Antimicrobial usage in German acute care hospitals: results of the third national point prevalence survey and comparison with previous national point prevalence surveys

Antimicrobial usage in German acute care hospitals: results of the third national point... Abstract Objectives Previous point prevalence surveys (PPSs) revealed the potential for improving antimicrobial usage (AU) in German acute care hospitals. Data from the 2016 German national PPS on healthcare-associated infections and AU were used to evaluate efforts in antimicrobial stewardship (AMS). Methods A national PPS in Germany was organized by the German National Reference Centre for Surveillance of Nosocomial Infections in 2016 as part of the European PPS initiated by the ECDC. The data were collected in May and June 2016. Results were compared with data from the PPS 2011. Results A total of 218 hospitals with 64 412 observed patients participated in the PPS 2016. The prevalence of patients with AU was 25.9% (95% CI 25.6%–26.3%). No significant increase or decrease in AU prevalence was revealed in the group of all participating hospitals. Prolonged surgical prophylaxis was found to be common (56.1% of all surgical prophylaxes on the prevalence day), but significantly less prevalent than in 2011 (P < 0.01). The most frequently administered antimicrobial groups were penicillins plus β-lactamase inhibitors (BLIs) (23.2%), second-generation cephalosporins (12.9%) and fluoroquinolones (11.3%). Significantly more penicillins plus BLIs and fewer second-generation cephalosporins and fluoroquinolones were used in 2016. Overall, an increase in the consumption of broad-spectrum antimicrobials was noted. For 68.7% of all administered antimicrobials, the indication was documented in the patient notes. Conclusions The current data reaffirm the points of improvement that previous data identified and reveal that recent efforts in AMS in German hospitals require further intensification. Introduction Five years after the second national point prevalence survey (PPS) of healthcare-associated infections (HAI) and antimicrobial usage (AU) in Germany in 2011, a third national PPS was conducted in 2016 following a similar methodology. The first such national survey was conducted in 1994.1 Results of the 2011 PPS were widely published.2–4 The prevalence of patients with AU was calculated to be 25.5% and various points for improvement such as insufficient documentation, extensive use of broad-spectrum antimicrobials and high frequency of prolonged surgical prophylaxis (SP) were described.2 The 2011 and the 2016 surveys were embedded in the European PPSs that were initiated by the ECDC. Similarly to the 2011/12 PPS, the ECDC organized the 2016/17 PPS in four periods. Germany was among the first countries to conduct the PPS in May and June 2016. European reference data from the 2016/17 ECDC PPS are, therefore, currently not available. In 2011/12, 29 EU countries and Croatia participated in the ECDC PPS. The prevalence of patients who received at least one antimicrobial on the day of the survey was found to be 35.0%.5 Efforts in promoting antimicrobial stewardship (AMS) in Germany have been intensified in recent years. For the year 2017, it is estimated that over 1000 AMS specialists will have been educated.6 Since results concerning data on HAI have been and will be published elsewhere,7 this article will focus solely on the findings regarding AU. Methods Protocol As in the 2011/12 survey, the ECDC provided a standardized study protocol that specified uniform definitions and a uniform methodology.8 The German National Reference Centre for Surveillance of Nosocomial Infections (NRZ) was responsible for organizing the survey in Germany. The ECDC protocol version 5.1 was implemented and materials were translated accordingly.9 The ECDC offered two different study protocols: a standard protocol and a light protocol. The most relevant difference was that patient data in the light protocol were collected only for patients who had at least one active HAI and/or who had received at least one antimicrobial on the day of the survey, whereas in the standard protocol, patient data were to be recorded for every patient. The light protocol version was applied in Germany in order to reduce the workload for the participating hospitals and to enable participation by a higher number of hospitals. Representative sample of hospitals The ECDC required all participating countries to invite a representative sample of hospitals to take part in the survey. Based on population and the structure of the healthcare system, 49 hospitals were calculated as the sample size for Germany. These were randomly selected by the number of beds from the German hospital register 2013.10 Only hospitals listed as acute care hospitals were included. They were invited to participate in the PPS on a voluntary basis. For every hospital selected, two other hospitals of comparable size were chosen as possible substitutes in case a selected hospital decided against participation. Moreover, all 1462 (as of the first quarter of 2016) hospitals participating in the German Krankenhaus-Infektions-Surveillance-System (KISS) were invited to participate. Data validation Ten of the 49 hospitals, which constituted the representative sample, were visited by two members of the NRZ, who represented the gold standard, to perform a validation study of the data collection in order to assess the sensitivity and specificity of the local PPS teams. Data collection Data on HAI and AU as well as structural and process indicators were collected. Local staff of the participating hospitals executed the data collection. To ensure application of consistent methodology and definitions, at least one member of the local data collection team per hospital was trained in one of the seven 1 day courses that were organized by the NRZ prior to the start of the PPS. The data were collected in the months of May and June 2016. The local PPS teams visited the wards of the hospital in succession and reviewed the charts of all patients who were present in the ward and who had been admitted to the ward before 08:00 on the day of survey. Patient data documented in the patient charts as well as further information available from the ward staff upon request were evaluated. Data from outpatients were not included in the survey. For every patient receiving at least one antimicrobial, the following data were collected: the WHO Anatomical Therapeutic Chemical (ATC5) code of the antimicrobial,11 the route of application, the indication for AU (therapy versus prophylaxis versus other versus unknown), the site of infection for therapeutic indication and whether documentation of the indication for AU was present in the patient notes. For therapeutic AU, a distinction was made between therapy for community-acquired infections, for infections acquired in long-term care facilities and for infections acquired in acute care hospitals. Antimicrobial prophylaxis was divided into SP and medical prophylaxis (MP). SP was broken down into single-dose applications (SP1) or multiple applications on 1 day (SP2) or multiple applications over more than 1 day (SP3). Unlike all other data on AU, which represent the antimicrobial consumption on the day of the survey, data on SP were to be collected for the 24 h prior to the day of the survey in order to properly assess its duration. Only anonymized data were collected and sent to the NRZ. Data collection was in accordance with German recommendations for good epidemiological practice with respect to data collection.12 At the end of the survey, all participating hospitals were reminded to delete all records containing personalized patient information that had been collected during the survey. Hospitals in Germany are required by the German Protection Against Infection Act (Infektionsschutzgesetz §23) to collect and analyse data on HAI;13 therefore no ethical approval or informed consent was needed for the survey. Data management and analysis In order to process data from the participating hospitals, a web portal was created by the NRZ specifically for the PPS. Participants transferred the collected data via the web portal to the NRZ. Upon receiving the data, the NRZ checked all variables for implausible or erroneous values and performed further validation. EpiR, part of the statistics program R, was used to calculate the 95% CI. To compare prevalence data from 2011 and 2016, the χ2 test was employed. The Mann–Whitney U-test (for unpaired samples) and the Wilcoxon signed-rank test (for paired samples) were used for comparison of structural variables. In addition to the software R, the web-based program OpenEpi, Version 3.01 was used. All tests were two-sided. Statistical significance was defined as a P value <0.05. Results Data on 64 412 patients from 218 hospitals were collected in the PPS 2016. The representative