Introduction
Leukemic patients undergoing hematopoietic stem cell transplantation (HSCT) receive significant risk factors of graft-versus-host disease, including immunosuppressant drugs, chemotherapy and broad-spectrum antibiotics (to avoid infectious complications). Empirical antibiotic therapy in HSCT patients increases the risk of developing Clostridium difficile infection (CDI) during hospitalisation [Reference Kinnebrew1]. Considerable CDI rates of 4–27% in HSCT recipients force the early detection and control of CDI in bone marrow transplanted patients [Reference Kamboj2]. Accurate and quick detection of toxin genes in samples could be applied for patients with antibiotic-associated diarrhoea, especially those who undergo transplantation or chemotherapy. Conventionally, detection of C. difficile is based on a two-staged culture and PCR algorithms that confirm C. difficile-specific gene gluD (glutamate dehydrogenase (GDH); diagnosis may be completed by detecting toxins by either ELISA or tracing toxin genes with PCR-based methods. The time-consuming procedure has been suggested in guidelines as the gold standard method for approaching CDI diagnosis [Reference Smits3].
Clostridium difficile pathogenesis depends on producing toxins A and/or B (TcdA and TcdB, respectively) which are interesting targets for the detection of pathogenic types. The main objective of the present study was to evaluate the CDI rates in leukaemic patients receiving stem cell transplantation by targeting tcdA and tcdB in a rapid multiplex real-time PCR method. This procedure may be suggesting a short-cut method for considerably rapid and accurate detection of toxigenic C. difficile.
Patients
From December 2017 to November 2018, 364 patients were admitted for hematopoietic stem cell transplantation. Isolation of the patients was fulfilled according to the CDC (Guideline for Isolation Precautions: Preventing Transmission of Infectious Agents in Health Care Settings) and the World Health Organisation (WHO Guidelines on Hand Hygiene in Health care). Eighty-six HSCT adult patients, receiving antibiotics and having diarrhoea (3 < unformed stools a day), were investigated for toxigenic CDIs. The Ethics Committee of Tarbiat Modares University (IR.TMU.REC.1395.403) and Digestive Disease Research Institute, Tehran University of Medical Sciences approved this survey.
Amongst 86 diarrhoeic patients, 42 had acute lymphoblastic leukaemia (ALL; 48.8%) as an underlying disease, followed by 26 chronic myelogenous leukaemia (CML; 30.2%) and 18 acute myeloid leukaemia (AML; 20.9%). According to the development of febrile periods mostly due to neutropenia, pulmonary and other relevant infections, the patients empirically received Imipenem (36%), Meropenem (24%), Ciprofloxacin (72%) and Cefixime (31%) as the most predominant antibacterial components. The history of antibiotic therapy starts from one day to one month before the diarrhoea happening.
Multiplex real-time PCR
Stool samples were spiked with internal control of RealStar® C. difficile PCR Kit (altona, Germany) before extraction of total DNA by QIAamp® DNA Stool Mini Kit (Qiagen, Germany). DNA from stool samples were investigated for the presence of tcdA and tcdB by RealStar® C. difficile PCR Kit using both LightCycler @96 (Roche, Germany) and Rotor-gene Q (Qiagen) machines simultaneously to ensure the accuracy of real-time PCR results. Multiplex real-time PCR assessment of the stool samples showed toxigenic CDI in nine patients (10.46%) amongst which seven (8.1%) were positive for both tcdA and tcdB and two (2.32%) were only positive for tcdB.
Toxigenic culture
Stool samples were collected; one spike (~1 g) of each specimen was treated by alcoholic shock for 1 h. Another one was transferred into the Clostridium difficile Brucella broth for 1 h under strict anaerobic conditions [Reference Cadnum4]. Treated specimens were inoculated onto the Cycloserine-Cefoxitin Fructose Agar, enriched by vitamin K1 (1 μg/ml) and hemin (5 μg/ml) and incubated in anaerobic condition for 2–5 days at 37 °C. Recovered colonies were confirmed by microbiological characteristics and the presence of C. difficile species-specific GDH gene (gluD) by PCR [Reference Origuen5]. The presence of tcdA and tcdB genes were explored by RealStar® C. difficile PCR Kit as described above. Toxigenic culture yielded 10 isolates, which were confirmed as C. difficile by positive PCR results for gluD. Real-time PCR assessment of the C. difficile isolates for tcdA and tcdB showed six (6.97%) having both tcdA and tcdB. Four other isolated C. difficile were non-toxigenic.
