Secondary Logo

Share this article on:

Analysis of Treatment Outcomes for Recurrent Clostridium difficile Infections and Fecal Microbiota Transplantation in a Pediatric Hospital

Aldrich, Aileen M., MD*; Argo, Taylor, MD; Koehler, Tracy J., PhD; Olivero, Rosemary, MD§

The Pediatric Infectious Disease Journal: January 2019 - Volume 38 - Issue 1 - p 32–36
doi: 10.1097/INF.0000000000002053
Original Studies

Background: Clostridium difficile infection (CDI) is one of the most common nosocomial infections in the United States, with an increasing incidence in children. Approximately 20% of pediatric patients develop recurrent infections. It’s imperative to further analyze the incidence of recurrent CDI in the pediatric population and determine the most effective treatments. The primary goal of this study is to characterize children with recurrent CDI at our institution, including both hospital-acquired CDI (HA-CDI) and community-acquired CDI (CA-CDI) cases, summarize the various treatments utilized, including fecal microbiota transplant (FMT) and compare their success rates.

Methods: A retrospective cohort study of pediatric patients 1–21 years of age treated for CDI at a single institution from January 2010 to December 2014 was performed.

Results: There were 175 subjects with 215 separate episodes of CDI. Oral metronidazole was the most common initial treatment (145/207, 70%) followed by oral vancomycin (30/207, 15%), with recurrence rates of 30% (42/145) and 37% (11/30), respectively. Twenty-nine percent (63/215) of all initial CDI cases had at least 1 documented recurrence. Using multivariate analysis, subjects with HA-CDI were 2.6 times less likely to recur than those with CA-CDI (odds ratio: 0.39; 95% confidence interval: 0.18–0.85; P = 0.018). The overall success rate for FMT at our institution was 10/12 (83%).

Conclusions: Our data show that cases of HA-CDI were less likely to recur compared with CA-CDI. Although currently reserved for multiply-recurrent cases, FMT was highly successful in our small cohort. More studies on FMT should be conducted to further evaluate its usefulness in the treatment of recurrent CDI in children.

From the *Spectrum Health/Michigan State University/Helen DeVos Children’s Hospital Pediatric Residency Program

Michigan State College of Human Medicine

Spectrum Health, Research, Graduate Medical Education

§Section of Pediatric Infectious Diseases, Helen DeVos Children’s Hospital of Spectrum Health, Grand Rapids, Michigan; and Department of Pediatrics and Human Development, College of Human Medicine, Michigan State University, East Lansing, Michigan.

Accepted for publication March 8, 2018.

The authors have no funding or conflicts of interest to disclose.

Supplemental digital content is available for this article. Direct URL citations appear in the printed text and are provided in the HTML and PDF versions of this article on the journal’s website (

The authors have no funding or conflicts of interest to disclose.

Address for correspondence: Aileen M. Aldrich, MD, Division of Pediatric Infectious Diseases, Children’s Hospital & Medical Center, 8200 Dodge Street, Omaha, NE 68114. E-mail: or Rosemary Olivero, MD, Section of Pediatric Infectious Diseases, Helen DeVos Children’s Hospital, 35 Michigan St NE, MC 177, Grand Rapids, MI 49503. E-mail:

The human digestive tract is home to more than 500 different species of bacteria.1 The population of microorganisms that inhabit the digestive tract is referred to as the gut microbiome.2 These diverse bacteria are vital to the health of an individual as they aid in digestion, energy storage, immune function and protection of the host from the invasion of pathogens.2 In healthy individuals, the microbiome is in a state of homeostasis. However, this balance can be disrupted with the introduction of antibiotics, which can enable resident strains of Clostridium difficile (CD) to proliferate and release toxin in the intestine, causing symptomatic disease.2 , 3 Gastric acidity has also been shown to be an important factor in maintaining a normal microbiome, with increased rates of CD infection (CDI) seen in those who use acid suppressants, particularly proton-pump inhibitors.4 Patients who are immunocompromised or have underlying inflammatory bowel disease (IBD) are at increased risk of complicated infections from CD due to disruption of colonic immune function, inflammatory status or treatment with immune-modulating drugs.3 While anticlostridial antimicrobials are the mainstay of treatment of CDI, other treatment alternatives are being investigated. One of those alternatives is to use the microbiome of a healthy host to restore the intestinal microbiota of a diseased individual, termed fecal microbiota transplant (FMT).2

