- Email: [email protected]
Reducing Clostridium difficile through early identification of clusters and the use of a standardised set of interventions
Journal of Hospital Infection 75 (2010) 277–281
Available online at www.sciencedirect.com
Journal of Hospital Infection journal homepage: www.elsevierhealth.com/journals/jhin
Reducing Clostridium difficile through early identification of clusters and the use of a standardised set of interventions K.J. Hardy a, b, *, S. Gossain a, D. Thomlinson a, D.G. Pillay a, P.M. Hawkey a, b a b
West Midlands Public Health Laboratory, Heart of England NHS Foundation Trust, Birmingham, UK School of Infection and Immunity, University of Birmingham, Birmingham, UK
a r t i c l e i n f o
s u m m a r y
Article history: Received 1 October 2009 Accepted 7 December 2009 Available online 12 March 2010
In recent years the rates of Clostridium difficile infection (CDI) have increased worldwide with several large outbreaks occurring within the UK. New guidance from the UK Department of Health describes measures to investigate periods of increased incidence (PII) of CDI which include informing staff, ribotyping isolates, enhanced cleaning, audits and monitoring of antibiotic prescribing. This study aimed to determine whether a standardised set of measures could be used to control the incidence of CDI within an acute hospital setting over an 18 month period. During the study period a total of 102 PII involving 439 patients were investigated. The number of PII per month ranged from 14 in February 2008 to one in June 2009. From January 2008 to September 2008, ribotyping of patient isolates was only carried out on PII involving more than 10 patients, but from October 2008 it was carried out on all PII. During the period October 2008 to June 2009, 28 PII were investigated on 21 different wards, with seven wards having two PII. Ribotyping of the isolates confirmed nine (32%) of these PII to be outbreaks, with three being due to ribotype 027, two ribotype 078 and the others distinct ribotypes. Use of a set of standardised interventions has resulted in a decrease in the incidence of PII and a reduction in the number of patients involved. By taking early action with a set of standardised measures the incidence of hospital-acquired CDI can be reduced. Ó 2009 The Hospital Infection Society. Published by Elsevier Ltd. All rights reserved.
Keywords: Clostridium difficile Control Ribotyping
Introduction Clostridium difficile infection (CDI) is the leading cause of healthcare-associated diarrhoea in the UK and the rest of the developed world. From 2002 a steep rise in the rates of CDI was observed across Europe and the Americas, with more than 50 000 cases of CDI being reported in England in 2007.1 This was accompanied by an increase in reported mortality, with a 28% rise in the number of death certificates citing C. difficile from 2006 to 2007 in the UK.2 The rise in C. difficile infection and associated mortality has been partly attributed to the introduction of a hypervirulent clone of C. difficile known as ribotype 027 or NAP1. This ribotype has been widely spread within the UK, accounting for nearly 50% of all isolates typed by the national ribotyping
* Corresponding author. Address: West Midlands Public Health Laboratory, Health Protection Agency, Heart of England NHS Foundation Trust, Bordesley Green East, Birmingham B9 5SS, UK. Tel: þ44 121 4240250; fax: þ44 121 7726229. E-mail address: [email protected] (K.J. Hardy).
service (C. difficile ribotyping network: CDRN) and has been associated with outbreaks.3 Control of CDI is multifactorial and the reduction in CDI in the UK since 2007 has been due to the introduction of numerous infection control measures. Successful strategies have involved antimicrobial stewardship, isolation of symptomatic patients, improving cleanliness, and improvements in hand hygiene.4 Muto et al. have used a ‘bundle’ approach combining the abovementioned measures along with education and management teams to control an outbreak of ribotype 027.5 As in the study by Muto et al., the majority of published studies have described the control of CDI during outbreak situations.6,7 Ideally measures should be instituted before a large outbreak develops, and the trigger for these interventions could be early localised detection of periods of increased incidence. Recently published guidance from the Department of Health describes measures to investigate periods of increased incidence of CDI on a ward (which were developed at The Heart of England NHS Foundation Trust).8 We now describe how this standardised set of interventions can be used to control the incidence of CDI within an acute hospital setting.