Overall results showed that the culture method missed three toxigenic C. difficile and so the calculated sensitivity and specificity of the culture method are 76.92% (95% CI 46.19–94.96%) and100.00% (95% CI 95.26–100.00%), respectively. Five patients who were diagnosed with the toxigenic C. difficile had ALL (55.5%), two had CML (22.2%) and the other two had AML (22.2%). Pairwise two-tailed regression showed no significant correlation between the antibiotics used nor leukaemia type and the CDI in subject patients.
Treatment
After the development of diarrhoea in patients who were receiving antibiotics, a single dose of Vancomycin and Metronidazole were administered orally (following stool sample collection) [Reference Bruminhent6]. Oral treatment with Metronidazole and Vancomycin was continued for the individuals for at least 7 days upon real-time PCR confirmation of the toxigenic CDI. CDI was successfully controlled and no mortality was recorded due to the toxigenic C. difficile during this study.
Discussion and conclusion
Leukaemic patients receiving allogeneic transplantation are amongst the most susceptible patient to all types of infections. Recent studies show the increasing rates of CDI from 5% to 20% and even rising higher to above 50% after the HSCT which may ruin the stem cell transplantation success [Reference Balsells7]. Allogeneic HSCT patients are exposed to many risk factors making them susceptible to the CDI such as broad-spectrum antibiotics, chemotherapy, immunosuppression and proton pump inhibitor medications.
Our study showed that 10.4% of patients with bone marrow transplantation were infected with the toxigenic C. difficile; the infection rate will be increased to 15.11% if four non-toxigenic C. difficile isolates are included. The history of antibiotic therapy for CDI patients was from one week to one month in the post-transplantation period. We did not detect or isolate C. difficile from any patient who received antibiotic therapy or being hospitalised less than a week and it is compatible with the known risk of prolonged antibiotic therapy as one of the most important predisposing factors for CDI. However, we could not correlate the antibiotics used in the subject patients with the incidence of CDI.
Toxins A and B – the C. difficile enterotoxin and cytotoxin, respectively – are the main pathogenicity factors of the organism. Toxigenic types of C. difficle cause the same clinical signs of illnesses ranging from mild diarrhoea to life-threatening inflammation of colon, toxic megacolon and deadly infectious colitis [Reference Chandrasekaran and Lacy8].
Detecting tcdA and tcdB directly in fesses leads to the diagnosis of pathogenic C. difficile rather than targeting gluD gene that only confirms the presence of C. difficile without distinguishing between pathogen and non-pathogen types. Timing of toxigenic culture and the isolation of C. difficile is unfavourable when the urgency of HSCT patient cases comes into account. However, in the current study, toxigenic culture results confirm the sensitivity and specificity of the multiplex-real-time PCR detection of the pathogen.
In addition, microbiological isolation and characterisation of the toxigenic C. difficile may fuel further investigations such as epidemiological tracing of the pathogen transmission and the infection control measurements. Antibiotic susceptibility testing of the C. difficile isolates also reported to be considered [Reference Rodriguez9].
Oral Metronidazole and Vancomycin were administered after collecting stool samples from HSCT patients with diarrhoea [Reference Bruminhent6]. This blind treatment continued if the results approve the infection with toxigenic C. difficile. Multiplex real-time PCR results were obtained on the same day (fewer than four hours) and helped the precise decision on the proper use of antibiotics. Although there are widely used serological methods for screening of the C. difficile toxins in stool samples, real-time PCR detection of the pathogen is significantly more sensitive, specific and rapid. It remarkably improves the proper management of critically infection-susceptible HSCT patients [Reference Collins and Auchtung10].
We assume that using real-time PCR for detection of the toxigenic C. difficile in leukaemic/HSCT patients with antibiotic-associated diarrhoea significantly increases the accuracy of diagnosis in a short time. This leads to efficient management and control of the disease and reduces mortality due to CDI.
Acknowledgements
This work was conducted in the Department of Bacteriology, Faculty of Medical Sciences at Tarbiat Modares University and in collaboration with Digestive Disease Research Institute, Tehran University of Medical Sciences.
Financial support
Tarbiat Modares University and Digestive Disease Research Institute, Tehran University of Medical Sciences as a collaborative project had funded this work.
Conflict of interest
The authors are agreeing with the content of the manuscript and declare there is no conflict of interest in all aspects of this work.
Data availability statement
The data that support the findings of this study are available from the corresponding author upon reasonable request.
Open access
