Abstract
Background:
Patients undergoing hematopoietic stem cell transplant (HSCT) possess numerous risk factors for Clostridioides (formerly Clostridium) difficile infection (CDI) and experience a high rate of diarrhea. Colonization rates of Clostridium difficile vary greatly among subgroup analyses with recent studies demonstrating colonization rates in the blood and marrow transplant units up to nine times that of the general population.
Methods:
The primary objectives of this study were to identify the rate of C difficile colonization and acquisition in HSCT patients admitted to the blood and marrow transplant unit. This was a prospective study that included all adult patients admitted for hematopoietic stem cell transplantation. Stool specimens were routinely collected on admission and weekly thereafter for a maximum of six samples per patient.
Results:
Forty-two patients met inclusion criteria and had baseline samples available for analysis. The rate of C difficile colonization on admission was 24%, and an additional 9% of patients acquired the organism during admission. Twelve percent of patients developed CDI that was diagnosed clinically. Univariate analysis showed an increased risk of colonization for patients with three or more prior chemotherapy cycles.
Conclusions:
Given high colonization rates coupled with high risk of CDI in this population, providers must be judicious when testing for CDI and interpreting test results for HSCT patients.
Keywords: acquisition, clostridioides difficile, clostridium difficile, colonization, hematopoietic stem cell transplantation, infection, rate
1 |. INTRODUCTION
Clostridioides (formerly Clostridium) difficile infection (CDI) is the most common cause of hospital-acquired infection in the United States.1 The Infectious Diseases Society of America (IDSA) defines CDI as “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.2” CDI is of significant concern in patients undergoing hematopoietic stem cell transplant (HSCT) because of increased exposure to healthcare facilities, immunosuppression, and frequent antibiotic use, all of which are established risk factors for disease.1 CDI is nine times more prevalent in HSCT patients, with a higher rate in allogeneic transplants versus autologous transplants.2–4 The purpose of this study was to identify rates of C difficile colonization on admission in our HSCT patients as well as to establish acquisition rates during hospitalization for their transplant. Secondary objectives were to elucidate risk factors for colonization and acquisition.
2 |. METHODS
2.1 |. Study population
We conducted a prospective study in the blood and marrow transplant unit (BMTU) at West Virginia University (WVU) Hospitals. The WVU Protocol Review & Monitoring Committee and the Institutional Review Board approved this study. All patients admitted for HSCT were screened on admission for inclusion. Inclusion criteria encompassed any adult patient ≥18 years admitted for transplantation with an expected hospital duration of at least 4 weeks. Patients with active CDI on admission, as defined by receiving antibiotics for the treatment of C difficile, were excluded.
2.2 |. Study methodology
Patients were screened and enrolled in the study by the treatment team. Stool samples were collected and sent to the clinical laboratory on admission and weekly (±72 hours) thereafter during their hospitalization, for a maximum of 6 tests. All results were recorded for research purposes only and were not disclosed in the patient chart or to the treatment team. Active CDI was diagnosed based on signs, symptoms, and clinical (ie, non-study) testing ordered at the discretion of the medical treatment team. Baseline data collection included demographics, prior documented CDI within 6 months, prior antibiotic agents within 30 days, prior antibiotic agents within 90 days, prior admissions within 90 days, and transplantation donor source. Additional prospective weekly data collection included organ and grade of graft-versus-host disease, presence and grade of mucositis, current anti-infective agents, proton-pump inhibitor use, histamine-2 antagonist use, corticosteroid use, and infectious complications.
2.3 |. Clostridioides (formerly Clostridium) difficile testing and typing
As per standard procedures at WVU Hospitals, the laboratory used a lateral flow immunoassay (Quik Chek Complete; Alere) to detect glutamate dehydrogenase (GDH), toxin A, and toxin B. Any specimen that was GDH(+)/toxin(−) or GDH(−)/toxin(+) was reflexed to a PCR for tcdB (Verigene CDF; Nanosphere, Northbrook, IL, USA, or Xpert C difficile; Cepheid, Sunnyvale, CA, USA). Testing was performed on all samples, whether formed or unformed stool.