CDI is one of the most common nosocomial infections in the United States, affecting half a million people per year.5 Since 2005, community-acquired CDI (CA-CDI) has been increasingly reported among young, healthy individuals.6 CA-CDI reflects an estimated 20%–28% of all cases of CDI, with an incidence of 20–50 cases per 100,000 people in the United States, Sweden and England.6 In 2011, of those who were diagnosed with CDI, approximately 20% had a recurrence and 29,000 patients died within 30 days of the initial diagnosis,5 indicating that CDI is a major public health concern.7 The economic burden of CDI is substantial, costing the United States’ healthcare system approximately $4.8 billion dollars per year in acute care facilities alone.5 There have been numerous studies in the adult population regarding the optimal treatment of recurrent CDIs; however, there is little to no information about the severity and treatment of recurrent CDI in children.

Although children were previously considered a low-risk population for developing CA-CDI, the incidence of CDI in children has increased significantly in the last 20 years, and the majority of these cases are CA-CDI.8 Up to 35% of pediatric patients develop recurrent infections after treatment with first-line agents2; however, there is a wide variation in recurrence rates depending on the population examined. One study of 75 total pediatric cases of CDI in the United Kingdom reported recurrence in only 2.6% of cases,9 while another study done in the United States of 92 pediatric CDI cases found a recurrence rate of approximately 20%.10 No studies of pediatric CDI report on treatments for recurrent CDI.11 Therefore, it is necessary to further analyze the relapse rates and treatment regimens of recurrent CDI in the pediatric population. The majority of the literature on recurrent CDI in adults focuses on recurrences after hospital-acquired CDI (HA-CDI).3 , 4 As a large proportion of pediatric CDIs occur in the outpatient setting, current data may underestimate the true incidence of recurrent CDI in children and misrepresent the optimal treatment strategies.4 , 8 The primary goal of this study is to characterize children with recurrent CDI at our institution, including both hospital- and community-acquired cases, summarize the various treatments utilized, including FMT, and compare their success rates.

Back to Top | Article Outline


A retrospective cohort study of patients 1–21 years of age treated for CDI at Helen DeVos Children’s Hospital of Spectrum Health hospitals and affiliated clinics in Grand Rapids, Michigan, from January 1, 2010, to December 31, 2014, was performed. Helen DeVos Children’s Hospital is a free-standing children’s hospital with 190 inpatient beds and is the largest pediatric referral center in its region. Children less than 12 months of age were excluded because of the high rate of CD colonization in this age group.12

Each case of CDI was categorized as either initial or recurrent CDI. Initial CDI was defined as positive stool CD toxin polymerase chain reaction (PCR; GeneXpert C. difficile; Cepheid, Sunnyvale, CA) without CDI diagnosis in the past 60 days. Recurrent CDI was defined as initial CDI followed by either lack of symptom resolution or return of symptoms within 60 days of completion of previous treatment, whether or not there was a documented repeat test. Details were collected regarding each treatment given, including antibiotic agent(s), and FMT, if applicable. Each case was classified as either hospital acquired or community acquired. HA-CDI was defined as any patient diagnosed with CDI more than 48 hours after hospitalization or within 30 days of hospital discharge. CA-CDI included CDI diagnosed within 48 hours of hospital admission, at greater than 30 days after a recent hospital discharge, or without any inpatient hospital exposure. Subsequent CDIs in subjects older than 21 years were included if they had an initial CDI before 22 years of age.

Data for CDI cases were abstracted from the electronic medical records (EMRs) for each eligible subject and included basic demographic information, CDI treatments used, recent antibiotic use (any antibiotics received in the 30 days preceding CDI diagnosis per documentation or the medication administration record), presence of IBD and immunocompromised status (receiving chemotherapy for cancer treatment or chronic immunosuppressive/immune-modulating medication, including systemic steroids and biologics for at least 2 weeks before diagnosis, or documented neutropenia). Inhaled steroids were not considered as immunosuppressants in this study.

Back to Top | Article Outline

Study Procedures

The study was approved by the Spectrum Health Institutional Review Board. Subjects with an International Classification of Diseases, 9th revision, code of 008.45 (the only diagnosis code available for CDI) who were not excluded based on their age during the above time period were included in the study. Subject data were collected from the inpatient and outpatient EMRs. Twenty cases coded as CDI were excluded because of the total absence of clinical data.