sample of 49 hospitals included data on 11 324 patients. A total 17 462 patients were part of a group of 46 hospitals (core group) that participated in 2011 as well as in 2016. Comparison of the overall AU prevalence in the respective groups as documented in the PPS 2016 with data from 2011 yielded diverse results. Whereas there was no significant change in the prevalence of patients with AU in the group of all hospitals, there was a significant decrease in the representative sample and a significant increase in the core group (Table 1). Similar to the results of the 2011 survey, the majority of hospitals (62.8%) had a prevalence of patients with AU between 20% and 35%. Table 1. Hospital characteristics and prevalence of patients with AU in 2016 compared with 2011 Group  Parameter  Survey 2016  Survey 2011  P  All hospitals  number of hospitals  218  132    median number of beds  305  359  0.17*  patients included  64 412  41 539    prevalence of patients with AU (%)  25.9 (95% CI 25.6–26.3)  25.5 (95% CI 25.1–26.0)  0.18**  Representative samplea  number of hospitals  49  46    median number of beds  205  216  0.97*  patients included  11 324  9626    prevalence of patients with AU (%)  21.5 (95% CI 20.8–22.3)  23.3 (95% CI 22.5–24.2)  <0.01**  Core groupb  number of hospitals  46  46    median number of beds  392  368  0.86***  patients included  17 462  17 009    prevalence of patients with AU (%)  27.3 (95% CI 26.6–28.0)  26.2 (95% CI 25.5–26.8)  0.02**  Group  Parameter  Survey 2016  Survey 2011  P  All hospitals  number of hospitals  218  132    median number of beds  305  359  0.17*  patients included  64 412  41 539    prevalence of patients with AU (%)  25.9 (95% CI 25.6–26.3)  25.5 (95% CI 25.1–26.0)  0.18**  Representative samplea  number of hospitals  49  46    median number of beds  205  216  0.97*  patients included  11 324  9626    prevalence of patients with AU (%)  21.5 (95% CI 20.8–22.3)  23.3 (95% CI 22.5–24.2)  <0.01**  Core groupb  number of hospitals  46  46    median number of beds  392  368  0.86***  patients included  17 462  17 009    prevalence of patients with AU (%)  27.3 (95% CI 26.6–28.0)  26.2 (95% CI 25.5–26.8)  0.02**  Values in boldface indicate statistical significance (P < 0.05). * Calculation of P value using Mann–Whitney U-test. ** Calculation of P value using χ2 test. *** Calculation of P value using Wilcoxon signed-rank test. a Representative sample as required by the ECDC. b Group of hospitals that participated in both surveys. All results that are illustrated in the following represent the data from the group of all participating hospitals (n = 218). Seven university hospitals participated in the PPS 2016. The prevalence of patients with AU in university hospitals (31.3%) was significantly higher than in non-university hospitals (25.1%) (P < 0.01). The mean AU rate in the ICU was 52.0%, significantly higher than in non-ICU wards (24.4%) (P < 0.01). Surgical (30.1%) and medical patients (27.8%) showed the highest AU prevalence, whereas paediatric (14.9%) and psychiatric (1.6%) were the patient specialties with the lowest AU prevalence. Parenteral application (72.0%) was the most frequent route of application of antimicrobials. Of all antimicrobials, 27.6% were given orally. Other routes of application (rectal, inhalative) played only a minor role (0.4%). Among indications for AU, treatment of infections was the most prevalent (73.0%). The treatment of community-acquired infections accounted for 51.3% and treatment of infections acquired in long-term care facilities was the reason for AU in 1.7% of applications. Treatment of infections acquired in acute care hospitals made up 20.0% of all AU and 21.7% of AU were administered as prophylaxis. Of all prophylaxes, 68.4% were for SP, whereas 31.6% were given as MP. A total of 5.4% of AU were given for unknown (4.7%) or other (0.7%) indications (e.g. erythromycin as a prokinetic agent). Regarding SP, a significant increase in the proportion of single-dose applications from 2011 to 2016 (23.3% versus 38.5%) was observed (P < 0.01). Correspondingly, a significant decrease in the proportion of prophylaxis for more than 1 day (70.3% versus 56.1%) was noted (P < 0.01) (Table 2). Table 2. Distribution of SP in all hospitals in 2016 (n = 218) compared with 2011 (n = 132) Duration of SP  No. of applications 2016  Percentage 2016 (%)  No. of applications 2011  Percentage 2011 (%)  P  Single dose  1262  38.5  619  23.3  <0.01*  Multiple applications <1 day  177  5.4  171  6.4  0.09*  Multiple applications >1 day  1839  56.1  1866  70.3  <0.01*  Total  3278  100.0  2656  100.0    Duration of SP  No. of applications 2016  Percentage 2016 (%)  No. of applications 2011  Percentage 2011 (%)  P  Single dose  1262  38.5  619  23.3  <0.01*  Multiple applications <1 day  177  5.4  171  6.4  0.09*  Multiple applications >1 day  1839  56.1  1866  70.3  <0.01*  Total  3278  100.0  2656  100.0    * Calculation of P value using χ2 test. Values in boldface indicate statistical significance (P < 0.05). There was significantly less documentation of indication in 2016 than in 2011. In 2011, the indication for the application was documented in the patient notes for 72.5% of antimicrobials. In 2016, that value was significantly lower (68.7%, P < 0.01). Penicillins plus β-lactamase inhibitors (BLIs) represented the most frequently administered antimicrobial group (23.2%) followed by second-generation cephalosporins (12.9%), fluoroquinolones (11.3%), third-generation cephalosporins (8.9%) and carbapenems (6.2%). A significant increase was noted in the application of penicillins plus BLIs. Conversely, significantly less usage of fluoroquinolones, second-generation cephalosporins, third-generation cephalosporins and penicillins with extended spectrum was observed. Table 3 illustrates the 10 most frequently administered antimicrobial groups of the PPS 2016 compared with data from the 2011 survey. Almost 99% of all second-generation cephalosporins were cefuroxime. Unlike the other four most frequently documented antimicrobial groups, which were predominantly used for therapeutic indications, second-generation cephalosporins were administered primarily for antimicrobial prophylaxis. Table 3. Most frequently administered antimicrobial groups in all hospitals in 2016 (n = 218) compared with 2011 (n = 132) Antimicrobial group  No. of applications 2016  Percentage 2016 (%)  No. of applications 2011  Percentage 2011 (%)  P  Penicillins plus BLIs  5119  23.2  1773  12.6  <0.01*  Second-generation cephalosporins  2856  12.9  2054  14.6  <0.01*  Fluoroquinolones  2494  11.3  1971  14.0  <0.01*  Third-generation cephalosporins  1971  8.9  1498  10.6  <0.01*  Carbapenems  1369  6.2  825  5.9  0.19*  Imidazole derivatives  1138  5.2  741  5.3  0.64*  Macrolides  833  3.8  545  3.9  0.63*  Lincosamides  699  3.2  487  3.5  0.15*  Penicillins with extended spectrum  682  3.1  765  5.4  <0.01*  Glycopeptide antibacterials  653  3.0  410  2.9  0.81*  Total  22 086  100.0  14 076  100.0    Antimicrobial group  No. of applications 2016  Percentage 2016 (%)  No. of applications 2011  Percentage 2011 (%)  P  Penicillins plus BLIs  5119  23.2  1773  12.6  <0.01*  Second-generation cephalosporins  2856  12.9  2054  14.6  <0.01*  Fluoroquinolones  2494  11.3  1971  14.0  <0.01*  Third-generation cephalosporins  1971  8.9  1498  10.6  <0.01*  Carbapenems  1369  6.2  825  5.9  0.19*  Imidazole derivatives  1138  5.2  741  5.3  0.64*  Macrolides  833  3.8  545  3.9  0.63*  Lincosamides  699  3.2  487  3.5  0.15*  Penicillins with extended spectrum  682  3.1  765  5.4  <0.01*  Glycopeptide antibacterials  653  3.0  410  2.9  0.81*  Total  22 086  100.0  14 076  100.0    * Calculation of P value using χ2 test. Values in boldface indicate statistical significance (P < 0.05). Cefuroxime (12.8%), piperacillin plus BLI (12.1%) and ciprofloxacin (8.0%) were the most frequently administered antimicrobial agents in 2016. Significantly more piperacillin plus BLI, ampicillin plus BLI, meropenem, clarithromycin and trimethoprim/sulfamethoxazole were used in 2016. A significant decrease was noted in the usage of cefuroxime and ciprofloxacin (Table 4). Trimethoprim/sulfamethoxazole was the most frequently used agent for MP (20.6% of all MP). Table 4. Most frequently administered antimicrobial agents in all hospitals in 2016 (n = 218) compared with 2011 (n = 132) Antimicrobial agent  No. of applications 2016  Percentage 2016 (%)  No. of applications 2011  Percentage 2011 (%)  P  Cefuroxime  2816  12.8  2006  14.3  <0.01*  Piperacillin plus BLI  2672  12.1  582  4.1  <0.01*  Ciprofloxacin  1774  8.0  1384  9.8  <0.01*  Ampicillin plus BLI  1563  7.1  689  4.9  <0.01*  Ceftriaxone  1562  7.1  1056  7.5  0.12*  Metronidazole (parenteral)  1136  5.1  740  5.3  0.63*  Meropenem  1031  4.7  524  3.7  <0.01*  Clindamycin  695  3.1  486  3.5  0.11*  Clarithromycin  612  