample was calculated, and the principal coordinate analysis (PCoA) was performed.

Composition analysis of the fecal microbiota of each sample is shown in Fig. 3a and c. Alpha diversity (Chao and Shannon indexes) was significantly higher at S6 than at S1, S2, S3, and S4 (P < 0.05, Fig. 3d and e). PCoA based on the Bray-Curtis distance showed a separation in the fecal microbiota structure at S5 and S6 relative to S2 (P < 0.005, Fig. 3f), indicating that the microbiota was significantly impacted by treatment. At the phylum level, Firmicutes was enriched at S5 and S6 (Fig. 3g). At the family level, Veillonellaceae and Lactobacillaceae were enriched (Fig. 3h). At the genus level, Veillonella and Lactobacillus were enriched (Fig. 3i). Bacterial phylotypes were identified at each stage using linear discriminant analysis effect size. The results showed no significant differences in phylotypes across stages (Fig. 3j and k). Fecal microbiota composition reflected by alpha and beta diversity changed over time, possibly due to CDI.

Follow-up and outcomes

Following three courses of norvancomycin treatment, stool frequency returned to normal, and diarrhea did not recur during three months of follow-up.

Discussion and conclusions

Early childhood is a crucial period during which the intestinal microbiota may impact current and future health status [16]. In adults, C. difficile primarily colonizes the lower intestinal tract and causes colonic inflammation by binding to toxins A and B to receptors on the plasma membrane [17, 18]. However, children’s intestines appear to be resistant to the effects of these toxins, and clinical infections are rare. C. difficile has been recovered from an average of 37% of stools in healthy infants younger than 1 month. The colonization rate decreases to approximately 30% between 1 and 6 months of age. During the first year of life, this rate declines until it reaches 10% in healthy infants. At 3 years old, the colonization rate is approximately 3%, similar to the adult carrier rate [19,20,21]. Neonatal resistance to CDI may be related to the absence of toxin receptors, downstream signaling pathways in the immature intestinal mucosa, and some protective factors in breast milk and the host intestinal microbiota [22, 23]. Therefore, the consensus guidelines recommend clinical testing for CDI only in the presence of clinical indications [8]. In recent studies, 26% of children hospitalized with CDI were younger than 1 year of age, and 5% were newborns [24]. Pediatric CDI is caused by a variety of factors, including age, gender, comorbidities, prolonged hospitalization and enteral feeding, but antibiotic exposure appears to be the main risk factor [25]. In spite of the fact that nearly all antibiotic classes can cause CDI, clindamycin, cephalosporins, and fluoroquinolones appear to pose the greatest threat [26]. Broad-spectrum antibiotic treatment can damage the host intestinal microbiota and reduce intestinal commensal flora diversity and beneficial bacteria abundance, leading to excessive growth and toxin production of C. difficile [27, 28]. Following the discontinuation of antibiotic treatment, these effects can persist for weeks or months, predisposing individuals to C. difficile infections.[29].

There is a wide range of clinical manifestations of CDI. It is possible for patients with a mild infection to have a single episode of diarrhea, as well as abdominal pain, bloody stools, purulent stools, or watery stools. In severe cases, patients may develop pseudomembranous colitis, fulminant colitis, toxic megacolon, and sepsis [4]. In this study, we searched Wanfang Data, China National Knowledge Infrastructure, and the Biomedical Literature Database (PubMed) for case reports on CDI in children. Studies that did not exclude other pathogens and had unclear detection methods related to C. difficile were excluded. A summary of the national and international studies is listed in Table 1 [30,31,32,33,34,35,36,37,38,39,40]. Combined with the literature review, the main clinical manifestations of CDI in children are fever, diarrhea, abdominal pain, vomiting and pseudomembranous colitis, and some children have rare clinical manifestations such as reactive arthritis. Increased white blood cell and CRP counts were found in laboratory tests. Our patient was treated with broad-spectrum antibiotics during hospitalization for hypoglycemia and NEC at birth. In the early stage of the onset, the main manifestations were diarrhea (loose yellow stool), repeated routine stool examinations were abnormal, and toxigenic C. difficile, GDH, and NAATs were positive. After excluding other causes, the patient was finally diagnosed with CDI.

