gms | German Medical Science

In 2010, the German Society for Hospital Hygiene (Deutsche Gesellschaft für Krankenhaushygiene, DGKH) launched a study to assess the prevalence of five emerging bacterial pathogens in volunteering hospitals to help to provide data on the epidemiology of carrier-ship and infections with these emerging nosocomial pathogens. The study was designed as a one-day point-prevalence study and conducted on the on the 10^th of February 2010.

For participating hospitals, the level of care, staffing with infection prevention personnel, availability of a MRSA-screening and microbiological support was assessed by a basic questionnaire and the prevalence of the five emerging bacterial pathogens: Methicillin-resistant Staphylococcus aureus (MRSA), extended-spectrum-lactamase-building (ESBL) Gram negative organisms, multiresistant Pseudomonas (MRP) and Acinetobacter species (MAB), Vancomycin-resistant Entoerococcus species (VRE) and toxin-building Clostridium difficile (CD) in intensive care, surgical and medical wards by three identical questionnaires (Table 1 [Tab. 1]).

Five tertiary and four secondary care hospitals participated in the study (Table 2 [Tab. 2]). Overall, 3411 patients were included. Questionnaires were mostly filled out by infection control nurses. None of the participating hospitals reported logistical problems.

Four tertiary care hospitals had an own infection control specialist and one was serviced by an external specialist. In contrast, one secondary care hospital had no infection control specialist and the other three had an external one.

All tertiary care hospitals had own infection control nurses: three hospitals had four, one three and one only one nurse. In secondary care hospitals, however, had only one, external, infection control nurse each (Table 2 [Tab. 2]). With one exception, the microbiological service was reported as “sufficient”.

Quality of MRSA screening regimes differed markedly between hospitals. In three tertiary hospitals, patients were screened in accordance with the recommendations by the Commission for Hospital Hygiene and Infection Prevention at the Robert Koch Institute (KRINKO) [1] if they had two or more risk factors. In two hospitals, patients with one or more risk factor and all patients admitted to intensive care wards were screened as established in the University Medicine Greifswald (“Greifswald model”) [2], [3]. In one tertiary care hospital an ESBL-screening for urological patient is established, too.

MRSA was the most frequently reported pathogen in all participating hospitals (Table 3 [Tab. 3], Table 4 [Tab. 4], Table 5 [Tab. 5], Table 6 [Tab. 6], Table 7 [Tab. 7], Table 8 [Tab. 8], Table 9 [Tab. 9], Table 10 [Tab. 10], Table 11 [Tab. 11]), followed by CD and ESBL. No patient with CD was reported to require intensive care.

Seen over all departments and hospitals, MRSA and CD had a higher prevalence in secondary care hospitals in comparison to tertiary care. ESBL prevalence was comparable for both groups and MRP and VRE were more frequently reported in tertiary care hospitals. MAB were reported in tertiary care hospitals only (Table 12 [Tab. 12], Table 13 [Tab. 13], Table 14 [Tab. 14], Table 15 [Tab. 15]).

Some hospitals additionally provided refined data that allow to between infection and colonisation (Table 16 [Tab. 16]).

Data on the epidemiology of emerging bacterial pathogens with significant impact on hospital epidemiology are still sparse in Germany. Voluntary prevalence studies as this one initiated by the DGKH are an attempt to improve the epidemiological knowledge in this field. While this type of study has several limitations, as it gives only a momentary image of the situation and the data can be compared to data gathered with the same method only, it still is an inexpensive, convenient and feasible way to generate valuable data and sensitize health care workers for the situation on their wards.

The overall prevalence of MRSA, the post frequently reported pathogen in this study, was 2.2%. While not directly comparable to other studies, this seems to be in the range of other surveys as from the county of Höxter in 2008 (3.4%) [4], the EUREGIO MRSA-net Twente/Münsterland in 2006 (1.6% for the German part) [5], the city of Essen in 2009 (2% in hospitals) [6] and the Saarland 2010 (2.18%) [7].

In the four secondary care hospitals the MRSA prevalence was 3.7% and therefore much higher as in the tertiary care hospitals (1.74%), which was unexpected.

Remarkably, all five studies report much higher prevalences than one would expect from the data provided by the German KISS (Krankenhaus-Infektions-Surveillance-System), that reported a mean prevalence of 0.98% for all participating hospitals (n=268) only, 0.96% for hospitals >600 beds and 1.00% for hospitals <600 beds in 2010 (Figure 1 [Fig. 1]) [8].

Our initial consideration was, that this may be due to the fact that our prevalence study was conducted on surgical, medical and intensive care wards only, and that therefore intensive care (which has a relatively high MRSA-prevalence) is overweighed in comparison to the KISS data that includes all wards. However, the MRSA prevalence on intensive care units in this study is also much higher than the one reported by the ITS-KISS with 8.39% versus 1.5%, respectively (Figure 2 [Fig. 2]) [9].

Unfortunately, the MRSA-KISS system does not support individual statistics for other medical specialities. Whether the obvious differences between the results from all five prevalence studies and the KISS is by chance or indicates a systematic underreporting in the KISS remains unclear.

As the tertiary care hospital number 5 provided the whole-year for MRSA-prevalence additionally, the point-prevalence and the overall prevalence for 2010 can be compared for this hospital. With 2.23%, the point-prevalence was higher than the mean prevalence for 2010 (1.28%), which is, again, much higher than to be expected from the MRSA-KISS. Just as for the relation between the prevalence studies and the KISS-data discussed above, this could be caused by overweighting intensive care units (see above). Still, the prevalence data from the intensive care units is also much higher (3.58%) than to be expected from ITS-KISS (1.5%), underlying the need to validate the KISS results by external studies [9], [10].

Compared to MRSA, the prevalence of the other included pathogens was much lower, but they were still frequently reported especially on ICUs. Table 17 [Tab. 17] compares the prevalences found in participating ICUs between levels of care and data from the ITS-KISS, if available [9]. Again, prevalences for all levels of care and all pathogens were much higher as to be expected from the KISS. For CD, too, the prevalence found in our study was more than twice as high (1.08%) as expected based on the CDAD-KISS (0.46%) [11].

As previously reported by other authors, our study shows that prevalences of emerging bacterial pathogens differ markedly between regions, departments and hospitals. While one-day point-prevalences have to be interpreted with caution, the prevalences found fit well to other prevalence studies from the last years. Remarkably, all of those studies have found much higher prevalences than to be expected from the data of the KISS. Further studies are needed to show whether this was by chance alone or indicates a systematic underreporting in the KISS.

Voluntary point-prevalence studies from routine data have been shown to be an inexpensive way to generate valuable data. By such initiatives, scientific societies as the DGKH can take part in collecting valuable epidemiological data. Future studies, using a fixed day and protocol, could be an interesting tool to describe the epidemiology of emerging bacterial pathogens and verify data from other sources.

The authors declare that they have no competing interests.