HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via CrossRef Citing Articles via Google Scholar Google Scholar Articles by Rotimi, V. O. Articles by Duerden, B. I. Search for Related Content PubMed PubMed Citation Articles by Rotimi, V. O. Articles by Duerden, B. I. Agricola Articles by Rotimi, V. O. Articles by Duerden, B. I.

Prevalent PCR ribotypes of clinical and environmental strains of Clostridium difficile isolated from intensive-therapy unit patients in Kuwait

¹Department of Microbiology, Faculty of Medicine, Kuwait University/Mubarak Al-Kabeer Hospital, Kuwait ²PHLS Anaerobe Reference Unit, Department of Medical Microbiology and PHLS, University of Wales College of Medicine, Cardiff, UK ³Department of Microbiology, Amiri Hospital, Kuwait

Correspondence V. O. Rotimi Vincent{at}hsc.kuniv.edu.kw

Received February 3, 2003
Accepted April 4, 2003

	Abstract

TOP Abstract INTRODUCTION METHODS RESULTS DISCUSSION Acknowledgements REFERENCES

 
Ninety-five isolates of Clostridium difficile from symptomatic and asymptomatic patients and 18 from their environment in the intensive-therapy units (ITUs) of four teaching hospitals in Kuwait were typed by PCR amplification of rRNA intergenic spacer regions (PCR ribotyping). A total of 32 different ribotypes was detected among the clinical isolates. The predominant ribotypes from the clinical isolates were types 097 and 078, which accounted for 40 % of all isolates in the ITUs in Kuwait. Ribotypes 097 (toxigenic), 078 (toxigenic) and 039 (non-toxigenic) were three distinct clones that were circulating in all four hospitals. Ribotypes 097, 078 and 076 (i.e. 50 % of isolates from symptomatic patients) were the predominant isolates associated with C. difficile-associated disease (CDAD). The environmental isolates belonged to a diverse range of ribotypes, with no particular types common to all the hospitals. Ribotype 078 was found only in the patient environment in Mubarak hospital, while ribotype 097 was restricted to Amiri hospital. The hospital environment occupied by symptomatic as well as symptom-free patients was contaminated with C. difficile. Eight new strains that did not match any in the PCR ribotype library established at the PHLS Anaerobe Reference Unit, Cardiff, UK, were assigned ribotypes 105, 125, 128, 129, 131, 134, 140 and 141. These findings show that the isolates associated with CDAD in Kuwait are different from those found in the UK and some other European countries.

	INTRODUCTION

TOP Abstract INTRODUCTION METHODS RESULTS DISCUSSION Acknowledgements REFERENCES

 
Clostridium difficile is an anaerobic, Gram-positive, spore-forming bacillus. It is often associated with a spectrum of diseases, referred as C. difficile-associated disease (CDAD), which manifest as self-limiting antibiotic-associated diarrhoea (AAD), antibiotic-associated colitis (AAC) and pseudomembranous colitis (PMC) with toxic megacolon and possible gut perforation. Rarely, it can present as extraintestinal infections, such as arthritis, osteomyelitis, soft-tissue infection and bacteraemia (Levett, 1986). CDAD is an important clinical problem because it is often acquired by hospitalized patients, and C. difficile is associated with outbreaks of diarrhoea and colitis in hospitalized adults receiving antibiotics (Bartlett et al., 1978; Djuretic et al., 1999).

In spite of major efforts to control the spread of CDAD in hospitals and nursing homes, this organism has remained a major problem worldwide, and it continues to be responsible for endemic and epidemic nosocomial diarrhoea (McFarland et al., 1989; Johnson et al., 1990; Barbut et al., 1996). C. difficile, or its toxins, has been identified in 8–10 % of cases of nosocomial diarrhoea, while other common bacterial enteric pathogens, Salmonella, Shigella and Campylobacter spp., are rarely isolated (Fan et al., 1993; Rohner et al., 1997).

It appears that the most important sources of C. difficile in a hospital setting are symptomatic patients and asymptomatic carriers who are the main reservoirs of C. difficile in the hospital. The environment of these patients is also an important source. There is evidence that environmental contamination in rooms of patients with diarrhoea is substantially greater than in rooms with asymptomatic patients (49 vs 29 %) (McFarland et al., 1989). In addition, rooms currently occupied by C. difficile-negative patients can, in spite of routine cleaning, be contaminated with C. difficile spores that can survive for several months in the hospital environment. Person-to-person transmission on hospital wards, especially geriatric wards, as well as environmental contamination and carriage on the hands of hospital workers have been documented (Kim et al., 1981; Malamou-Ladas et al., 1983; Savage & Alford, 1983; McFarland et al., 1989). Transmission occurs mainly via the faecal oral route and direct contact with contaminated surfaces (Barbut & Petit, 2001).