0195-6701/$ – see front matter Ó 2009 The Hospital Infection Society. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.jhin.2009.12.004
278
K.J. Hardy et al. / Journal of Hospital Infection 75 (2010) 277–281
Methods Setting and definition of a period of increased incidence The Heart of England NHS Foundation Trust (HoEFT) is a large teaching hospital with a total of 1800 beds at three different sites. All cases of CDI occurring at HoEFT were defined as pre or post 48 h (those occurring <48 or >48 h after admission as defined by the Department of Health criteria). On a weekly basis all wards that had two or more post 48 h cases of C. difficile within a 28 day period were defined as having a period of increased incidence (PII). The Trust has antimicrobial guidelines for the empirical treatment of most common infections, with an emphasis on the use of narrow spectrum agents consistent with national guidance. Within the policies there is restriction on the use of cephalosporins and quinolones. There were no significant changes to antibiotic prescribing guidelines or compliance with these guidelines during the study period. The Trust has a C. difficile cohort ward which was opened in July 2007. Actions taken upon identification of a period of increased incidence Cases of CDI notified to the infection control service were reviewed at an infection control weekly operational meeting attended by the infection control nurses (ICNs), medical microbiologists, data analyst and clinical scientist. Following the confirmation of a PII, a standardised set of interventions was introduced. The first of these was notification of the PII to the matron, ward manager, directorate manager and clinical lead for the ward. A weekly C. difficile audit was carried out by the ICNs on each ward with a PII. The audit aimed to identify any areas of poor practice and covered the following: (i) compliance with hand hygiene; (ii) adequacy of environmental decontamination in sluice, bathrooms and toilets, manual handling equipment; (iii) compliance with isolation and clinical management of patients with confirmed or suspected CDI, including reviewing and stopping or changing antibiotic therapy as necessary. If the audit identified failure to adequately clean commodes or other toileting equipment, this automatically resulted in a 10% deduction from their audit score along with rectification. The wards with a PII were audited weekly until they had achieved three consecutive passes (audit score >90%) and there had been no further cases of post 48 h CDI. All audit results were discussed with the matron and ward manager for the area with training given where necessary to facilitate rectification of the issues. During the period in which a ward remained on weekly audits, the whole ward was cleaned with Chlorclean (1000 ppm available chlorine), with each bed space being cleaned with a separate cloth. All symptomatic C. difficile patients were either nursed within a side room or were transferred to the CDI cohort ward. Total antibiotic usage, measured in defined daily dose per 100 bed-days, within the Trust was monitored. For the period January 2008 to September 2008 all of the above measures were implemented, with PCR ribotyping of isolates being carried out on those PII with more than 10 cases. From October 2008 to July 2009 in addition to the measures detailed, isolates from all PII were ribotyped (dependent on the isolate being grown). Isolates were typed using PCR ribotyping as previously described.9 A PII was classified as an outbreak of CDI if there were two or more cases of the same PCR ribotype within a 28 day period. All confirmed outbreaks were reported as serious untoward incidents to the Primary Care Trust and Strategic Health Authority. All PII were reviewed on a weekly basis at the infection control operational meeting and incident meetings held as appropriate. For those wards that repeatedly failed the C. difficile audit and had additional post 48 h cases, an additional, more detailed
environmental audit was completed by one of the two head nurses responsible for nursing across the whole Trust. This audit concentrated on the built environment and equipment (e.g. state of repair of the ward, beds and medical equipment). Repair or replacement of equipment/wards arose from this audit. Data analysis For data analysis the PII was assigned to the month in which the second post 48 h case on that ward occurred. All of the C. difficile cases that occurred within 28 days of the previous case were classified as belonging to the same PII. If additional C. difficile cases occurred more than 28 days after the previous case, this was defined as a new PII. Results Overall From January 2008 to June 2009 the Trust had 584 post 48 h C. difficile cases, of which 439 (75%) were associated with PII. A total of 102 PII were investigated. The number of PII