2.4 |. Definitions
Clostridioides (formerly Clostridium) difficile colonization was defined as a positive laboratory result (either GDH− and toxin immunoassay-positive or PCR-positive) in the absence of significant diarrhea. C difficile acquisition was defined by a positive result for C difficile after admission, where the admission sample was negative. CDI was defined as clinical symptoms, positive C difficile testing ordered by a clinician, and treatment with therapeutic agents (ie, metronidazole or oral vancomycin).
2.5 |. Statistical analysis
All patients who met study criteria and produced at least one sample were included in analysis. Fisher’s exact test was used in the univariate data analysis on the following variables to evaluate the risk of colonization: gender, prior CDI, prior hospitalization, prior antibiotic therapy, cancer diagnosis, transplantation type, conditioning regimen, prior rituximab treatment, acquired colonization, and number of prior chemotherapy cycles. Statistical inferences were based on two-sided tests at a significance level of P < .05.
3 |. RESULTS
Forty-five patients admitted to the BMTU between October 2015 and June 2017 met inclusion criteria and were enrolled in the study. Two patients did not receive a transplant and one patient did not provide an admission week sample so were excluded, leaving a total of forty-two patients included in the data analysis. Table 1 shows the baseline characteristics of the study population. The average age of the population was 51 years (range: 22–70 years), the majority were Caucasian (95%), and there were 24 females (57%) and 18 males (43%). The most prevalent cancer diagnosis was non-Hodgkin lymphoma (41%). Patients had a median length of stay of 22 days (range: 10–37 days). Median neutrophil engraftment (neutrophil count: ≥0.5 × 109/L) occurred at 13 days (range: 9–29 days) post-transplant. Two patients had prior CDI within 6 months of study inclusion, and these were the only patients with any prior CDI at any time point. Defined daily doses (per 1000 patient days) of pertinent antibiotic agents in the study population were doxycycline (56), linezolid (9), vancomycin IV (125), vancomycin PO (17), piperacillin/tazobactam (108), ceftriaxone (26), ceftazidime (2), cefepime (141), meropenem (146), ertapenem (10), levofloxacin (326), and metronidazole (47).
TABLE 1.
Patient demographics
| Demographic | N=42 |
|---|---|
| Age (years), median | 51 (range: 22–70) |
| Race (Caucasian) | 40 (95%) |
| Gender (male) | 18 (43%) |
| Length of stay (days), median | 22 (range: 10–37) |
| Cancer diagnosis | |
| NHL | 17 (41%) |
| AML | 11 (26%) |
| MDS | 7 (17%) |
| ALL | 3 (7%) |
| HL | 3 (7%) |
| MM | 1 (2%) |
| Donor type | |
| Autologous | 16 (38%) |
| Matched related donor | 12 (29%) |
| Matched unrelated donor | 10 (24%) |
| Cord blood | 3 (7%) |
| Haploidentical | 1 (2%) |
| Prior chemotherapy regimens, median | 2 (range: 0–7) |
| Antibiotics in prior 30 d | 3 (15%) |
| Antibiotics in prior 90 d | 19 (45%) |
| Prior CDI within 6 mo | 2 (5%) |
| Prior admission | |
| Within 30 d | 10 (24%) |
| Within 60 d | 14 (33%) |
| Within 90 d | 18 (43%) |
Abbreviations: ALL, acute lymphoblastic leukemia; AML, acute myeloid leukemia; HL, Hodgkin lymphoma; MDS, myelodysplastic syndrome; MM, multiple myeloma; NHL, non-Hodgkin lymphoma.
A total of 127 stool samples were collected with a median of 3 samples per patient [range: 1–6]. Full details of individualized results are included in Table 2. Ten patients had positive tests on admission, giving a baseline colonization rate of 24% (95% CI, 12%−35%). Five patients (12%) were clinically diagnosed with CDI, all of whom exhibited colonization on admission. Among the 32 patients not colonized at admission, 3 patients (9%) acquired C difficile colonization at some point during their hospitalization.
TABLE 2.