Epidemiological studies of C. difficile strains isolated from different environmental and clinical sources involve detailed comparison of the different isolates from all sites. Several typing schemes have been developed to determine the relatedness of strains of C. difficile associated with infections. However, of these, PCR ribotyping offers several advantages over the other methods because it is highly discriminative, reproducible and relatively rapid and easy to perform (O'Neill et al., 1996; Stubbs et al., 1999).

Our main aim was to study the epidemiology of C. difficile isolates from patients admitted into the intensive-therapy units (ITUs) of the four main teaching hospitals in Kuwait by PCR ribotyping.

	METHODS

TOP Abstract INTRODUCTION METHODS RESULTS DISCUSSION Acknowledgements REFERENCES

 
Bacterial isolates from patients.

Stool samples were collected from all patients admitted to the ITUs of Mubarak Al-Kabeer hospital (ITU-1), Ibn Sina Burn Unit (ITU-2), haematology wards of Kuwait Cancer Control Centre (ITU-3) and Amiri hospital (ITU-4) who had stayed in the units for a minimum period of 3 days. The study was carried out over a period of 1 year (February 2001–January 2002). Freshly passed stool (rectal swabs, in Amies transport medium, if collection of stool was not feasible) was taken on the day of admission and weekly thereafter until the patient was discharged or developed diarrhoea secondary to C. difficile infection/colonization.

Environmental and other samples.

A sterile swab, pre-moistened with sterile normal saline, was wiped over selected surfaces in the units, e.g. bed sheets, mattress, bed edges, bed ledges, surfaces of the side table next to the patients, suction regulator, oxygen regulator, ventilator surfaces, IV stand and the floor under the bed. Hands of doctors, nurses and physiotherapists who had had contact with the positive cases were cultured, by contact agar plates, for evidence of C. difficile contamination.

Inoculation, isolation and identification.

The stool and environmental samples were inoculated into Robertson cooked meat (RCM) medium containing 25 ml fastidious anaerobe broth (FAB; Lab M) and incubated anaerobically for 48 h at 37 °C. Five-hundred microlitres of the FAB from the RCM was heated for 10 min at 80 °C. Next, cycloserine-cefoxitin egg yolk agar (CCEYA; Oxoid) and cycloserine-cefoxitin fructose agar (CCFA; Oxoid) plates were inoculated with one loop of the heated broth and incubated anaerobically for 48 h in an anaerobic chamber (H₂ 10 %, CO₂ 10 %, N₂ 80 %). Isolates that were Gram-positive bacilli with characteristic horse-dung smell and fluoresced yellowish-green under long-wave UV light (365 nm) were selected and their identity was confirmed as C. difficile by API 20A (bioMérieux).

Toxin detection.

Single colonies were subcultured on pre-reduced Columbia agar base (Oxoid) supplemented with 5 % horse blood, vitamin K and haemin and incubated at 37 °C under anaerobic conditions for toxin A detection. Toxin A was detected by ELISA (TOX-A) kits (Tech Lab); the procedure was carried out according to the manufacturer's instructions. Toxin B was detected by cytotoxicity assay on Vero cells. Production of cytopathic effects by filtered supernate of C. difficile RCM broth culture on Vero cells indicated toxin B production. Toxins A and B were detected in the stool by the TOX-A/B kits (Tech Lab).

PCR ribotyping.

All isolates were typed by the PCR ribotyping method described by O'Neill et al. (1996). Briefly, after obtaining a pure culture, a single colony was subcultured on fastidious anaerobe agar (FAA; Lab M) supplemented with 6 % horse blood and incubated for 24 h at 37 °C. DNA was extracted from a suspension of 10 colonies in 100 µl 5 % Chelex 100 (Bio-Rad) by heating at 100 °C for 10 min. Cell debris was removed from the suspension by centrifugation for 10 min at 17 000 g. The supernate obtained was then used as the DNA template. Primers P3 (5'-CTGGGGTGAAGTCGTAACAAGG) and P4 (5'-GCGCCCTTTGT AGCTTTGACC) were used for the PCR amplification. The reaction mixture (final volume, 100 µl) contained 1.5 mM MgCl₂, 10 mM Tris/HCl (pH 9), 50 mM KCl, 0.1 % Triton X-100, 2.5 U Taq polymerase, 200 mM of each dNTP, 50 pmol of each primer and 10 µl DNA template. The PCR programme was 35 cycles of denaturation at 95 °C for 1 min, annealing at 56 °C for 1 min and extension at 72 °C for 2 min. A negative control was included in each run. Amplification products were concentrated to a final volume of 25 µl by heating at 75 °C for 90 min before electrophoresis at 200 V, 100 A in Metaphore agarose gel (FMC) for 3 h at room temperature. DNA fragments were then visualized by staining the gel for 20 min in 0.5 % ethidium bromide. Gel images were analysed with GelCompar image analysis software (version 4.0; Applied Maths). Our results were then compared with the library of PCR ribotypes already established at the PHLS Anaerobe Reference Unit, Cardiff, UK.

	RESULTS

TOP Abstract INTRODUCTION METHODS RESULTS DISCUSSION Acknowledgements REFERENCES

 
C. difficile was isolated from 95 of 922 patients (430 in ITU-1, 63 in ITU-2, 88 in ITU-3 and 341 in ITU-4) screened during this period. Thirty (31.6 %) of these 95 patients were symptomatic with CDAD; two patients had PMC, four had AAC and 24 had AAD. The overall incidence of C. difficile-positive patients was 10.3 %, with incidence in each hospital being 10.5 % for ITU-1, 6.3 % in ITU-2, 18 % in ITU-3 and 9.6 % in ITU-4. The organism was isolated from 18 of 380 environmental samples; six from ITU-1, three from ITU-3 and nine from ITU-4.

The distribution of PCR ribotypes among the clinical and environmental isolates is shown in Tables 1 and 2, and the banding of representative ribotypes is demonstrated in Fig. 1. Of the 95 isolates, 72 (75.8 %) were toxin-producers, while 23 (24.2 %) were non-toxigenic. A total of 28 (38.9 %) of the 72 toxigenic and two (8.7 %) of the 23 non-toxigenic isolates were associated with diarrhoea. As shown in Table 1, 32 distinct, genotypically different ribotypes were identified among the 95 clinical isolates. Of these, three distinct clones were detected in all four hospitals, ribotypes 097 (toxigenic), 078 (toxigenic) and 039 (non-toxigenic). The remaining 58 isolates belonged to a set of diverse ribotypes. In ITU-1, the ribotypes associated with diarrhoea in the 18 symptomatic patients were ribotypes 078 (4, 22 %), 076 (3, 17 %), 012 (2, 11 %), 056 (2, 11 %), 094 (2, 11 %), 097 (2, 11 %), 105 (2, 11 %) and 046 (1, 6 %). The five isolates associated with diarrhoea in ITU-3 represented ribotypes 077 (2, 40 %), 014 (1, 20 %), 097 (1, 20 %) and 113 (1, 20 %), while the seven associated with diarrhoea in ITU-4 represented ribotypes 017 (1, 14.3 %), 039 (1, 14.3 %) and 097 (5, 71.4 %).

View larger version (123K): [in this window] [in a new window]   Fig. 1. Representative PCR ribotype profiles of C. difficile isolated from patients in Kuwait. Lanes: 1, 6, 11 and 14, 100 bp ladder; 2, ribotype 029; 3, 039; 4, 097; 5, 039; 7, 039; 8, 039; 9, 078; 10, 078; 12, 097; 13, 051.

All symptomatic patients with toxigenic ribotypes had detectable toxins in their stools; two patients with non-toxigenic ribotypes 039 and 113 had no detectable level of toxin in their stools. Two PMC cases in ITU-1 were infected with ribotype 076. Single AAC cases in ITU-1 and ITU-3 were infected with ribotypes 046 and 097. Two AAC cases in ITU-4 were infected with ribotype 097.

As demonstrated in Table 2, the 18 environmental isolates were assigned to nine genotypically distinct ribotypes (six toxigenic and three non-toxigenic); the predominant types were 010 (3), 078 (3) and 097 (4). Ribotypes 078 and 097 were isolated primarily from the environment of symptomatic patients infected with the same ribotypes in ITU-1 and ITU-4, respectively. Ribotype 010 was isolated from the environment of asymptomatic patients in ITU-1. Ribotype 105, associated with diarrhoea in ITU-1, was absent in its environment but present in the environment of ITU-3 and ITU-4, where it did not contribute to CDAD. Ribotypes 001 and 144 were present in the environment of ITU-4, but were not isolated from any patient in the unit. Similarly, ribotypes 120 and 125 were isolated from the floor of ITU-3, but were not isolated from the patients in the unit.

No C. difficile was isolated from the hands of healthcare providers.

	DISCUSSION

TOP Abstract INTRODUCTION METHODS RESULTS DISCUSSION Acknowledgements REFERENCES

 
In spite of the growing number of studies devoted to CDAD in Western countries, studies on CDAD in the Middle East are lacking, especially in Kuwait, where information on the incidence of C. difficile carriage and CDAD is almost non-existent. This is partly as a result of inertia in anaerobic bacteriology prompted, until recently, by lack of expertise, technology and facilities for culturing anaerobic pathogens. We have, in the recent past, encountered a relatively high proportion of asymptomatic carriers of C. difficile in our hospitals, and the interpretation of positive cultures was equivocal. C. difficile or its toxin has been reported in one study from nearby Saudi Arabia (Akhter et al., 1994), in Turkey (Soyletir et al., 1996) and in a few reports from Israel (Rudensky et al., 1993; Rivlin et al., 1998; Boaz et al., 2000). However, this is the first report on PCR ribotyping of C. difficile strains isolated from symptomatic and asymptomatic patients in the Middle East.

The data generated from the present study showed that all 95 and 18 C. difficile isolates respectively originating from patients and their environment were typable by the PCR ribotyping method, and they revealed some interesting epidemiological findings. Secondly, none of the patients admitted to the ITUs carried C. difficile into the units on admission. Thus, an incidence rate of hospital-acquired C. difficile infection/colonization of 10.3 % was established for the four hospital ITUs in Kuwait, ranging from 6.3 to 18 % in different hospitals. Interestingly, the 95 C. difficile culture-positive patients harboured 32 different, highly diverse PCR ribotypes. Three different DNA clones (PCR ribotypes 097, 078 and 039) were detected among patient isolates in all hospitals. The other isolates were assigned to 29 distinct and diverse types. Ribotype 097 was the single most prevalent type and was responsible for about 27 % of CDAD, while ribotype 078 was responsible for about 13 % of CDAD. Thus, ribotypes 097 and 078 were responsible for over one-third of the cases of CDAD seen. From an epidemiological point of view, this is an interesting finding, in that the dominant ribotypes causing diarrhoea in Kuwait are completely different from those seen in Europe, particularly the UK (Stubbs et al., 1999), Hungary (Urban et al., 2001) and Poland (Martirosian et al., 1995). Fifty-five per cent of C. difficile infections seen in UK hospitals are caused by ribotype 001 (Stubbs et al., 1999), while ribotype 087 accounted for 39 % of all isolates in Hungary (Urban et al., 2001). In the Polish study, all the environmental isolates and 11 of 31 neonatal isolates belonged to ribotype 001.

It is noteworthy that one of our patients in ITU-4 who had AAD was infected with ribotype 017, which is toxin-variable (toxin A-negative, B-positive). He gave a history of travel to Thailand and he was hospitalized there for unrelated illness. An outbreak of toxin A-negative, B-positive C. difficile associated with diarrhoea has been reported in a Canadian tertiary-care hospital (Al-Barrak et al., 1999). This is not our experience in Kuwait so far, as the single isolate was restricted to only one hospital and one patient. Why some patients have more marked symptoms than others and some strains are more associated with outbreaks of CDAD in different countries is unclear. Systemic symptoms are caused mainly by toxin-mediated inflammatory mediators released in the colon (Dallal et al., 2002). Thus, it is conceivable that the ability of the host to mount an effective antibody-mediated response to the C. difficile toxin plays a major role in this regard. However, the bacterial factors involved in the geographical differences seen in the two locations are worthy of further investigation.

The environmental strains were heterogeneous in each hospital. The 18 isolates were assigned to nine different ribotypes. The environment of two patients each in ITU-1 and ITU-4 was contaminated by the same ribotypes as found in the patients’ clinical samples, confirming that the patient's environment is a potential source of C. difficile and that cross-contamination may play an important role in the acquisition of nosocomial CDAD (Cohen et al., 2000). In some cases, there was no correlation between the environmental and patient isolates. For instance, the environmental isolates in ITU-3 and some in ITU-4 differed from those found in the patients. Our finding is not an isolated one. Earlier, Cohen et al. (1997) found none of the environmental C. difficile genotypes among isolates from patients, and Simor et al. (1993) found no cross-transmission in their institution in a survey of C. difficile infection in a long-term care facility.

In conclusion, the prevalent PCR ribotypes of C. difficile strains circulating in Kuwait hospitals are different from those found in many hospitals in Europe and are also different from those associated with CDAD in many patients. Further work is needed to elucidate the factors responsible for the geographical differences and the ability of one strain to cause more systemic symptoms than another.

	Acknowledgements

TOP Abstract INTRODUCTION METHODS RESULTS DISCUSSION Acknowledgements REFERENCES

 
This work was supported by Kuwait University research grant no. MI 110, which is gratefully acknowledged. We are grateful to Ms Tina Verghese, Ms May Shahin and Ms Fatima Khodakhast for their excellent technical support. Dr Micaela Gal is thanked for her technical assistance. We also thank our colleagues and the nursing staff in the four ITUs who helped us in the collection of the samples.

	REFERENCES

TOP Abstract INTRODUCTION METHODS RESULTS DISCUSSION Acknowledgements REFERENCES