investigated per month decreased from a peak of 14 per month in February 2008 to one in June 2009 (Figure 1). Seven PII from January 2008 to September 2008 had more than 10 cases and five of these were confirmed as outbreaks of ribotype 027 and one as an outbreak of ribotype 106. The isolates from the seventh PII had been inadvertently discarded before typing was complete. October 2008 to June 2009 During the period October 2008 to June 2009 the Trust had 157 post 48 h cases of C. difficile, of which 79 (50.3%) were associated with PII. Twenty-eight PII were investigated on 21 different wards, with seven wards having two PII. Ribotyping of the isolates confirmed nine (32%) of these PII to be outbreaks, with three being due to ribotype 027, two ribotype 078 and all the others distinct ribotypes (Figure 2). A total of 11 different ribotypes were identified on wards that were not confirmed as outbreaks, with the most predominant types being 027 (seven), 020 (four), 014 (three) and 015 (three) and 023 (three). The mean number of patients involved in the PII confirmed as outbreaks was 3.3 with a mean of 2.6 in the non-outbreak PII. The mean number of weeks that wards remained on the audit was greater for the wards where outbreaks were confirmed; 10 weeks as opposed to 8 weeks for those PII which were not confirmed as outbreaks. The longest duration a ward remained on the audit was 16 weeks. Only two wards, one of which was confirmed as having an outbreak, passed the weekly audit in the minimum of three weeks. The most common reason for failure of the audit was dirty commodes. Nine out of the 28 wards (32%) had further C. difficile cases once a PII had been identified and measures implemented. A total of 57 of the 157 (36%) C. difficile positive patients were transferred to the cohort ward, the remaining patients either being nursed within a side room or being asymptomatic by the time the result was received. The overall antibiotic usage from April 2008 to June 2009 did not show any marked reductions (Figure 3). Discussion Worldwide, the rates of C. difficile have been increasing over the last decade, with numerous large scale outbreaks occurring. This study demonstrates that by identifying local (ward-based) increased incidences of C. difficile early and instituting a standardised
K.J. Hardy et al. / Journal of Hospital Infection 75 (2010) 277–281
279
16
120
14
100
80
10 8
60
6
Number of PIIs
Number of patients
12
40 4 20
Jun 09
May 09
Apr 09
Mar 09
Feb 09
Jan 09
Dec 08
Nov 08
Oct 08
Sep 08
Aug 08
Jul 08
Jun 08
May 08
Apr 08
Mar 08
Feb 08
Jan 08
0
2 0
Month Figure 1. Total number of periods of increased incidence (PII) recognised per month (solid line) and the corresponding number of patients with C. difficile (dashed line).
set of measures, control can be achieved. The measures introduced resulted in a reduction in both the number of PII and the number of patients involved in each PII. It is well recognised that control of CDI requires a multifactorial approach. There is the need first to ensure that patients are not predisposed to CDI by the use of antibiotic stewardship programmes and second to interrupt the transmission process using good infection control procedures.10 The relative importance of each of these factors in the control of C. difficile has been widely debated, with some claiming control of CDI with antibiotic stewardship alone, whereas others state that infection control measures alone can reduce the incidence.11,12
Many of the studies claiming to achieve a reduction in CDI through good antibiotic stewardship alone have been retrospective and have not controlled for changes in infection control procedures that were implemented during the outbreak.6,13 One prospective study did demonstrate that the introduction of a narrow spectrum antibiotic policy and feedback led to a reduction in the incidence of CDI, but this was against a background of good isolation and infection control practice.14 A time-interrupted study by Valiquette et al. during an outbreak of ribotype 027 saw no changes in the rate of CDI after the implementation of infection control procedures, but saw a reduction after a modification in
6
5 078
078
Number of PII
4 027
027
3 003 2 171
027
029
1 052 0
Oct 08
Nov 08
Dec 08
Jan 09
Feb 09 Month
Mar 09
Apr 09
May 09
Jun 09
Figure 2. Number of periods of increased incidence (PII) identified per month. White bars: PII, confirmed not to be an outbreak; black bars: PII, confirmed to be an outbreak with corresponding ribotype.
280
K.J. Hardy et al. / Journal of Hospital Infection 75 (2010) 277–281
100
DDD/100 bed days
80
60
40
Jun 09
May 09
Apr 09
Mar 09
Feb 09
Jan 09
Dec 08
Nov 08
Oct 08
Sep 08
Aug 08
Jul 08
Jun 08
May 08
0
Apr 08
20
Figure 3. The total antibiotic consumption as defined daily dose (DDD) for the Trust from April 2008 to June 2009. Total DDD per 100 bed-days (dashed line: mean).