Back to Top | Article Outline

FMT Protocol

Subjects considered for FMT for recurrent CDI at our facility complete an informed consent for this procedure, including discussions about procedural sedation and that, per the Food and Drug Administration, FMT is an investigational procedure with undefined long-term effects. At our facility, FMT is performed by providers in both pediatric infectious diseases and pediatric gastroenterology using a similar screening and stool preparation protocol. The stool donor is usually a close relative, frequently a parent or sibling. Banked donor stool has not been used by our facility. Screening labs for the stool donors include serum hepatitis panel (Hepatitis A IgM, Hepatitis B surface antigen, Hepatitis B core antibody IgM and Hepatitis C antibody), HIV-1/2 antibody and syphilis IgG antibody, as well as stool testing for CD toxin by PCR, complete ova and parasite screen and stool culture. Families are provided 6 catheter tip 60 mL syringes, gloves, masks, collection hat and large biohazard bags. Families are asked to purchase a blender which will subsequently be discarded. Families are instructed to collect the donor stool specimen in their home within 6 hours of procedure. A designated family member prepares the specimen at home by mixing the donor stool specimen with 240 mL distilled water, blending until in suspension, then drawing up liquid contents into six 60 mL catheter syringes. Subjects undergoing FMT with a pediatric gastroenterologist frequently have the specimen infused via colonoscopy in the pediatric sedation unit, distributing the donor sample throughout the colon. Occasionally the donor sample is given by temporary nasojejunal (NJ) tube. Subjects undergoing FMT with a pediatric infectious diseases specialist undergo temporary NJ tube placement in the pediatric sedation unit, and the donor stool sample is instilled through the NJ tube.

The term success was used for patients who had resolution of symptoms and no need for further interventions for CDI for 60 days after date of FMT. FMT failure was defined as return of symptoms and requirement for further CDI treatment within 60 days from date of FMT (a recurrence).

Back to Top | Article Outline

Statistical Analysis

Summary statistics were calculated for the study sample. Quantitative data are expressed as the mean ± standard deviation, while nominal data are expressed as a percentage. Analyses comparing HA-CDI and CA-CDI cases were performed using univariate logistic and multiple regression, with clustering on the subjects to obtain robust standard errors to adjust for nonindependent observations. Analyses were performed using Stata v.14.2 (StataCorp, College Station, TX). Selected patients were excluded from analyses when data were missing. Significance was assessed at P < 0.05.

Back to Top | Article Outline


Subject Characteristics

There were 175 subjects that fit our inclusion criteria, accounting for 215 separate episodes of CDI during the 5-year study period. Treatment data were available for 207 CDI episodes. Study sample characteristics are summarized in Table, Supplemental Digital Content 1, Overall mean age was 11.8 ± 6.8 years, with a median age of 13.8 years and a range of 1.1–22.8 years. There were approximately equal numbers of males and females, and the subjects were of predominantly Caucasian race. Approximately 34% of subjects were immunocompromised, and almost half had recent antibiotic exposure. Fourteen percent of the CDI cases were in subjects that had known or were later found to have IBD. The number of cases per year steadily increased from 2010 to 2014 (Fig., Supplemental Digital Content 2,

Back to Top | Article Outline


Table 1 outlines the treatments used for each CDI episode and recurrence. Oral metronidazole was the most common initial treatment, followed by oral vancomycin and combination of metronidazole and oral vancomycin. Recurrent CDI occurred in 30% of those initially treated with metronidazole compared with 37% of those initially treated with vancomycin (P = 0.356; Fig. 1). Twenty-nine percent (63/215) of all initial CDI cases had at least 1 documented recurrence (Table 2). Cases with more than 2 recurrences were often treated with novel or combination drug therapies such as rifaximin, nitazoxanide and other medications (Table 1). There were 12 documented FMTs performed for recurrent CDI in 10 subjects.







Recurrence of CDI was significantly more likely to be associated with CA-CDI (34%) cases than HA-CDI (22%) cases (Table 3). Those with HA-CDI were also significantly more likely to be immunocompromised and were more likely to have had recent antibiotic use and be given combination therapy for their initial CDI treatment. There was no significant difference in age, gender or IBD diagnoses between the 2 groups.