2.8  341  2.4  0.04*  Trimethoprim/sulfamethoxazole  606  2.7  325  2.3  0.01*  Total  22 086  100.0  14 076  100.0    Antimicrobial agent  No. of applications 2016  Percentage 2016 (%)  No. of applications 2011  Percentage 2011 (%)  P  Cefuroxime  2816  12.8  2006  14.3  <0.01*  Piperacillin plus BLI  2672  12.1  582  4.1  <0.01*  Ciprofloxacin  1774  8.0  1384  9.8  <0.01*  Ampicillin plus BLI  1563  7.1  689  4.9  <0.01*  Ceftriaxone  1562  7.1  1056  7.5  0.12*  Metronidazole (parenteral)  1136  5.1  740  5.3  0.63*  Meropenem  1031  4.7  524  3.7  <0.01*  Clindamycin  695  3.1  486  3.5  0.11*  Clarithromycin  612  2.8  341  2.4  0.04*  Trimethoprim/sulfamethoxazole  606  2.7  325  2.3  0.01*  Total  22 086  100.0  14 076  100.0    * Calculation of P value using χ2 test. Values in boldface indicate statistical significance (P < 0.05). Figures 1 and 2 illustrate the most frequently administered antimicrobial groups for the treatment of lower respiratory infections (LRI) and urinary tract infections (UTI). Figure 1. View largeDownload slide Most frequently administered antimicrobial groups for the treatment of LRI. Data from the PPS 2016. HI, infections acquired in acute care hospitals; CI, community-acquired infections; LI, infections acquired in long-term care facilities; PENβlacI, penicillins plus BLIs; Macrol, macrolides; C3G, third-generation cephalosporins; FLQ, fluoroquinolones; Carb, carbapenems; ExtSpPEN, penicillins with extended spectrum; C2G, second-generation cephalosporins; GlycopAB, glycopeptide antibacterials; Imidaz, imidazole derivatives; OTHAminogly, other aminoglycosides. Figure 1. View largeDownload slide Most frequently administered antimicrobial groups for the treatment of LRI. Data from the PPS 2016. HI, infections acquired in acute care hospitals; CI, community-acquired infections; LI, infections acquired in long-term care facilities; PENβlacI, penicillins plus BLIs; Macrol, macrolides; C3G, third-generation cephalosporins; FLQ, fluoroquinolones; Carb, carbapenems; ExtSpPEN, penicillins with extended spectrum; C2G, second-generation cephalosporins; GlycopAB, glycopeptide antibacterials; Imidaz, imidazole derivatives; OTHAminogly, other aminoglycosides. Figure 2. View largeDownload slide Most frequently administered antimicrobial groups for the treatment of UTI. Data from the PPS 2016. HI, infections acquired in acute care hospitals; CI, community-acquired infections; LI, infections acquired in long-term care facilities; FLQ, fluoroquinolones; PENβlacI, penicillins plus BLIs; C3G, third-generation cephalosporins; C2G, second-generation cephalosporins; Carb, carbapenems; SXT, trimethoprim/sulfamethoxazole; ExtSpPEN penicillins with extended spectrum; Nitro, nitrofuran derivatives; GlycopAB, glycopeptide antibacterials; OTHAminogly, other aminoglycosides. Figure 2. View largeDownload slide Most frequently administered antimicrobial groups for the treatment of UTI. Data from the PPS 2016. HI, infections acquired in acute care hospitals; CI, community-acquired infections; LI, infections acquired in long-term care facilities; FLQ, fluoroquinolones; PENβlacI, penicillins plus BLIs; C3G, third-generation cephalosporins; C2G, second-generation cephalosporins; Carb, carbapenems; SXT, trimethoprim/sulfamethoxazole; ExtSpPEN penicillins with extended spectrum; Nitro, nitrofuran derivatives; GlycopAB, glycopeptide antibacterials; OTHAminogly, other aminoglycosides. The validation of the collected data, for which members of the NRZ visited 10 hospitals including 432 patient cases, revealed a sensitivity of 96.7% and a specificity of 99.7% in identifying patients with AU. Discussion The PPS 2011 revealed a prevalence of patients with AU of 25.5%, which was substantially higher than the prevalence that was determined in the first national prevalence survey in 1994 (17.7%).14,15 In 2016, the survey was repeated following a largely identical methodology. Therefore, comparisons of the 2011 and the 2016 national surveys are most reliable. Various subgroups were identified for data analysis, of which the group of all participating hospitals, the representative sample as required by the ECDC, and the core group of hospitals participating both in 2011 and 2016 are of particular interest. When comparing prevalence data from 2011 and 2016, a significant decrease in AU prevalence was noted in the representative sample. However, AU prevalence remained at a rather constant level in the group of all participating hospitals. Moreover, it increased significantly in the core group. It is important to note that sampling for the representative sample was based solely on hospital size given in the German hospital register. Consequently, the representative sample is prone to random effects, which reduce the comparability of its data. The core group provides data for the most reliable comparison. However, it is possible that it does not accurately represent the overall situation in Germany. The 46 hospitals might have participated in the PPS again because of a higher than usual interest in surveillance and aspects of AMS or because they identified relevant problems in previous surveys. Patient demographics, such as the average age of patients hospitalized in Germany, remained at a constant level from 2011 to 2016.16 The average length of stay of patients in German hospitals has not changed substantially between 2011 and 2016 either,17 which increases the comparability of the 2011 and 2016 surveys. The high sensitivity and specificity in identifying patients with AU, which was shown by the validation study in 10 hospitals, corroborate the validity of the results on AU prevalence as assessed by the local data collection teams. The frequent parenteral application of antimicrobials that was observed in the current survey may be linked to the trend of patients being quickly discharged from hospitals after the conclusion of parenteral antimicrobial therapy.18 The most frequent indication for AU was the treatment of community-acquired infections. Roughly one-fifth of all administered antimicrobials were given to treat hospital-acquired infections, which highlights the importance of infection prevention in healthcare facilities. Regarding AU for SP, a prolonged administration of SP is still frequently observed in German hospitals, which illustrates the substantial differences between the reality of SP in German hospitals and nationally and internationally established guidelines.19,20 Nevertheless, a significant decrease in the frequency of SP3 among all SP was noted from 2011 to 2016. This development was possibly caused by improved awareness and implementation of guidelines as well as by intensified efforts in AMS in many hospitals in recent years. The documentation of the reason for AU in the patient charts has decreased significantly. The insufficient documentation corresponds with the relevant proportion of almost 5% of AU for which the data collection teams could not identify the reason for the AU even after consulting the ward staff. With regard to the antimicrobial groups and agents, one of the most remarkable results was the significant increase in the percentage of penicillins plus BLIs among all administered antimicrobials, which was almost twice as high in 2016 as in 2011. Penicillins plus BLIs seem to have replaced other frequently applied antimicrobial groups, such as second-generation cephalosporins, third-generation cephalosporins, fluoroquinolones and penicillins with extended spectrum. These antimicrobial groups cover a comparable range of pathogens and represented a significantly smaller proportion in 2016 than in 2011. Almost 50% of all AU could be attributed to the broad-spectrum antimicrobials: penicillins plus BLIs, fluoroquinolones, third-generation cephalosporins and carbapenems. The share of these antimicrobial groups among all AU has grown further since 2011. The significant increase in penicillins plus BLIs was largely due to the substantial and statistically significant increase in the use of piperacillin plus BLI (mostly tazobactam), which has almost tripled. Noticeably, the use of ampicillin plus BLI (usually sulbactam) has also grown significantly from 2011 to 2016. Both for community-acquired and hospital-acquired LRI, penicillins plus BLIs are the most commonly used antimicrobial group. Furthermore, penicillins plus BLIs represent the second most prevalent antimicrobial group for the treatment of UTI (for community-acquired and hospital-acquired UTI combined). The high percentage of anti-pseudomonal antimicrobials for community-acquired infections is surprising and goes counter to the German guidelines for the treatment of community-acquired LRI and UTI.21,22,Pseudomonas spp. are not routinely found as pathogens in community-acquired LRI and UTI in patients without corresponding risk factors. Moreover, the significant increase in the use of penicillins plus BLIs may also be linked to the reported increase in Escherichia coli that is resistant to third-generation cephalosporins in Germany.23 Carbapenems represent the fifth most frequently used antimicrobial group. An increase was observed from 2011 to 2016 for all carbapenems; in particular, the use of meropenem has grown significantly. A possible explanation can also be seen in the increase in resistance to third-generation cephalosporins.24 In times of emerging carbapenem-resistant organisms,25 the high consumption of carbapenems reveals significant potential for improvement and for reducing the use of broad-spectrum antimicrobials. The significant increase in the use of trimethoprim/sulfamethoxazole is owing to the fact that this agent was the most frequently used antimicrobial for MP. With a share of 6.8%,7 MP continues to be a relevant indication for AU in German acute care hospitals. The prevalence of patients with AU is rather low in Germany compared with other European countries.5 Nevertheless, the results of the PPS 2016 re-emphasize the areas for improvement that previous surveys already identified: the prolonged administration of SP, the insufficient documentation of indication, the relevant proportion of antimicrobials for which no indication can be identified (even after consulting the ward staff) and an increase in the use of broad-spectrum antimicrobials. Despite recent efforts to promote AMS in German hospitals,26 the results of the PPS 2016 do not reveal substantial measurable progress on the matter yet. It is therefore crucial to invest further in AMS education and training and in hospital-wide AMS activities.27 Acknowledgements Our thanks go to all participating hospitals for their dedication and willingness to participate in this survey and to the ECDC for initiating and supporting the PPS at the European level. We wish to give special thanks to all hospitals that participated in the validation study for welcoming the NRZ validation team and allowing us to visit their hospital sites. Furthermore, we thank Dr Janine Zweigner (Universitätsklinikum Köln), PD Dr Roland Schulze-Röbbecke (Universitätsklinikum Düsseldorf), Dr Gerhard Schwarzkopf-Steinhauser (Klinikum München), PD Dr Christian Brandt (Director of the Institute of Hygiene and Environmental Medicine of Vivantes Netzwerk für Gesundheit GmbH, Berlin, formerly Universitätsklinikum Frankfurt) and Dr Susanne Huggett (Asklepios Klinik Altona) for hosting 1-day PPS introductory courses at their institutions. Funding The German Ministry of Health funded the national PPS 2016. Transparency declarations None to declare. References 1 Ruden H, Gastmeier P, Daschner F et al.   [Nosocomial infections in Germany. Their epidemiology in old and new Federal Lander]. Dtsch Med Wochenschr  1996; 121: 1281– 7. Google Scholar CrossRef Search ADS PubMed  2 Hansen S, Sohr D, Piening B et al.   Antibiotic usage in German hospitals: results of the second national prevalence study. J Antimicrob Chemother  2013; 68: 2934– 9. Google Scholar CrossRef Search ADS PubMed  3 Behnke M, Hansen S, Leistner R et al.   Nosocomial infection and antibiotic use: a second national prevalence study in Germany. Dtsch Arztebl Int  2013; 110: 627– 33. Google Scholar PubMed  4 Bundesamt für Verbraucherschutz und Lebensmittelsicherheit, Paul-Ehrlich-Gesellschaft für Chemotherapie e.V., Infektiologie Freiburg. GERMAP 2012 - Bericht über den Antibiotikaverbrauch und die Verbreitung von Antibiotikaresistenzen in der Human- und Veterinärmedizin in Deutschland. Rheinbach, Germany: Antiinfectives Intelligence, 2014. 5 ECDC. Point Prevalence Survey of Healthcare-Associated Infections and Antimicrobial Use in European Acute Care Hospitals . Stockholm, Sweden: ECDC, 2013. 6 Walger P, Fätkenheuer G, Herrmann M et al.   Infektionsmedizin: klinische expertise fördern. Dtsch Arztebl International  2017; 114: A948– 50. 7 Nationales Referenzzentrum für die Surveillance von nosokomialen Infektionen. German National Point Prevalence Survey of Healthcare-Associated Infections and Antimicrobial Usage 2016, Final Report. http://www.nrz-hygiene.de/fileadmin/nrz/download/pps2016/PPS_2016_Abschlussbericht_20.07.2017.pdf. 8 ECDC. Point Prevalence Survey of Healthcare-Associated Infections and Antimicrobial Use in European Acute Care Hospitals—Protocol Version 5.3 . Stockholm, Sweden: ECDC, 2016. 9 Nationales Referenzzentrum für die Surveillance von nosokomialen Infektionen. Protocol of the German Point Prevalence Survey. http://www.nrz-hygiene.de/fileadmin/nrz/download/pps2016/EUPPS2016DE_Protokoll_Version_4.6.pdf. 10 Statistisches Bundesamt (Destatis). List of Hospitals and Rehabilitation Clinics in Germany. https://www.destatis.de/DE/Publikationen/Thematisch/Gesundheit/Krankenhaeuser/Krankenhausverzeichnis.html. 11 WHO Collaborating Centre for Drug Statistics Methodology. Guidelines for ATC Classification and DDD Assignment 2013. Oslo, Norway: WHO Collaborating Centre for Drug Statistics Methodology, 2012. https://www.whocc.no/filearchive/publications/1_2013guidelines.pdf. 12 German Recommendations of Good Epidemiological Practice. https://dgepi.de/fileadmin/pdf/GEP_LL_english_f.pdf. 13 German Protection Against Infection Act. http://www.gesetze-im-internet.de/ifsg/BJNR104510000.html. 14 Ruden H. Nosocomial Infections in Germany—Surveillance and Prevention. Band 56 der Schriftenreihe des Bundesministeriums für Gesundheit . Baden-Baden, Germany: Nomos Verlagsgesellschaft, 1995. 15 Hauer T, Lacour M, Gastmeier P et al.   [Nosocomial infections in Germany. Microbiological diagnosis, preventive antibiotics and antibiotic therapy]. Med Klin (Munich)  1996; 91: 681– 6. Google Scholar PubMed  16 Diagnosedaten der Patienten und Patientinnen in Krankenhäusern (einschließlich Sterbe- und Stundenfälle).  Wiesbaden, Germany: Statistisches Bundesamt, 2011. 17 Statistisches Bundesamt. Average Length of Stay in German Hospitals in the Years 1992 to 2015 (in days). https://de.statista.com/statistik/daten/studie/2604/umfrage/durchschnittliche-verweildauer-im-krankenhaus-seit-1992/. 18 Lee RW, Lindstrom ST. Early switch to oral antibiotics and early discharge guidelines in the management of community-acquired pneumonia. Respirology  2007; 12: 111– 6. Google Scholar CrossRef Search ADS PubMed  19 Allegranzi B, Zayed B, Bischoff P et al.   New WHO recommendations on intraoperative and postoperative measures for surgical site infection prevention: an evidence-based global perspective. Lancet Infect Dis  2016; 16: e288– 303. Google Scholar CrossRef Search ADS PubMed  20 Berrios-Torres SI, Umscheid CA, Bratzler DW et al.   Centers for Disease Control and Prevention guideline for the prevention of surgical site infection, 2017. JAMA Surg  2017; 152: 784– 91. Google Scholar CrossRef Search ADS PubMed  21 Wagenlehner F, Schmiemann G, Hoyme U et al.   Nationale S3-Leitlinie “Unkomplizierte Harnwegsinfektionen”. Der Urologe  2011; 50: 153– 69. Google Scholar CrossRef Search ADS PubMed  22 Ewig S, Höffken G, Kern W et al.   Behandlung von erwachsenen Patienten mit ambulant erworbener Pneumonie und Prävention–update 2016. Pneumologie  2016; 70: 151– 200. Google Scholar CrossRef Search ADS PubMed  23 Maechler F, Geffers C, Schwab F et al.   [Development of antimicrobial resistance in Germany: What is the current situation?]. Med Klin Intensivmed Notfmed  2017; 112: 186– 91. Google Scholar CrossRef Search ADS PubMed  24 Hamprecht A, Rohde AM, Behnke M et al.   Colonization with third-generation cephalosporin-resistant Enterobacteriaceae on hospital admission: prevalence and risk factors. J Antimicrob Chemother  2016; 71: 2957– 63. Google Scholar CrossRef Search ADS PubMed  25 Grundmann H, Glasner C, Albiger B et al.   Occurrence of carbapenemase-producing Klebsiella pneumoniae and Escherichia coli in the European survey of carbapenemase-producing Enterobacteriaceae (EuSCAPE): a prospective, multinational study. Lancet Infect Dis  2017; 17: 153– 63. Google Scholar CrossRef Search ADS PubMed  26 de With K, Allerberger F, Amann S et al.   Strategies to enhance rational use of antibiotics in hospital: a guideline by the German Society for Infectious Diseases. Infection  2016; 44: 395– 439. Google Scholar CrossRef Search ADS PubMed  27 Hansen S, Schwab F, Gropmann A et al.   Hygiene und sicherheitskultur in deutschen krankenhäusern. Bundesgesundheitsbl  2016; 59: 908– 15. Google Scholar CrossRef Search ADS   © The Author(s) 2018. Published by Oxford University Press on behalf of the British Society for Antimicrobial Chemotherapy. All rights reserved. For Permissions, please email: journals.permissions@oup.com. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Journal of Antimicrobial Chemotherapy Oxford University Press

Antimicrobial usage in German acute care hospitals: results of the third national point prevalence survey and comparison with previous national point prevalence surveys

Loading next page...