antibiotic prescribing.15 However, the authors showed no data on compliance with infection control precautions during the study and the decrease may have partly been due to either improved infection control measures or an accumulative effect of the introduction of improved control. The majority of studies describing control of CDI have used a multifactorial approach combining both prudent antimicrobial prescribing and infection control measures, with the majority in recent years describing the control of outbreaks due to ribotype 027.3,5 Unlike the current study which used a set of standardised interventions to prevent increased incidences of CDI becoming outbreaks, Muto et al. used a ‘bundle’ consisting of early casefinding, infection control measures, a CDI management team and monitoring of antimicrobial usage to control an outbreak of ribotype 027. Although the authors were unable to ascertain which part of the ‘bundle’ contributed to the decrease in CDI, they felt that control was not achieved due to the control of antimicrobial usage alone. Salgado et al. have similarly described sustained control of an outbreak with only a change in infection control measures.7 As in the current study, the measures were aimed at reducing the transmission of C. difficile through cleaning with a chlorine-based product, soap and water for hand washing and cohort or isolation nursing of patients. One of the major differences between previously described studies and the current study is that we aimed to pre-empt and prevent outbreaks of CDI from becoming established, as opposed to being reactive and trying to control CDI once an outbreak was evident. The early identification and notification of PII enabled actions to be prompt and targeted. The initiation of cleaning with a chlorine-based agent ensured that C. difficile was removed from the environment, thereby preventing onward transmission. The audits were invaluable in establishing areas of poor practice that may have contributed to transmission, and continuous monitoring ensured that these practices were rectified and improvements sustained. The wards that were confirmed as having outbreaks of CDI had a greater number of failed audits, and this may therefore imply that transmission of CDI occurred on these wards due to poor infection control practices.
This study focused on enhanced interventions on localised wards, in addition to the hospitalwide infection control measures. By concentrating on selected PII wards the potential environmental sources of CDI transmission to the rest of the hospital was reduced. Ribotype 027 is widespread throughout North America and Europe and was the predominant cause of outbreaks in this study. This study has used PCR ribotyping for defining and controlling outbreaks, in contrast to previous studies that have described management of outbreaks due to ribotype 027.5,6 Although ribotype 027 was predominant during the initial phase of the study and remained within our hospital for the entire period, it was not the only cause of CDI outbreaks, with six other ribotypes also causing outbreaks. This is similar to findings by Belmare et al. who demonstrated that during a period of 10 years several different restriction endonuclease analysis (REA) types were responsible for large clusters.16 This study demonstrates that if early action is taken using a standardised set of measures in accordance with the Department of Health (UK) guidance, periods of increased incidence of C. difficile can be controlled. CDI is a growing problem in healthcare facilities in both Europe and North America, and the application of the simple measures described in this paper could have a significant impact on the mortality and morbidity associated with CDI. Acknowledgements We thank all members of the infection control nursing team who completed the weekly audits and the laboratory staff for PCR ribotyping of isolates. Conflict of interest statement None declared. Funding source The study was funded by the West Midlands Public Health Laboratory.
K.J. Hardy et al. / Journal of Hospital Infection 75 (2010) 277–281
References 1. Redelings MD, Sorvillo F, Mascola L. Increase in Clostridium difficile-related mortality rates, United States, 1999–2004. Emerg Infect Dis 2007;13: 1417–1419. 2. Deaths involving Clostridium difficile; England and Wales, 2003–07. Health Stat Q 2008;39:67–76. 3. Weiss K, Boisvert A, Chagnon M, et al. Multipronged intervention strategy to control an outbreak of Clostridium difficile infection (CDI) and its impact on the rates of CDI from 2002 to 2007. Infect Control Hosp Epidemiol 2009; 30:156–162. 4. Gerding DN, Muto CA, Owens Jr RC. Measures to control and prevent Clostridium difficile infection. Clin Infect Dis 2008;46(Suppl. 1):S43–S49. 5. Muto CA, Blank MK, Marsh JW, et al. Control of an outbreak of infection with the hypervirulent Clostridium difficile BI strain in a university hospital using a comprehensive ‘‘bundle’’ approach. Clin Infect Dis 2007;45:1266–1273. 6. Debast SB, Vaessen N, Choudry A, Wiegers-Ligtvoet EA, van den Berg RJ, Kuijper EJ. Successful combat of an outbreak due to Clostridium difficile PCR ribotype 027 and recognition of specific risk factors. Clin Microbiol Infect 2009;15:427–434. 7. Salgado CD, Mauldin PD, Fogle PJ, Bosso JA. Analysis of an outbreak of Clostridium difficile infection controlled with enhanced infection control measures. Am J Infect Control 2009;37:458–464. 8. Department of Health Working Party. Clostridium difficile infection: how to deal with the problem. London: DoH; 2009.