Table 1 Cases of Clostridioides difficile infection in children
Full size table

Clinical history and laboratory tests are essential for the accurate diagnosis of CDI. This patient had diarrhea symptoms, abnormal stool routine examinations, and increased white blood cell count, CRP and muscle enzymes, which may be related to acute enteritis at the early stage of the disease. This clinical manifestation is difficult to distinguish from intestinal diseases caused by other predisposing factors. Therefore, the diagnosis of CDI in children remains an incredible clinical challenge. CDI is defined by the presence of symptoms (usually diarrhea) and either a stool test positive for C. difficile toxins or detection of toxigenic C. difficile or colonoscopic or histopathologic findings revealing pseudomembranous colitis [8]. However, no single laboratory test is considered the best. An immunoassay for GDH detects a highly conserved metabolic enzyme (common antigen) present in all isolates of C. difficile. GDH immunoassays cannot distinguish toxigenic C. difficile strains and lack specificity, so it is often used as a primary screening test for CDI [8]. TC has a high sensitivity (94–100%) and specificity (99%) which makes it a gold standard for laboratory diagnosis, but it has high experimental requirements and is not suitable for widespread use in clinical laboratories [41]. In addition, NAATs are capable of detecting the genes encoding C. difficile toxins A and B, making them an effective way to detect C. difficile [42]. However, when used in populations with high rates of C. difficile colonization, they may cause overdiagnosis of CDI due to their sensitivity [43]. Toxins in stool can distinguish colonization from infection more precisely, although recent studies have shown that asymptomatic children are more likely to have positive toxins in stool [44, 45]. The toxin EIA for C. difficile toxins A and/or B is inexpensive and easy to perform, but it is less sensitive than NAATs for detecting CDI and should not be used as a standalone test [8]. Consequently, the Infectious Diseases Society of America (IDSA) and the Society for Healthcare Epidemiology of America (SHEA) have provided some valuable recommendations for the laboratory diagnosis of C. difficile in adults and children [8]. To begin with, C. difficile testing should only be performed in patients with three or more unexplained unformed stools within 24 h. As a second recommendation, routine testing should be avoided in children under the age of 12 months unless other possible causes have been ruled out. Furthermore, ESCMID recommends multiple-step test for the accurate diagnosis of CDI [46]. The first test should be GDH assay or NAAT, which has a high negative predictive value. If the result is positive, the second assay should be highly specific, such as toxin EIA, which has a high positive predictive value. If the second test is positive, the final diagnosis is CDI. Patients with a negative second test for toxins should be re-evaluated (TC or NAATs) for the possibility of true infection. In this study, after excluding other potential causes, the infant was first tested for GDH combined with toxins by EIA, and the results showed GDH (+) and Toxins (-). Subsequently, to further evaluate whether the infant was truly infected, we performed TC and NAATs, and the results were positive. Finally, combined with the clinical symptoms, the patient was diagnosed with CDI.

Antibiotic therapy remains the first line of treatment of CDI, and antibiotics should be chosen according to guidelines and severity of the infection [41]. Currently, the treatment of CDI in children is based on clinical data from adults [47]. For children with their first episode or first recurrence of non-severe CDI, metronidazole and vancomycin are recommended. Oral vancomycin is preferred over metronidazole for children with a first episode of severe CDI [8]. Notably, new antibiotics may further disrupt intestinal microbiota, with approximately 25% of patients experiencing relapses within 4 weeks after antibiotic treatment [48]. FMT helps re-establish the intestinal microbiota and has a higher success rate than vancomycin when treating CDI that has relapsed/refractory [49]. Moreover, several studies have shown that probiotics reduce the incidence of antibiotic-associated diarrhea, and Saccharomyces boulardii, Lactobacillus acidophilus, Lactobacillus casei, and Lactobacillus rhamnosus can prevent primary or recurrent CDI [50,51,52]. In this case, after 3 days of intravenous norvancomycin (North China Pharmaceutical Company, China) according to the instructions, the frequency of diarrhea decreased. After continuing treatment with norvancomycin for 9 days, the infant’s stool frequency was normal, about 3 times/day, mainly viscous stools, and the condition was significantly improved. Stool culture of C. difficile and NAATs were still positive. After discharge, the patient was given oral norvancomycin and probiotic consolidation therapy. However, during this period, the patient had two relapses of diarrhea, accompanied by abnormal stool routine examination, C. difficile toxigenic culture, and NAATs were positive. Based on the clinical manifestations and laboratory diagnosis of the patient, the duration of the final treatment stage was extended, and the dose was adjusted in time according to the clinical status. Finally, her stool frequency returned to normal after three rounds of oral norvancomycin treatment. The patient had no recurrence of diarrhea, stool culture of C. difficile, GDH EIA, toxin EIA, and NAATs were negative during the 3-month follow-up period.

The ecology of the intestinal microbiota determines C. difficile colonization and virulence [53]. It has been reported that the alpha diversity of the intestinal microbiota decreased significantly in infants with NEC, and the abundance of Bifidobacterium and Lactobacillus decreased [54, 55]. The strictly anaerobes in the intestinal of patients with NEC are Clostridium species (C. butyricum, C. neonatale, C. perfringens, C. paraputrificum, and C. difficile), which are associated with NEC in preterm infants [56, 57]. Therefore, in our case, we hypothesized that the intestinal microbiota was altered by NEC and antibiotic therapy, facilitating the development of CDI. However, the relationship between NEC and CDI still needs to be further elucidated. On the other hand, the microbiota differs between caesarean section born and vaginally delivered infants over the first year of life, showing enrichment of Enterococcus, Enterobacter, Clostridium perfringens and Klebsiella, and reduction of Bacteroides and Bifidobacterium in caesarean section born infants [58,59,60,61]. Our patient was born by caesarian section. These data suggested that caesarian section may also be a risk factor for CDI in infants.

In the recovery of diseases, the intestinal microbiota plays an equally important role. Studies have shown that the abundance of Firmicutes and Bacteroidetes decreased while the abundance of Proteobacteria increased in children with toxin-positive C. difficile [62]. In the present study, the intestinal microbiota improved in the late stage of treatment and recovery period, demonstrated by the enrichment of Firmicutes and Lactobacillus and the increase in alpha diversity. Lactobacillus species, used as probiotics, regulate the intestinal microbiota, reduce intestinal inflammation, enhance host immune function, maintain the integrity of the intestinal barrier, and inhibit the production of toxins A and B [63, 64]. On the other hand, increased alpha diversity in the intestinal microbiota was also associated with age and complementary feeding [16, 65]. The effect of age and formula feeding in recovery was also considered during the study period. Further studies are necessary to determine the evolution and characteristics of the intestinal microbiota to better understand the relationship between C. difficile-associated dysbiosis and infant development.

In summary, based on the literature review and this case report, clinicians should also pay attention to diarrhea caused by C. difficile in infants and young children. More strong evidence is needed to explain the true prevalence of CDI in this population and to better understand the C. difficile-associated diarrhea in infants.