 
/lp/ou_press/antimicrobial-usage-in-german-acute-care-hospitals-results-of-the-Ep804DDvxj
Publisher
Oxford University Press
ISSN
0305-7453
eISSN
1460-2091
D.O.I.
10.1093/jac/dkx494
Publisher site
See Article on Publisher Site

Abstract

Abstract Objectives Previous point prevalence surveys (PPSs) revealed the potential for improving antimicrobial usage (AU) in German acute care hospitals. Data from the 2016 German national PPS on healthcare-associated infections and AU were used to evaluate efforts in antimicrobial stewardship (AMS). Methods A national PPS in Germany was organized by the German National Reference Centre for Surveillance of Nosocomial Infections in 2016 as part of the European PPS initiated by the ECDC. The data were collected in May and June 2016. Results were compared with data from the PPS 2011. Results A total of 218 hospitals with 64 412 observed patients participated in the PPS 2016. The prevalence of patients with AU was 25.9% (95% CI 25.6%–26.3%). No significant increase or decrease in AU prevalence was revealed in the group of all participating hospitals. Prolonged surgical prophylaxis was found to be common (56.1% of all surgical prophylaxes on the prevalence day), but significantly less prevalent than in 2011 (P < 0.01). The most frequently administered antimicrobial groups were penicillins plus β-lactamase inhibitors (BLIs) (23.2%), second-generation cephalosporins (12.9%) and fluoroquinolones (11.3%). Significantly more penicillins plus BLIs and fewer second-generation cephalosporins and fluoroquinolones were used in 2016. Overall, an increase in the consumption of broad-spectrum antimicrobials was noted. For 68.7% of all administered antimicrobials, the indication was documented in the patient notes. Conclusions The current data reaffirm the points of improvement that previous data identified and reveal that recent efforts in AMS in German hospitals require further intensification. Introduction Five years after the second national point prevalence survey (PPS) of healthcare-associated infections (HAI) and antimicrobial usage (AU) in Germany in 2011, a third national PPS was conducted in 2016 following a similar methodology. The first such national survey was conducted in 1994.1 Results of the 2011 PPS were widely published.2–4 The prevalence of patients with AU was calculated to be 25.5% and various points for improvement such as insufficient documentation, extensive use of broad-spectrum antimicrobials and high frequency of prolonged surgical prophylaxis (SP) were described.2 The 2011 and the 2016 surveys were embedded in the European PPSs that were initiated by the ECDC. Similarly to the 2011/12 PPS, the ECDC organized the 2016/17 PPS in four periods. Germany was among the first countries to conduct the PPS in May and June 2016. European reference data from the 2016/17 ECDC PPS are, therefore, currently not available. In 2011/12, 29 EU countries and Croatia participated in the ECDC PPS. The prevalence of patients who received at least one antimicrobial on the day of the survey was found to be 35.0%.5 Efforts in promoting antimicrobial stewardship (AMS) in Germany have been intensified in recent years. For the year 2017, it is estimated that over 1000 AMS specialists will have been educated.6 Since results concerning data on HAI have been and will be published elsewhere,7 this article will focus solely on the findings regarding AU. Methods Protocol As in the 2011/12 survey, the ECDC provided a standardized study protocol that specified uniform definitions and a uniform methodology.8 The German National Reference Centre for Surveillance of Nosocomial Infections (NRZ) was responsible for organizing the survey in Germany. The ECDC protocol version 5.1 was implemented and materials were translated accordingly.9 The ECDC offered two different study protocols: a standard protocol and a light protocol. The most relevant difference was that patient data in the light protocol were collected only for patients who had at least one active HAI and/or who had received at least one antimicrobial on the day of the survey, whereas in the standard protocol, patient data were to be recorded for every patient. The light protocol version was applied in Germany in order to reduce the workload for the participating hospitals and to enable participation by a higher number of hospitals. Representative sample of hospitals The ECDC required all participating countries to invite a representative sample of hospitals to take part in the survey. Based on population and the structure of the healthcare system, 49 hospitals were calculated as the sample size for Germany. These were randomly selected by the number of beds from the German hospital register 2013.10 Only hospitals listed as acute care hospitals were included. They were invited to participate in the PPS on a voluntary basis. For every hospital selected, two other hospitals of comparable size were chosen as possible substitutes in case a selected hospital decided against participation. Moreover, all 1462 (as of the first quarter of 2016) hospitals participating in the German Krankenhaus-Infektions-Surveillance-System (KISS) were invited to participate. Data validation Ten of the 49 hospitals, which constituted the representative sample, were visited by two members of the NRZ, who represented the gold standard, to perform a validation study of the data collection in order to assess the sensitivity and specificity of the local PPS teams. Data collection Data on HAI and AU as well as structural and process indicators were collected. Local staff of the participating hospitals executed the data collection. To ensure application of consistent methodology and definitions, at least one member of the local data collection team per hospital was trained in one of the seven 1 day courses that were organized by the NRZ prior to the start of the PPS. The data were collected in the months of May and June 2016. The local PPS teams visited the wards of the hospital in succession and reviewed the charts of all patients who were present in the ward and who had been admitted to the ward before 08:00 on the day of survey. Patient data documented in the patient charts as well as further information available from the ward staff upon request were evaluated. Data from outpatients were not included in the survey. For every patient receiving at least one antimicrobial, the following data were collected: the WHO Anatomical Therapeutic Chemical (ATC5) code of the antimicrobial,11 the route of application, the indication for AU (therapy versus prophylaxis versus other versus unknown), the site of infection for therapeutic indication and whether documentation of the indication for AU was present in the patient notes. For therapeutic AU, a distinction was made between therapy for community-acquired infections, for infections acquired in long-term care facilities and for infections acquired in acute care hospitals. Antimicrobial prophylaxis was divided into SP and medical prophylaxis (MP). SP was broken down into single-dose applications (SP1) or multiple applications on 1 day (SP2) or multiple applications over more than 1 day (SP3). Unlike all other data on AU, which represent the antimicrobial consumption on the day of the survey, data on SP were to be collected for the 24 h prior to the day of the survey in order to properly assess its duration. Only anonymized data were collected and sent to the NRZ. Data collection was in accordance with German recommendations for good epidemiological practice with respect to data collection.12 At the end of the survey, all participating hospitals were reminded to delete all records containing personalized patient information that had been collected during the survey. Hospitals in Germany are required by the German Protection Against Infection Act (Infektionsschutzgesetz §23) to collect and analyse data on HAI;13 therefore no ethical approval or informed consent was needed for the survey. Data management and analysis In order to process data from the participating hospitals, a web portal was created by the NRZ specifically for the PPS. Participants transferred the collected data via the web portal to the NRZ. Upon receiving the data, the NRZ checked all variables for implausible or erroneous values and performed further validation. EpiR, part of the statistics program R, was used to calculate the 95% CI. To compare prevalence data from 2011 and 2016, the χ2 test was employed. The Mann–Whitney U-test (for unpaired samples) and the Wilcoxon signed-rank test (for paired samples) were used for comparison of structural variables. In addition to the software R, the web-based program OpenEpi, Version 3.01 was used. All tests were two-sided. Statistical significance was defined as a P value <0.05. Results Data on 64 412 patients from 218 hospitals were collected in the PPS 2016. The representative sample of 49 hospitals included data on 11 324 patients. A total 17 462 patients were part of a group of 46 hospitals (core group) that participated in 2011 as well as in 2016. Comparison of the overall AU prevalence in the respective groups as documented in the PPS 2016 with data from 2011 yielded diverse results. Whereas there was no significant change in the prevalence of patients with AU in the group of all hospitals, there was a significant decrease in the representative sample and a significant increase in the core group (Table 1). Similar to the results of the 2011 survey, the majority of hospitals (62.8%) had a prevalence of patients with AU between 20% and 35%. Table 1. Hospital characteristics and prevalence of patients with AU in 2016 compared with 2011 Group  Parameter  Survey 2016  Survey 2011  P  All hospitals  number of hospitals  218  132    median number of beds  305  359  0.17*  patients included  64 412  41 539    prevalence of patients with AU (%)  25.9 (95% CI 25.6–26.3)  25.5 (95% CI 25.1–26.0)  0.18**  Representative samplea  number of hospitals  49  46    median number of beds  205  216  0.97*  patients included  11 324  9626    prevalence of patients with AU (%)  21.5 (95% CI 20.8–22.3)  23.3 (95% CI 22.5–24.2)  <0.01**  Core groupb  number of hospitals  46  46    median number of beds  392  368  0.86***  patients included  17 462  17 009    prevalence of patients with AU (%)  27.3 (95% CI 26.6–28.0)  26.2 (95% CI 25.5–26.8)  0.02**  Group  Parameter  Survey 2016  Survey 2011  P  All hospitals  number of hospitals  218  132    median number of beds  305  359  0.17*  patients included  64 412  41 539    prevalence of patients with AU (%)  25.9 (95% CI 25.6–26.3)  25.5 (95% CI 25.1–26.0)  0.18**  Representative samplea  number of hospitals  49  46    median number of beds  205  216  0.97*  patients included  11 324  9626    prevalence of patients with AU (%)  21.5 (95% CI 20.8–22.3)  23.3 (95% CI 22.5–24.2)  <0.01**  Core groupb  number of hospitals  46  46    median number of beds  392  368  0.86***  patients included  17 462  17 009    prevalence of patients with AU (%)  27.3 (95% CI 26.6–28.0)  26.2 (95% CI 25.5–26.8)  0.02**  Values in boldface indicate statistical significance (P < 0.05). * Calculation of P value using Mann–Whitney U-test. ** Calculation of P value using χ2 test. *** Calculation of P value using Wilcoxon signed-rank test. a Representative sample as required by the ECDC. b Group of hospitals that participated in both surveys. All results that are illustrated in the following represent the data from the group of all participating hospitals (n = 218). Seven university hospitals participated in the PPS 2016. The prevalence of patients with AU in university hospitals (31.3%) was significantly higher than in non-university hospitals (25.1%) (P < 0.01). The mean AU rate in the ICU was 52.0%, significantly higher than in non-ICU wards (24.4%) (P < 0.01). Surgical (30.1%) and medical patients (27.8%) showed the highest AU prevalence, whereas paediatric (14.9%) and psychiatric (1.6%) were the patient specialties with the lowest AU prevalence. Parenteral application (72.0%) was the most frequent route of application of antimicrobials. Of all antimicrobials, 27.6% were given orally. Other routes of application (rectal, inhalative) played only a minor role (0.4%). Among indications for AU, treatment of infections was the most prevalent (73.0%). The treatment of community-acquired infections accounted for 51.3% and treatment of infections acquired in long-term care facilities was the reason for AU in 1.7% of applications. Treatment of infections acquired in acute care hospitals made up 20.0% of all AU and 21.7% of AU were administered as prophylaxis. Of all prophylaxes, 68.4% were for SP, whereas 31.6% were given as MP. A total of 5.4% of AU were given for unknown (4.7%) or other (0.7%) indications (e.g. erythromycin as a prokinetic agent). Regarding SP, a significant increase in the proportion of single-dose applications from 2011 to 2016 (23.3% versus 38.5%) was observed (P < 0.01). Correspondingly, a significant decrease in the proportion of prophylaxis for more than 1 day (70.3% versus 56.1%) was noted (P < 0.01) (Table 2). Table 2. Distribution of SP in all hospitals in 2016 (n = 218) compared with 2011 (n = 132) Duration of SP  No. of applications 2016  Percentage 2016 (%)  No. of applications 2011  Percentage 2011 (%)  P  Single dose  1262  38.5  619  23.3  <0.01*  Multiple applications <1 day  177  5.4  171  6.4  0.09*  Multiple applications >1 day  1839  56.1  1866  70.3  <0.01*  Total  3278  100.0  2656  100.0    Duration of SP  No. of applications 2016  Percentage 2016 (%)  No. of applications 2011  Percentage 2011 (%)  P  Single dose  1262  38.5  619  23.3  <0.01*  Multiple applications <1 day  177  5.4  171  6.4  0.09*  Multiple applications >1 day  1839  56.1  1866  70.3  <0.01*  Total  3278  100.0  2656  100.0    * Calculation of P value using χ2 test. Values in boldface indicate statistical significance (P < 0.05). There was significantly less documentation of indication in 2016 than in 2011. In 2011, the indication for the application was documented in the patient notes for 72.5% of antimicrobials. In 2016, that value was significantly lower (68.7%, P < 0.01). Penicillins plus β-lactamase inhibitors (BLIs) represented the most frequently administered antimicrobial group (23.2%) followed by second-generation cephalosporins (12.9%), fluoroquinolones (11.3%), third-generation cephalosporins (8.9%) and carbapenems (6.2%). A significant increase was noted in the application of penicillins plus BLIs. Conversely, significantly less usage of fluoroquinolones, second-generation cephalosporins, third-generation cephalosporins and penicillins with extended spectrum was observed. Table 3 illustrates the 10 most frequently administered antimicrobial groups of the PPS 2016 compared with data from the 2011 survey. Almost 99% of all second-generation cephalosporins were cefuroxime. Unlike the other four most frequently documented antimicrobial groups, which were predominantly used for therapeutic indications, second-generation cephalosporins were administered primarily for antimicrobial prophylaxis. Table 3. Most frequently administered antimicrobial groups in all hospitals in 2016 (n = 218) compared with 2011 (n = 132) Antimicrobial group  No. of applications 2016  Percentage 2016 (%)  No. of applications 2011  Percentage 2011 (%)  P  Penicillins plus BLIs  5119  23.2  1773  12.6  <0.01*  Second-generation cephalosporins  2856  12.9  2054  14.6  <0.01*  Fluoroquinolones  2494  11.3  1971  14.0  <0.01*  Third-generation cephalosporins  1971  8.9  1498  10.6  <0.01*  Carbapenems  1369  6.2  825  5.9  0.19*  Imidazole derivatives  1138  5.2  741  5.3  0.64*  Macrolides  833  3.8  545  3.9  0.63*  Lincosamides  699  3.2  487  3.5  0.15*  Penicillins with extended spectrum  682  3.1  765  5.4  <0.01*  Glycopeptide antibacterials  653  3.0  410  2.9  0.81*  Total  22 086  100.0  14 076  100.0    Antimicrobial group  No. of applications 2016  Percentage 2016 (%)  No. of applications 2011  Percentage 2011 (%)  P  Penicillins plus BLIs  5119  23.2  1773  12.6  <0.01*  Second-generation cephalosporins  2856  12.9  2054  14.6  <0.01*  Fluoroquinolones  2494  11.3  1971  14.0  <0.01*  Third-generation cephalosporins  1971  8.9  1498  10.6  <0.01*  Carbapenems  1369  6.2  825  5.9  0.19*  Imidazole derivatives  1138  5.2  741  5.3  0.64*  Macrolides  833  3.8  545  3.9  0.63*  Lincosamides  699  3.2  487  3.5  0.15*  Penicillins with extended spectrum  682  3.1  765  5.4  <0.01*  Glycopeptide antibacterials  653  3.0  410  2.9  0.81*  Total  22 086  100.0  14 076  100.0    * Calculation of P value using χ2 test. Values in boldface indicate statistical significance (P < 0.05). Cefuroxime (12.8%), piperacillin plus BLI (12.1%) and ciprofloxacin (8.0%) were the most frequently administered antimicrobial agents in 2016. Significantly more piperacillin plus BLI, ampicillin plus BLI, meropenem, clarithromycin and trimethoprim/sulfamethoxazole were used in 2016. A significant decrease was noted in the usage of cefuroxime and ciprofloxacin (Table 4). Trimethoprim/sulfamethoxazole was the most frequently used agent for MP (20.6% of all MP). Table 4. Most frequently administered antimicrobial agents in all hospitals in 2016 (n = 218) compared with 2011 (n = 132) Antimicrobial agent  No. of applications 2016  Percentage 2016 (%)  No. of applications 2011  Percentage 2011 (%)  P  Cefuroxime  2816  12.8  2006  14.3  <0.01*  Piperacillin plus BLI  2672  12.1  582  4.1  <0.01*  Ciprofloxacin  1774  8.0  1384  9.8  <0.01*  Ampicillin plus BLI  1563  7.1  689  4.9  <0.01*  Ceftriaxone  1562  7.1  1056  7.5  0.12*  Metronidazole (parenteral)  1136  5.1  740  5.3  0.63*  Meropenem  1031  4.7  524  3.7  <0.01*  Clindamycin  695  