281
9. Stubbs SL, Brazier JS, O’Neill GL, Duerden BI. PCR targeted to the 16S–23S rRNA gene intergenic spacer region of Clostridium difficile and construction of a library consisting of 116 different PCR ribotypes. J Clin Microbiol 1999;37:461–463. 10. Vonberg RP, Kuijper EJ, Wilcox MH, et al. Infection control measures to limit the spread of Clostridium difficile. Clin Microbiol Infect 2008;14(Suppl. 5):2–20. 11. Beaulieu M, Thirion DJ, Williamson D, Pichette G. Clostridium difficile-associated diarrhea outbreaks: the name of the game is isolation and cleaning. Clin Infect Dis 2006;42:725–729. 12. Pear SM, Williamson TH, Bettin KM, Gerding DN, Galgiani JN. Decrease in nosocomial Clostridium difficile-associated diarrhea by restricting clindamycin use. Ann Intern Med 1994;120:272–277. 13. Pepin J, Saheb N, Coulombe MA, et al. Emergence of fluoroquinolones as the predominant risk factor for Clostridium difficile-associated diarrhea: a cohort study during an epidemic in Quebec. Clin Infect Dis 2005;41: 1254–1260. 14. Fowler S, Webber A, Cooper BS, et al. Successful use of feedback to improve antibiotic prescribing and reduce Clostridium difficile infection: a controlled interrupted time series. J Antimicrob Chemother 2007;59:990–995. 15. Valiquette L, Cossette B, Garant MP, Diab H, Pepin J. Impact of a reduction in the use of high-risk antibiotics on the course of an epidemic of Clostridium difficileassociated disease caused by the hypervirulent NAP1/027 strain. Clin Infect Dis 2007;45(Suppl. 2):S112–S121. 16. Belmares J, Johnson S, Parada JP, et al. Molecular epidemiology of Clostridium difficile over the course of 10 years in a tertiary care hospital. Clin Infect Dis 2009;49:1141–1147.
Available online at www.sciencedirect.com
Journal of Hospital Infection journal homepage: www.elsevierhealth.com/journals/jhin
Reducing Clostridium difficile through early identification of clusters and the use of a standardised set of interventions K.J. Hardy a, b, *, S. Gossain a, D. Thomlinson a, D.G. Pillay a, P.M. Hawkey a, b a b
West Midlands Public Health Laboratory, Heart of England NHS Foundation Trust, Birmingham, UK School of Infection and Immunity, University of Birmingham, Birmingham, UK
a r t i c l e i n f o
s u m m a r y
Article history: Received 1 October 2009 Accepted 7 December 2009 Available online 12 March 2010
In recent years the rates of Clostridium difficile infection (CDI) have increased worldwide with several large outbreaks occurring within the UK. New guidance from the UK Department of Health describes measures to investigate periods of increased incidence (PII) of CDI which include informing staff, ribotyping isolates, enhanced cleaning, audits and monitoring of antibiotic prescribing. This study aimed to determine whether a standardised set of measures could be used to control the incidence of CDI within an acute hospital setting over an 18 month period. During the study period a total of 102 PII involving 439 patients were investigated. The number of PII per month ranged from 14 in February 2008 to one in June 2009. From January 2008 to September 2008, ribotyping of patient isolates was only carried out on PII involving more than 10 patients, but from October 2008 it was carried out on all PII. During the period October 2008 to June 2009, 28 PII were investigated on 21 different wards, with seven wards having two PII. Ribotyping of the isolates confirmed nine (32%) of these PII to be outbreaks, with three being due to ribotype 027, two ribotype 078 and the others distinct ribotypes. Use of a set of standardised interventions has resulted in a decrease in the incidence of PII and a reduction in the number of patients involved. By taking early action with a set of standardised measures the incidence of hospital-acquired CDI can be reduced. Ó 2009 The Hospital Infection Society. Published by Elsevier Ltd. All rights reserved.
Keywords: Clostridium difficile Control Ribotyping
Introduction Clostridium difficile infection (CDI) is the leading cause of healthcare-associated diarrhoea in the UK and the rest of the developed world. From 2002 a steep rise in the rates of CDI was observed across Europe and the Americas, with more than 50 000 cases of CDI being reported in England in 2007.1 This was accompanied by an increase in reported mortality, with a 28% rise in the number of death certificates citing C. difficile from 2006 to 2007 in the UK.2 The rise in C. difficile infection and associated mortality has been partly attributed to the introduction of a hypervirulent clone of C. difficile known as ribotype 027 or NAP1. This ribotype has been widely spread within the UK, accounting for nearly 50% of all isolates typed by the national ribotyping
* Corresponding author. Address: West Midlands Public Health Laboratory, