Hypervirulent epidemic strains of Clostridium difficile belonging to ribotype 027 (toxinotype III, toxin A+B+binary+) have been reported to cause important outbreaks of severe infection in the United States and Canada since 2001.1,2 Since 2003, C. difficile ribotype 027 (Cd027) has been detected in major outbreaks or sporadic cases in several European countries.3 In Spain, data describing the molecular epidemiology of C. difficile are scarce, and the importance of Cd027 is still unknown.4,5

In this report, we describe the first Spanish autochthonous case of Cd027 infection and present an additional 3 cases of diarrhea caused by Cd027 strains isolated at our institution.

Since January 2007, our laboratory has performed active surveillance of Cd027 based on multiplex polymerase chain reaction (PCR) to detect C. difficile toxin genes in all strains of C. difficile isolated in our hospital. tdcA conserved fragment, tcdA deleted fragment, tcdB, cdtA, cdtB, and 16SrRNA (internal control) were amplified by PCR following a method adapted from other authors.6,7

As Cd027 is characterized by binary toxin production, all binary toxin-positive strains were ribotyped.8 Strains belonging to ribotype 027 were further characterized by toxinotyping,9 sequencing of tcdC,10 and multilocus-variable-number of tandem repeat analysis (MLVA).11 MLVA patterns were interpreted using summed tandem repeat differences (STRD).11 Susceptibility to erythromycin, clindamycin, moxifloxacin, rifampin, imipenem, metronidazole, and vancomycin was assessed using the E-test in Brucella agar.

By April 1st, 2012 we had studied 3666 strains of C. difficile of which 444 (12%) had binary toxin genes. A total of 360 strains (10%) belonged to toxinotype V ribotype 078/126, and only 8 (0.21%) belonged to toxinotype III ribotype 027 (Fig. 1, ribotyping pattern compared to epidemic strains kindly provided by E.J. Kuijper). All Cd027 strains had a single 1-bp deletion at position 117 and an 18-bp deletion in the tcdC gene that is characteristic of the epidemic C. difficile strain ribotype 027. These strains were isolated from 4 patients (Table 1).

Fig. 1.

Similarity dendrogram of ribotyping of the Cd027 strains.

(0,12MB).

All our Cd027 isolates were susceptible to metronidazole. The isolates for the first 2 patients were highly resistant to erythromycin, clindamycin, moxifloxacin, rifampin, and imipenem, although the Cd027 isolates of the last 2 patients were not so (Table 1). The vancomycin MIC varied from 0.75μg/ml to 2μg/ml.

MLVA typing of Cd027 isolates revealed 4 different patterns (Table 2). The strains from the first episode of patient 1 and the strain from patient 2 were grouped in the same clonal cluster (STRD≤2). Strains from patients 3 and 4 were genetically unrelated to each other and to strains from patients 1 and 2 (STRD≥10). Surprisingly, MLVA typing demonstrated that the strains from the 2 episodes of patient 1 were different and did not belong to the same clone (STRD≥10), probably reflecting sequential isolation of different Cd027 strains from an initial polyclonal infection acquired in the hospital of origin in the UK.

To our knowledge, we report the first autochthonous case of C. difficile infection caused by Cd027 in Spain. The patient had no previous contact with health care institutions in Spain or abroad; therefore, we were unable to track the origin of the strain and we ignore if this patient had contact with people (households, relatives, etc.) from environments where Cd027 is frequent.14 The review of the other 3 episodes shows that Cd027 mainly appears as imported cases. The first 2 cases of Cd027 in our hospital were detected in 2007. The first was considered to be imported from the UK; the second involved horizontal transmission to a laboratory technician handling the first patient's stool samples.13

Between 2000 and 2006, rates of C. difficile infection increased markedly in the USA and Canada,15,16 mainly because of the epidemic Cd027 strain.1 Although some outbreaks of Cd027 have been described in Europe,17,18 a surveillance study performed in 34 European nations in 2008 showed a mean incidence of nosocomial C. difficile infection of 4.1 cases per 10,000 patient-days (range 0.0–36.3); only 5% were caused by Cd027.5 Similarly, a pan-European Cd027 survey revealed a relatively low prevalence in Europe that varied from country to country.19 Cd027 outbreaks in Europe were contained thanks to awareness of the American experience and the introduction of strict control measures in several European countries.

It is difficult to know the real importance of Cd027 in Spain. Our results indicate that Cd027 infection could be distinctly uncommon in Spain. Although, it is necessary to consider that not all clinical microbiology laboratories in Spain currently culture C. difficile, and diagnosis of C. difficile infection is based mainly on the results of enzyme immunoassay as demonstrated in a recent paper by our group.20 Besides, Spain does not have a national surveillance program to investigate C. difficile infection or a reference laboratory where hospitals could send C. difficile isolates for further characterization. These limitations would determine that Cd027 could go undetected mainly considering that only a few hospitals perform molecular characterization of C. difficile.

These limitations are overcome in a recently reported prospective nationwide diagnostic study, in which confirmatory cultures and molecular characterization of C. difficile isolates were performed on all diarrheic stools arriving at 118 Spanish microbiology laboratories. An incidence of 3.8 cases of C. difficile infection per 10,000 patient-days was reported without isolation of Cd027.4 In other studies recently conducted in Spanish hospitals, Cd027 was also not detected.21,22 After intensive searching at our center, we detected a very low incidence of Cd027 strains. The reasons for this finding are both unclear and unexpected, since, annually, Spain receives millions of tourists from all over the world.

The incorporation in the routine diagnosis of C. difficile infection of easy-to-use systems such as GeneExpert C. diff® (Cepheid, CA, USA), which detects toxigenic C. difficile and presumptive Cd027, could improve detection of Cd027 strains in Spain, although laboratories should culture C. difficile in order to have a stock of isolates for epidemiological studies.

In conclusion, although Cd027 does not seem to be a major problem in Spain, it is important to recommend that clinical microbiology laboratories perform C. difficile culture, at least from toxin-positive stools or in severe infections, to enable epidemiological studies to know the real situation of Cd027 and other epidemic strains.

This project was partially financed by a grant (project number PS09-02389) from the FIS (Fondo de Investigaciones Sanitarias). Fragment analysis to obtain MLVA patterns and sequencing of tcdC were performed in a 3130×1 Genetic Analyzer that was financed in part by grants from the FIS (IF01-3624 and IF08-36173).

The authors declare no conflict of interest.