3.1  486  3.5  0.11*  Clarithromycin  612  2.8  341  2.4  0.04*  Trimethoprim/sulfamethoxazole  606  2.7  325  2.3  0.01*  Total  22 086  100.0  14 076  100.0    Antimicrobial agent  No. of applications 2016  Percentage 2016 (%)  No. of applications 2011  Percentage 2011 (%)  P  Cefuroxime  2816  12.8  2006  14.3  <0.01*  Piperacillin plus BLI  2672  12.1  582  4.1  <0.01*  Ciprofloxacin  1774  8.0  1384  9.8  <0.01*  Ampicillin plus BLI  1563  7.1  689  4.9  <0.01*  Ceftriaxone  1562  7.1  1056  7.5  0.12*  Metronidazole (parenteral)  1136  5.1  740  5.3  0.63*  Meropenem  1031  4.7  524  3.7  <0.01*  Clindamycin  695  3.1  486  3.5  0.11*  Clarithromycin  612  2.8  341  2.4  0.04*  Trimethoprim/sulfamethoxazole  606  2.7  325  2.3  0.01*  Total  22 086  100.0  14 076  100.0    * Calculation of P value using χ2 test. Values in boldface indicate statistical significance (P < 0.05). Figures 1 and 2 illustrate the most frequently administered antimicrobial groups for the treatment of lower respiratory infections (LRI) and urinary tract infections (UTI). Figure 1. View largeDownload slide Most frequently administered antimicrobial groups for the treatment of LRI. Data from the PPS 2016. HI, infections acquired in acute care hospitals; CI, community-acquired infections; LI, infections acquired in long-term care facilities; PENβlacI, penicillins plus BLIs; Macrol, macrolides; C3G, third-generation cephalosporins; FLQ, fluoroquinolones; Carb, carbapenems; ExtSpPEN, penicillins with extended spectrum; C2G, second-generation cephalosporins; GlycopAB, glycopeptide antibacterials; Imidaz, imidazole derivatives; OTHAminogly, other aminoglycosides. Figure 1. View largeDownload slide Most frequently administered antimicrobial groups for the treatment of LRI. Data from the PPS 2016. HI, infections acquired in acute care hospitals; CI, community-acquired infections; LI, infections acquired in long-term care facilities; PENβlacI, penicillins plus BLIs; Macrol, macrolides; C3G, third-generation cephalosporins; FLQ, fluoroquinolones; Carb, carbapenems; ExtSpPEN, penicillins with extended spectrum; C2G, second-generation cephalosporins; GlycopAB, glycopeptide antibacterials; Imidaz, imidazole derivatives; OTHAminogly, other aminoglycosides. Figure 2. View largeDownload slide Most frequently administered antimicrobial groups for the treatment of UTI. Data from the PPS 2016. HI, infections acquired in acute care hospitals; CI, community-acquired infections; LI, infections acquired in long-term care facilities; FLQ, fluoroquinolones; PENβlacI, penicillins plus BLIs; C3G, third-generation cephalosporins; C2G, second-generation cephalosporins; Carb, carbapenems; SXT, trimethoprim/sulfamethoxazole; ExtSpPEN penicillins with extended spectrum; Nitro, nitrofuran derivatives; GlycopAB, glycopeptide antibacterials; OTHAminogly, other aminoglycosides. Figure 2. View largeDownload slide Most frequently administered antimicrobial groups for the treatment of UTI. Data from the PPS 2016. HI, infections acquired in acute care hospitals; CI, community-acquired infections; LI, infections acquired in long-term care facilities; FLQ, fluoroquinolones; PENβlacI, penicillins plus BLIs; C3G, third-generation cephalosporins; C2G, second-generation cephalosporins; Carb, carbapenems; SXT, trimethoprim/sulfamethoxazole; ExtSpPEN penicillins with extended spectrum; Nitro, nitrofuran derivatives; GlycopAB, glycopeptide antibacterials; OTHAminogly, other aminoglycosides. The validation of the collected data, for which members of the NRZ visited 10 hospitals including 432 patient cases, revealed a sensitivity of 96.7% and a specificity of 99.7% in identifying patients with AU. Discussion The PPS 2011 revealed a prevalence of patients with AU of 25.5%, which was substantially higher than the prevalence that was determined in the first national prevalence survey in 1994 (17.7%).14,15 In 2016, the survey was repeated following a largely identical methodology. Therefore, comparisons of the 2011 and the 2016 national surveys are most reliable. Various subgroups were identified for data analysis, of which the group of all participating hospitals, the representative sample as required by the ECDC, and the core group of hospitals participating both in 2011 and 2016 are of particular interest. When comparing prevalence data from 2011 and 2016, a significant decrease in AU prevalence was noted in the representative sample. However, AU prevalence remained at a rather constant level in the group of all participating hospitals. Moreover, it increased significantly in the core group. It is important to note that sampling for the representative sample was based solely on hospital size given in the German hospital register. Consequently, the representative sample is prone to random effects, which reduce the comparability of its data. The core group provides data for the most reliable comparison. However, it is possible that it does not accurately represent the overall situation in Germany. The 46 hospitals might have participated in the PPS again because of a higher than usual interest in surveillance and aspects of AMS or because they identified relevant problems in previous surveys. Patient demographics, such as the average age of patients hospitalized in Germany, remained at a constant level from 2011 to 2016.16 The average length of stay of patients in German hospitals has not changed substantially between 2011 and 2016 either,17 which increases the comparability of the 2011 and 2016 surveys. The high sensitivity and specificity in identifying patients with AU, which was shown by the validation study in 10 hospitals, corroborate the validity of the results on AU prevalence as assessed by the local data collection teams. The frequent parenteral application of antimicrobials that was observed in the current survey may be linked to the trend of patients being quickly discharged from hospitals after the conclusion of parenteral antimicrobial therapy.18 The most frequent indication for AU was the treatment of community-acquired infections. Roughly one-fifth of all administered antimicrobials were given to treat hospital-acquired infections, which highlights the importance of infection prevention in healthcare facilities. Regarding AU for SP, a prolonged administration of SP is still frequently observed in German hospitals, which illustrates the substantial differences between the reality of SP in German hospitals and nationally and internationally established guidelines.19,20 Nevertheless, a significant decrease in the frequency of SP3 among all SP was noted from 2011 to 2016. This development was possibly caused by improved awareness and implementation of guidelines as well as by intensified efforts in AMS in many hospitals in recent years. The documentation of the reason for AU in the patient charts has decreased significantly. The insufficient documentation corresponds with the relevant proportion of almost 5% of AU for which the data collection teams could not identify the reason for the AU even after consulting the ward staff. With regard to the antimicrobial groups and agents, one of the most remarkable results was the significant increase in the percentage of penicillins plus BLIs among all administered antimicrobials, which was almost twice as high in 2016 as in 2011. Penicillins plus BLIs seem to have replaced other frequently applied antimicrobial groups, such as second-generation cephalosporins, third-generation cephalosporins, fluoroquinolones and penicillins with extended spectrum. These antimicrobial groups cover a comparable range of pathogens and represented a significantly smaller proportion in 2016 than in 2011. Almost 50% of all AU could be attributed to the broad-spectrum antimicrobials: penicillins plus BLIs, fluoroquinolones, third-generation cephalosporins and carbapenems. The share of these antimicrobial groups among all AU has grown further since 2011. The significant increase in penicillins plus BLIs was largely due to the substantial and statistically significant increase in the use of piperacillin plus BLI (mostly tazobactam), which has almost tripled. Noticeably, the use of ampicillin plus BLI (usually sulbactam) has also grown significantly from 2011 to 2016. Both for community-acquired and hospital-acquired LRI, penicillins plus BLIs are the most commonly used antimicrobial group. Furthermore, penicillins plus BLIs represent the second most prevalent antimicrobial group for the treatment of UTI (for community-acquired and hospital-acquired UTI combined). The high percentage of anti-pseudomonal antimicrobials for community-acquired infections is surprising and goes counter to the German guidelines for the treatment of community-acquired LRI and UTI.21,22,Pseudomonas spp. are not routinely found as pathogens in community-acquired LRI and UTI in patients without corresponding risk