Full description of C difficile testing results
| Patient | Baseline | Week 1 | Week 2 | Week 3 | Week 4 | Week 5 |
|---|---|---|---|---|---|---|
| 1 | Positive | Positive | Negative | Positive | Discharged | Discharged |
| 4 | Negative | Negative | Negative | Negative | Discharged | Discharged |
| 5 | Negative | Negative | Negative | Negative | NC | Discharged |
| 6 | Positive | Positive | Negative | Negative | NC | Discharged |
| 7 | Negative | Negative | Negative | NC | Discharged | Discharged |
| 8 | Negative | Negative | NC | Discharged | Discharged | Discharged |
| 9 | Negative | Negative | Negative | NC | Discharged | Discharged |
| 10 | Negative | NC | Negative | NC | Negative | Positive |
| 11 | Negative | Negative | Negative | Negative | NC | Discharged |
| 12 | Positive (CDI) | Negative | NC | Negative | Negative | Discharged |
| 14 | Negative | Negative | Negative | NC | NC | Discharged |
| 15 | Negative | NC | Negative | NC | Discharged | Discharged |
| 16 | Negative | Negative | Negative | NC | NC | Discharged |
| 17 | Negative | NC | Negative | Negative | Discharged | Discharged |
| 18 | Negative | Negative | Negative | NC | Discharged | Discharged |
| 19 | Positive | Positive | Positive | Positive (CDI) | Negative | Negative |
| 20 | Negative | Negative | NC | NC | Discharged | Discharged |
| 21 | Negative | Negative | Negative | Discharged | Discharged | Discharged |
| 22 | Positive (CDI) | Negative | Negative | NC | Discharged | Discharged |
| 23 | Negative | Negative | Negative | Discharged | Discharged | Discharged |
| 24 | Positive | Positive (CDI) | Negative | Discharged | Discharged | Discharged |
| 26 | Negative | Negative | Negative | NC | Negative | Discharged |
| 27 | Negative | Negative | Negative | Discharged | Discharged | Discharged |
| 28 | Negative | NC | Negative | Negative | Discharged | Discharged |
| 29 | Negative | Negative | Negative | Negative | Discharged | Discharged |
| 30 | Negative | NC | Negative | Discharged | Discharged | Discharged |
| 31 | Negative | Negative | Positive | NC | Discharged | Discharged |
| 32 | Negative | Negative | Negative | Discharged | Discharged | Discharged |
| 33 | Negative | Positive | Negative | NC | NC | Discharged |
| 34 | Negative | Negative | Negative | Discharged | Discharged | Discharged |
| 35 | Negative | Negative | Negative | Discharged | Discharged | Discharged |
| 36 | Positive | NC | NC | Discharged | Discharged | Discharged |
| 37 | Positive | Negative | NC | Discharged | Discharged | Discharged |
| 38 | Negative | Negative | Negative | Negative | NC | Discharged |
| 39 | Negative | Negative | NC | NC | Discharged | Discharged |
| 40 | Negative | Negative | Negative | Discharged | Discharged | Discharged |
| 41 | Positive | NC | Positive | Positive | NC | Positive (CDI) |
| 42 | Positive | Negative | Negative | NC | Discharged | Discharged |
| 43 | Negative | Negative | Negative | Negative | NC | NC |
| 45 | Negative | Negative | Negative | Negative | Discharged | Discharged |
| 46 | Negative | NC | Negative | NC | Discharged | Discharged |
| 47 | Negative | NC | NC | NC | NC | Discharged |
Abbreviations: CDI, Clostridioides difficile infection requiring treatment; NC, not collected.
Univariate analysis showed an increased risk of colonization for patients with three or more prior chemotherapy cycles (10% colonization with ≤2 prior cycles versus 58% with ≥3 prior cycles; P = .003). There was a trend towards increased colonization in patients with prior hospitalization within 90 days of admission (13% vs 39%; P = .07) and decreased risk with prior rituximab treatment (0% vs 31%; P = .084), but these did not reach statistical significance. There were no significant differences in the other variables evaluated.
4 |. DISCUSSION
Our study demonstrated a rate of C difficile colonization on admission of 24% in the HSCT population. This rate aligns with other studies that have reported rates of 10.7% to 29%.5,6 Compared to the general population, this rate of colonization is substantially higher, although carriage rates for C difficile continue to rise with one recent study of general medicine patients demonstrating a colonization rate on admission of 15%.1 Our observed rate of C difficile acquisition was 9% over a median of 22 hospital admission days, which is similar to acquisition rates among general medicine ward patients reported in one prior study (7.9%)7 but substantially lower than that of another (21%).8
Our study found that 50% of patients who were colonized with C difficile on admission developed infection, as documented by the treatment team. It is difficult to assess whether these patients truly had CDI, or whether diarrhea was from other causes and C difficile was detected in the stool sample because of colonization. Considering the high rate of HSCT patients colonized with C difficile, judicious use of testing should be considered with the high likelihood of baseline asymptomatic colonization. It is important to ensure patients are off all laxatives, diarrhea is persistent, and other causes of diarrhea are considered.