A multivariate logistic regression analysis was performed, with CDI recurrence as the dependent variable, and outpatient status, immunocompromised status, recent antibiotic use, presence of IBD and initial treatment type (oral metronidazole, oral vancomycin and combination of oral metronidazole and oral vancomycin; reference variable was oral metronidazole) as the independent variables. Subjects with HA-CDI were 2.6 times less likely to recur than those with CA-CDI (odds ratio: 0.39; 95% confidence interval: 0.18–0.85; P = 0.018). None of the other independent variables were significant predictors of recurrence.

Back to Top | Article Outline

Fecal Microbiota Transplant

There were a total of 12 FMTs performed on 10 different subjects. Eight out of 10 FMTs were documented as successful on the first attempt. Two subjects underwent a second FMT, both of which were successful in curing their recurrent CDI. Details of FMTs are listed in Table, Supplemental Digital Content 3, and Figure 2. The overall success rate for FMT at our facility was 83% (10/12). All 7 of the FMTs performed by colonoscopy were successful. The 2 failures were FMT given via enteric tube (1 gastric tube and 1 NJ tube). FMT was used in only 4.7% (10/215) of the overall cases of CDI. The majority of subjects who received FMT did so after at least 3 other treatments (Table 1), with the exception of 1 subject, who had a prior history of complicated recurrent CDI.



Back to Top | Article Outline


Our study shows that CDI is a disease affecting children in both the community and hospital settings. Our results reflect a slightly higher recurrence rate (29%) compared with recent pediatric literature (2.6%–20%).9 , 10 Recent antibiotic use, although thought to be the major trigger for the majority of CDI, was a risk factor in only half of CDI cases, suggesting that there may be other important underlying factors involved in the pathophysiology of CDI in children.

A striking finding in our study is the significantly lower rate of recurrence in HA-CDI cases. Subjects with HA-CDI were 3 times more likely to be immunocompromised and were over twice as likely to receive combination therapy with at least 1 drug via intravenous (IV) route as their initial treatment for their CDI. Several things in the hospital setting may explain those findings, including a lower threshold for CD testing, quicker initiation of treatment and ready access to IV metronidazole therapy. Incomplete medication adherence to outpatient treatment regimens may be a factor in the higher recurrence rates in the CA-CDI group. Although metronidazole is recommended to be given via the enteric route for treatment of CDI, those with HA-CDI were more likely to be treated with IV metronidazole compared with the CA-CDI group, suggesting that many of these subjects may have indeed had severe CDI disease. Given the retrospective nature of our study, we were unable to classify the severity of CDI within our cohort, which may have impacted antimicrobial treatment decisions and subsequent recurrence risk. In addition, our study did not have sufficient recurrent CDI cases to meaningfully address the question of whether combination antimicrobial therapy may reduce the risk of recurrent CDI, which is a practice not supported by the current literature.13

A possible bias in our study was in the classification of HA-CDI and CA-CDI cases. As the data were collected retrospectively, there is the possibility that patients were misclassified, perhaps leading to a falsely elevated rate of recurrence in the CA-CDI group. Furthermore, we relied on International Classification of Diseases, 9th revision, codes for diagnosis, which may have inadvertently excluded some otherwise eligible patients or misdiagnosed a patient with CDI who may have actually been a colonizer. Patients who may have had positive CD PCR, but no diagnosis code, were not reviewed. Another limitation of our study is that recent antibiotic use data were obtained by retrospective chart review and were limited to provider documentation in the EMR. This may have led to an underrepresentation of recent antibiotic use among our subjects. Due to the low number of subjects treated with FMT and incomplete documentation of FMT details, we are unable to provide a comprehensive comparison between all the variables of the different FMT techniques. However, our limited numbers do show a higher success rate for colonoscopy compared with all of the upper gastrointestinal instillation methods. A standardized protocol at individual institutions and ideally on a national level would help assure more uniform practice and clinical outcomes of FMT in the future.