factors. Moreover, the significant increase in the use of penicillins plus BLIs may also be linked to the reported increase in Escherichia coli that is resistant to third-generation cephalosporins in Germany.23 Carbapenems represent the fifth most frequently used antimicrobial group. An increase was observed from 2011 to 2016 for all carbapenems; in particular, the use of meropenem has grown significantly. A possible explanation can also be seen in the increase in resistance to third-generation cephalosporins.24 In times of emerging carbapenem-resistant organisms,25 the high consumption of carbapenems reveals significant potential for improvement and for reducing the use of broad-spectrum antimicrobials. The significant increase in the use of trimethoprim/sulfamethoxazole is owing to the fact that this agent was the most frequently used antimicrobial for MP. With a share of 6.8%,7 MP continues to be a relevant indication for AU in German acute care hospitals. The prevalence of patients with AU is rather low in Germany compared with other European countries.5 Nevertheless, the results of the PPS 2016 re-emphasize the areas for improvement that previous surveys already identified: the prolonged administration of SP, the insufficient documentation of indication, the relevant proportion of antimicrobials for which no indication can be identified (even after consulting the ward staff) and an increase in the use of broad-spectrum antimicrobials. Despite recent efforts to promote AMS in German hospitals,26 the results of the PPS 2016 do not reveal substantial measurable progress on the matter yet. It is therefore crucial to invest further in AMS education and training and in hospital-wide AMS activities.27 Acknowledgements Our thanks go to all participating hospitals for their dedication and willingness to participate in this survey and to the ECDC for initiating and supporting the PPS at the European level. We wish to give special thanks to all hospitals that participated in the validation study for welcoming the NRZ validation team and allowing us to visit their hospital sites. Furthermore, we thank Dr Janine Zweigner (Universitätsklinikum Köln), PD Dr Roland Schulze-Röbbecke (Universitätsklinikum Düsseldorf), Dr Gerhard Schwarzkopf-Steinhauser (Klinikum München), PD Dr Christian Brandt (Director of the Institute of Hygiene and Environmental Medicine of Vivantes Netzwerk für Gesundheit GmbH, Berlin, formerly Universitätsklinikum Frankfurt) and Dr Susanne Huggett (Asklepios Klinik Altona) for hosting 1-day PPS introductory courses at their institutions. Funding The German Ministry of Health funded the national PPS 2016. Transparency declarations None to declare. References 1 Ruden H, Gastmeier P, Daschner F et al.   [Nosocomial infections in Germany. Their epidemiology in old and new Federal Lander]. Dtsch Med Wochenschr  1996; 121: 1281– 7. Google Scholar CrossRef Search ADS PubMed  2 Hansen S, Sohr D, Piening B et al.   Antibiotic usage in German hospitals: results of the second national prevalence study. J Antimicrob Chemother  2013; 68: 2934– 9. Google Scholar CrossRef Search ADS PubMed  3 Behnke M, Hansen S, Leistner R et al.   Nosocomial infection and antibiotic use: a second national prevalence study in Germany. Dtsch Arztebl Int  2013; 110: 627– 33. Google Scholar PubMed  4 Bundesamt für Verbraucherschutz und Lebensmittelsicherheit, Paul-Ehrlich-Gesellschaft für Chemotherapie e.V., Infektiologie Freiburg. GERMAP 2012 - Bericht über den Antibiotikaverbrauch und die Verbreitung von Antibiotikaresistenzen in der Human- und Veterinärmedizin in Deutschland. Rheinbach, Germany: Antiinfectives Intelligence, 2014. 5 ECDC. Point Prevalence Survey of Healthcare-Associated Infections and Antimicrobial Use in European Acute Care Hospitals . Stockholm, Sweden: ECDC, 2013. 6 Walger P, Fätkenheuer G, Herrmann M et al.   Infektionsmedizin: klinische expertise fördern. Dtsch Arztebl International  2017; 114: A948– 50. 7 Nationales Referenzzentrum für die Surveillance von nosokomialen Infektionen. German National Point Prevalence Survey of Healthcare-Associated Infections and Antimicrobial Usage 2016, Final Report. http://www.nrz-hygiene.de/fileadmin/nrz/download/pps2016/PPS_2016_Abschlussbericht_20.07.2017.pdf. 8 ECDC. Point Prevalence Survey of Healthcare-Associated Infections and Antimicrobial Use in European Acute Care Hospitals—Protocol Version 5.3 . Stockholm, Sweden: ECDC, 2016. 9 Nationales Referenzzentrum für die Surveillance von nosokomialen Infektionen. Protocol of the German Point Prevalence Survey. http://www.nrz-hygiene.de/fileadmin/nrz/download/pps2016/EUPPS2016DE_Protokoll_Version_4.6.pdf. 10 Statistisches Bundesamt (Destatis). List of Hospitals and Rehabilitation Clinics in Germany. https://www.destatis.de/DE/Publikationen/Thematisch/Gesundheit/Krankenhaeuser/Krankenhausverzeichnis.html. 11 WHO Collaborating Centre for Drug Statistics Methodology. Guidelines for ATC Classification and DDD Assignment 2013. Oslo, Norway: WHO Collaborating Centre for Drug Statistics Methodology, 2012. https://www.whocc.no/filearchive/publications/1_2013guidelines.pdf. 12 German Recommendations of Good Epidemiological Practice. https://dgepi.de/fileadmin/pdf/GEP_LL_english_f.pdf. 13 German Protection Against Infection Act. http://www.gesetze-im-internet.de/ifsg/BJNR104510000.html. 14 Ruden H. Nosocomial Infections in Germany—Surveillance and Prevention. Band 56 der Schriftenreihe des Bundesministeriums für Gesundheit . Baden-Baden, Germany: Nomos Verlagsgesellschaft, 1995. 15 Hauer T, Lacour M, Gastmeier P et al.   [Nosocomial infections in Germany. Microbiological diagnosis, preventive antibiotics and antibiotic therapy]. Med Klin (Munich)  1996; 91: 681– 6. Google Scholar PubMed  16 Diagnosedaten der Patienten und Patientinnen in Krankenhäusern (einschließlich Sterbe- und Stundenfälle).  Wiesbaden, Germany: Statistisches Bundesamt, 2011. 17 Statistisches Bundesamt. Average Length of Stay in German Hospitals in the Years 1992 to 2015 (in days). https://de.statista.com/statistik/daten/studie/2604/umfrage/durchschnittliche-verweildauer-im-krankenhaus-seit-1992/. 18 Lee RW, Lindstrom ST. Early switch to oral antibiotics and early discharge guidelines in the management of community-acquired pneumonia. Respirology  2007; 12: 111– 6. Google Scholar CrossRef Search ADS PubMed  19 Allegranzi B, Zayed B, Bischoff P et al.   New WHO recommendations on intraoperative and postoperative measures for surgical site infection prevention: an evidence-based global perspective. Lancet Infect Dis  2016; 16: e288– 303. Google Scholar CrossRef Search ADS PubMed  20 Berrios-Torres SI, Umscheid CA, Bratzler DW et al.   Centers for Disease Control and Prevention guideline for the prevention of surgical site infection, 2017. JAMA Surg  2017; 152: 784– 91. Google Scholar CrossRef Search ADS PubMed  21 Wagenlehner F, Schmiemann G, Hoyme U et al.   Nationale S3-Leitlinie “Unkomplizierte Harnwegsinfektionen”. Der Urologe  2011; 50: 153– 69. Google Scholar CrossRef Search ADS PubMed  22 Ewig S, Höffken G, Kern W et al.   Behandlung von erwachsenen Patienten mit ambulant erworbener Pneumonie und Prävention–update 2016. Pneumologie  2016; 70: 151– 200. Google Scholar CrossRef Search ADS PubMed  23 Maechler F, Geffers C, Schwab F et al.   [Development of antimicrobial resistance in Germany: What is the current situation?]. Med Klin Intensivmed Notfmed  2017; 112: 186– 91. Google Scholar CrossRef Search ADS PubMed  24 Hamprecht A, Rohde AM, Behnke M et al.   Colonization with third-generation cephalosporin-resistant Enterobacteriaceae on hospital admission: prevalence and risk factors. J Antimicrob Chemother  2016; 71: 2957– 63. Google Scholar CrossRef Search ADS PubMed  25 Grundmann H, Glasner C, Albiger B et al.   Occurrence of carbapenemase-producing Klebsiella pneumoniae and Escherichia coli in the European survey of carbapenemase-producing Enterobacteriaceae (EuSCAPE): a prospective, multinational study. Lancet Infect Dis  2017; 17: 153– 63. Google Scholar CrossRef Search ADS PubMed  26 de With K, Allerberger F, Amann S et al.   Strategies to enhance rational use of antibiotics in hospital: a guideline by the German Society for Infectious Diseases. Infection  2016; 44: 395– 439. Google Scholar CrossRef Search ADS PubMed  27 Hansen S, Schwab F, Gropmann A et al.   Hygiene und sicherheitskultur in deutschen krankenhäusern. Bundesgesundheitsbl  2016; 59: 908– 15. Google Scholar CrossRef Search ADS   © The Author(s) 2018. Published by Oxford University Press on behalf of the British Society for Antimicrobial Chemotherapy. All rights reserved. For Permissions, please email: journals.permissions@oup.com.

Journal

Journal of Antimicrobial ChemotherapyOxford University Press

Published: Apr 1, 2018

There are no references for this article.

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off