Further exploration revealed that in March 2007, a severe case of a C. difficile infection had already occurred in this hospital. The strain isolated from this patient in April could be further characterised as PCR ribotype 027, toxinotype III, PFGE NAP1, and was shown to be PCR-positive for the binary toxin and an 18 bp deletion in the tcdC regulatory gene. The strain demonstrated resistance to erythromycin and moxifloxacin, among other antimicrobials, but was susceptible to clindamycin, thereby exhibiting a similar profile to that seen for the highly virulent strains that have recently caused outbreaks in North America and several European countries [1,2,3].

Based on these findings, a retrospective case search was initiated, including a systematic review of patient details, history, and known risk factors for CDAD. In addition, a prospective surveillance system was implemented for all hospitals in Trier. Stool samples of all patients with possible CDAD are now being systematically tested for CDAD, as well as cultured to isolate C. difficile. Isolates from positive samples are being ribotyped. The investigations are still ongoing.

As of 5 November 2007, the situation was as follows: since January 2007, eight confirmed and 28 probable cases of C. difficile PCR ribotype 027 (definitions of probable and confirmed cases can be found in the box) were identified in six hospitals in the region of Trier (Figure). The cases include 16 male and 20 female patients. The mean age was 74 years. Six patients died due to a cause attributable directly or indirectly to the CDAD. Two small clusters comprising a total of six cases were identified in one hospital. We have not yet been able to establish linkage between the other cases. An additional infection with C. difficile PCR ribotype 027 was identified in an asymptomatic carrier.

A neighbouring administrative district in which surveillance has not yet been established, reported another case of CDAD due to PCR ribotype 027.

Agreement was reached upon a nationwide notification of severe cases of CDAD between chief representatives of the public health authorities of the federal states and the consulting experts from the Robert Koch Institute.

As isolates are seldom routinely cultured or typed in Germany, and samples are only rarely sent to the national consultant laboratory or the few specialised laboratories, it is probable that there have been previous cases in Germany before the case reported above (see also [4]). However, C. difficile PCR ribotype 027 with the characteristics of the highly virulent strains described above was not found in a set of approximately 900 isolates collected between January 2000 and September 2006 and sent to the consultant laboratory for gastrointestinal infections, Freiburg, Germany [5]. The results of the investigations performed so far suggest that C. difficile PCR ribotype 027 may already be endemic in Germany, at least in the Trier region. Further investigations to determine the extent to which this new strain and possibly other highly virulent strains have spread in Germany are ongoing.

Acknowledgements
We would like to thank E.J. Kuijper, National Reference Laboratory for Clostridium difficile, Leiden, The Netherlands for providing us with detailed information about the first confirmed strain of PCR ribotype 027 in Germany and its laboratory processing, as well as C. v. Eichel-Streiber, Consultant Laboratory for Clostridium difficile, Mainz, Germany, Prof. M Mielke, Robert-Koch Institut, Wernigerode, Germany, Prof. W. Witte, Robert-Koch Institut, Wernigerode, Germany, and Prof. M. Kist, Consultant laboratory for gastrointestinal infections, Freiburg, Germany for their collaboration in the ongoing investigations.