Current adult guidelines recommend 10–14 days of oral metronidazole as the first-line agent for initial episode of mild to moderate CDI.13 Vancomycin given orally for 10–14 days is the initial regimen of choice for patients with severe CDI.13 Combination therapy with oral vancomycin and IV metronidazole should be used for severe, complicated cases of CDI, with colectomy reserved for those severely ill patients.13 For recurrent cases, the guidelines are less clear. Therapy for initial recurrence should be a repeat of the first regimen, again stratified by disease severity.13 Cases with further recurrences should trial vancomycin pulse or taper,13 but there are no evidence-based recommendations beyond that. The data in our cohort cannot speak to the differences in oral vancomycin dosing or duration for recurrent CDI. Delineating the severity of CDI was not feasible in our cohort; therefore, no assessment as to the appropriateness of therapy could be determined retrospectively. The standard initial treatment for CDI, metronidazole, was used most often in our cohort and had a high success rate as a first-line agent, slightly more so than for vancomycin. This is in keeping with the use of metronidazole as a first-line agent as it is well studied and very cost-effective for mild to moderate cases.13 However, it is known that metronidazole is not as effective in more severe cases and is not recommended beyond 2 treatment courses because of concern for neurotoxicity with ongoing use.13 , 14 Therefore, other treatment options must be chosen for cases with multiple recurrences.

Antibiotic therapy may not be the only way to successfully treat CDI. Several recent treatment advances have promising efficacy in preventing recurrent CDI. Monoclonal antibodies directed at CD toxins, such as bezlotoxumab, have been shown to reduce the rate of recurrence of CDI in treatment of initial or recurrent cases with 80% initial cure rate and 64% sustained cure rate (for 12 weeks).15 Although currently reserved for multiple resistant cases, our data on a small number of patients show that FMT is highly successful and can be considered earlier in the clinical course as a definitive treatment option for children with recurrent CDI. However, a recent study of recurrent CDI in adults comparing oral vancomycin followed by FMT to an oral vancomycin taper in a large adult cohort with recurrent CDI showed no significant difference in recurrence rates between the 2 groups.16 This study was limited by the lack of blinding and a majority of the randomized patients had previously failed at least 1 vancomycin taper, and the authors also noted that further study into the details of FMT protocols is needed.16 Another means of restoring a more physiologic gut microbiome is using commercial probiotics.17 While this has not proven on its own to be an effective preventive or treatment modality, probiotics may be a useful adjunct in treatment of CDI.17 We did not attempt to quantify or compare the use of probiotics in this study.

When considering novel therapies to treat recurrent CDI, cost must be evaluated. Gabriel and Beriot-Mathiot18 published a review article in 2014 estimating the cost and length of stay burdens for adult patients with CDI. Estimated total costs were $6774–$10,212 for CDI requiring admission and $2992–$29,000 for HA-CDI.18 The ranges for hospital length of stay were 5–13.6 and 2.7–21.3 days, respectively.18 One dose of bezlotoxumab costs approximately $3000–$4000.19–21 The mean cost of a single fecal transplant procedure, including screening donor and recipient for potential infectious risks as well as procedure and facility costs, was $1086 (range, $815–$1358) in 201122. The mean cost of a tapered oral vancomycin course (2-week course followed by 6-week taper) was estimated to be $2069 (range, $1836–$2303).22 Taking these cost data into consideration, the newest therapies for recurrent CDI may indeed prove more cost-effective in the long term, particularly for more severe cases that would otherwise have increased costs from prolonged hospital stays.

In summary, our study presents success rates of various treatment modalities for CDI in children. Our data show a higher recurrence rate of CDI compared with prior studies. Our study also shows a significantly lower recurrence in cases of HA-CDI, which is a population that is more likely to be immunocompromised. We had overall great success with FMT treatment in our patients with recurrent CDI. This, in addition to a review of costs in the literature, may show FMT to be a cost-effective alternative for successful treatment of patients with recurrent CDI. This study may serve as evidence for further development of FMT programs for children.

Back to Top | Article Outline


We thank Alan Davis, PhD, for his phenomenal contributions with statistical analyses, interpretations and manuscript review. We also thank George Fogg, MD, PhD, for his assistance in review of this manuscript.