With 50% of colonized patients developing CDI, prophylaxis could be evaluated in these patients. Recent studies have evaluated CDI prophylaxis in HSCT patients with good success. Two single-center studies have reported oral vancomycin prophylaxis compared with no prophylaxis and found a reduction in CDI occurrence.9,10 Another study compared fidaxomicin with placebo for CDI prophylaxis for HSCT patients receiving fluoroquinolones and showed an incidence of C difficile-associated diarrhea through 30 days of 4.3% in the fidaxomicin group versus 10.7% in the placebo group, which was statistically significant.11 A final study evaluated oral vancomycin use in any patients previously diagnosed with CDI who were hospitalized for treatment with antibiotics. This study showed a CDI rate of 4% in the prophylaxis group and 27% in the control group.12 Further studies are needed to determine the role of CDI prophylaxis or de-colonization in clinical practice.
Interestingly, some patients colonized on admission became negative on subsequent testing. Some of these patients received treatment for documented CDI (patients 12, 19, 22, 24), which could account for this finding (Table 2). Yet, four additional patients colonized on admission who did not receive CDI treatment had subsequent negative testing results (patients 1, 6, 36, and 42), while one patient (33) exhibited intermittent positivity throughout hospitalization. These findings could reflect transient colonization, low bacterial burden near assay limits of detection, or suboptimal test performance for asymptomatic patients providing formed stool specimens (a non-validated source for our lab).
Potential explanations for the seemingly low acquisition rate noted in our study include the broad preventative measures in place in our facility (eg, decontamination of CDI patient rooms with hydrogen peroxide vapor and UV lights and increased surveillance of hand hygiene) as well as stricter isolation policies among transplant wards. We have an active Antimicrobial Stewardship Program at our hospital. Substantial efforts are in place to monitor antibiotic usage and antimicrobial resistance rates. We utilize antibiotic restrictions to the limit usage of agents that promote CDIs such as quinolones and carbapenems. We also utilize 72-hour electronic alerts to reevaluate the necessity of all antibiotics and prompt de-escalation. To help understand our prescribing patterns, we have mandatory indications for use when ordering antibiotics. We feel without these initiatives the colonization and acquisition rates would be even higher than described in our study.
Limitations of our study include small sample size, missed sample collections at some time points (see Table 2), and difficulty obtaining timely admission samples. Even though we found higher number of prior chemotherapy cycles to be a risk factor for colonization, some other variables might have been significant as well with a larger sample size. Though our colonization on admission rates was similar to other studies, we allowed up to eight days after admission to collect the initial sample. Some of our colonized-on-admission patients could have actually acquired colonization during the first week, making our admission colonization rates falsely elevated and acquired colonization falsely low. Eight patients (19%) had admission stool samples collected after three days of hospitalization. Two patients had samples collected on day 4, one on day 5, two each on days 6 and 7, and one on day 8. Only two of these patients (one sample on day 4 and one on day 6) had positive results on the admission sample. Another limitation was the use of immunoassay and PCR by the laboratory because of resource constraints rather than toxigenic culture (ie, enriched or heat/ethanol-shocked anaerobic culture followed by cell-culture cytotoxin neutralization assay), which is the gold standard method, particularly for carrier detection. A previous study of asymptomatic colonization reported a lower sensitivity by PCR compared with culture,13 which would suggest that our study may have underestimated the actual burden of colonization and acquisition.
5 |. CONCLUSION
Patients undergoing hematopoietic stem cell transplantation have a high rate of colonization with C difficile, as evidenced by this and other studies. The acquisition of C difficile during hospital admission was lower than expected in this population. Because of high rate of non-infectious diarrhea, this patient population may be prone to asymptomatic colonization receiving CDI diagnoses. C difficile testing should be conducted judiciously, assessing each patient for other causes of diarrhea before sending a specimen for testing. Further research on C difficile colonization and CDI prophylaxis would be useful in this patient population.
ACKNOWLEDGMENTS
SW acknowledges the WVCTSI support from the National Institute of General Medical Sciences grant U54GM104942.
Funding information
National Institute of General Medical Sciences, Grant/Award Number: U54GM104942
Footnotes
CONFLICT OF INTEREST
None to report.
REFERENCES
- 1.Alasmari F, Seiler SM, Hink T, Burnham CD, Dubberke ER. Prevalence and risk factors for asymptomatic clostridium difficile carriage. Clin Infect Dis. 2014;59:216–222. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.McDonald LC, Gerding DN, Johnson S, et al. Clinical Practice Guidelines for Clostridium difficile Infection in Adults and Children: 2017 Update by the Infectious Diseases Society of America (IDSA) and Society for Healthcare Epidemiology of America (SHEA). Clin Infect Dis. 2018;66:1–48. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Chopra T, Chandrasekar P, Salimnia H, Heilbrun LK, Smith D, Alangaden GJ. Recent epidemiology of Clostridium difficile infection during hematopoietic stem cell transplantation. Clin Transplant. 2011;24:E82–E87. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Zacharioudakis IM, Ziakas PD, Mylonakis E. Clostridium difficile Infection in the Hematopoietic Unit: A meta-analysis of published studies. Biol Blood Marrow Transplant. 2014;20:1650–1654. [DOI] [PubMed] [Google Scholar]
- 5.Jain T, Croswell C, Urday-Cornejo V, et al. Clostridium difficile colonization in hematopoietic stem cell transplant recipients: a prospective study of the epidemiology and outcomes involving toxigenic and nontoxigenic strains. Biol Blood Marrow Transplant. 2016;22:157–163. [DOI] [PubMed] [Google Scholar]
- 6.Bruminhent J, Wang ZX, Hu C, et al. Clostridium difficile colonization and disease in patients undergoing hematopoietic stem cell transplantation. Biol Blood Marrow Transplant. 2014;20:1329–1334. [DOI] [PubMed] [Google Scholar]
- 7.Clabots CR, Johnson S, Olson MM, Peterson LR, Gerding DN. Acquisition of Clostridium difficile by hospitalized patients: evidence for colonized new admissions as a source of infection. J Infect Dis. 1992;166(3):561–567. [DOI] [PubMed] [Google Scholar]
- 8.McFarland LV, Mulligan ME, Kwok RY, et al. Nosocomial acquisition of Clostridium difficile infection. N Engl J Med. 1989;320(4):204–210. [DOI] [PubMed] [Google Scholar]
- 9.Ganetsky A, Han JH, Hughes ME, et al. Oral vancomycin prophylaxis is highly effective in preventing Clostridium difficile infection in allogeneic hematopoietic cell transplant recipients. Clin Infect Dis. 2019;68(12):2003–2009. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Morrisette T, Van Matre AG, Miller MA, et al. Oral vancomycin prophylaxis as secondary prevention against Clostridioides difficile infection in the hematopoietic stem cell transplantation and hematologic malignancy population. Biol Blood Marrow Transplant. 2019;25(10):2091–2097. [DOI] [PubMed] [Google Scholar]
- 11.Mullane KM, Winston DJ, Nooka A, et al. A randomized, placebo-controlled trial of fidaxomicin for prophylaxis of Clostridium difficile-associated diarrhea in adults undergoing hematopoietic stem cell transplantation. Clin Infect Dis. 2019;68:196–203. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Van Hise NW, Bryant AM, Hennessey EK, et al. Efficacy of oral vancomycin in preventing recurrent Clostridium difficile infection in patients treated with systemic antimicrobial agents. Clin Infect Dis. 2016;63:651–653. [DOI] [PubMed] [Google Scholar]
- 13.Galdys AL, Nelson JS, Shutt KA, et al. Prevalence and Duration of Asymptomatic Clostridium difficile Carriage among Healthy Subjects in Pittsburgh, Pennsylvania. J Clin Microbiol. 2014;52(7):2406–2409. [DOI] [PMC free article] [PubMed] [Google Scholar]