Back to Top | Article Outline


1. Canny GO, McCormick BA. Bacteria in the intestine, helpful residents or enemies from within? Infect Immun. 2008;76:3360–3373.
2. Gough E, Shaikh H, Manges AR. Systematic review of intestinal microbiota transplantation (fecal bacteriotherapy) for recurrent Clostridium difficile infection. Clin Infect Dis. 2011;53:994–1002.
3. Garey KW, Sethi S, Yadav Y, et al. Meta-analysis to assess risk factors for recurrent Clostridium difficile infection. J Hosp Infect. 2008;70:298–304.
4. Dial S, Delaney JA, Barkun AN, et al. Use of gastric acid-suppressive agents and the risk of community-acquired Clostridium difficile-associated disease. JAMA. 2005;294:2989–2995.
5. Lessa FC, Mu Y, Bamberg WM, et al. Burden of Clostridium difficile infection in the United States. N Engl J Med. 2015;372:825–834.
6. Chitnis AS, Holzbauer SM, Belflower RM, et al. Epidemiology of community-associated Clostridium difficile infection, 2009 through 2011. JAMA Intern Med. 2013;173:1359–1367.
7. Russell G, Kaplan J, Ferraro M, et al. Fecal bacteriotherapy for relapsing Clostridium difficile infection in a child: a proposed treatment protocol. Pediatrics. 2010;126:e239–e242.
8. Khanna S, Pardi DS, Aronson SL, et al. The epidemiology of community-acquired Clostridium difficile infection: a population-based study. Am J Gastroenterol. 2012;107:89–95.
9. Pai S, Aliyu SH, Enoch DA, et al. Five years experience of Clostridium difficile infection in children at a UK tertiary hospital: proposed criteria for diagnosis and management. PLoS One. 2012;7:e51728.
10. Khanna S, Baddour LM, Huskins WC, et al. The epidemiology of Clostridium difficile infection in children: a population-based study. Clin Infect Dis. 2013;56:1401–1406.
11. McFarland LV, Ozen M, Dinleyici EC, et al. Comparison of pediatric and adult antibiotic-associated diarrhea and Clostridium difficile infections. World J Gastroenterol. 2016;22:3078–3104.
12. Pant C, Deshpande A, Altaf MA, et al. Clostridium difficile infection in children: a comprehensive review. Curr Med Res Opin. 2013;29:967–984.
13. Cohen SH, Gerding DN, Johnson S, et al; Society for Healthcare Epidemiology of America; Infectious Diseases Society of America. Clinical practice guidelines for Clostridium difficile infection in adults: 2010 update by the society for healthcare epidemiology of America (SHEA) and the infectious diseases society of America (IDSA). Infect Control Hosp Epidemiol. 2010;31:431–455.
14. Kapoor K, Chandra M, Nag D, et al. Evaluation of metronidazole toxicity: a prospective study. Int J Clin Pharmacol Res. 1999;19:83–88.
15. Wilcox MH, Gerding DN, Poxton IR, et al. Bezlotoxumab for prevention of recurrent Clostridium difficile infection. N Engl J Med. 2017;376:305–317.
16. Hota SS, Sales V, Tomlinson G, et al. Oral vancomycin followed by fecal transplantation versus tapering oral vancomycin treatment for recurrent Clostridium difficile infection: an open-label, randomized controlled trial. Clin Infect Dis. 2017;64:265–271.
17. Crow JR, Davis SL, Chaykosky DM, et al. Probiotics and fecal microbiota transplant for primary and secondary prevention of Clostridium difficile infection. Pharmacotherapy. 2015;35:1016–1025.
18. Gabriel L, Beriot-Mathiot A. Hospitalization stay and costs attributable to Clostridium difficile infection: a critical review. J Hosp Infect. 2014;88:12–21.
19. Sakoulas G. FDA Approval of Bezlotoxumab in Prevention of Recurrent Clostridium difficile Infection. NEJM Journal Watch. May 14, 2017. Available at: Accessed May 14, 2017.
20. Blink Health. Zinplava (bezlotoxumab). 2017. Available at: Accessed May 14, 2017.
21. Zinplava Prices, Coupons and Patient Assistance Programs. 2017. Available at: Accessed May 14, 2017.
22. Varier RU, Biltaji E, Smith KJ, et al. Cost-effectiveness analysis of fecal microbiota transplantation for recurrent Clostridium difficile infection. Infect Control Hosp Epidemiol. 2015;36:438–444.

Clostridium difficile; pediatric; fecal transplant

Supplemental Digital Content

Back to Top | Article Outline
Copyright © 2019 Wolters Kluwer Health, Inc. All rights reserved.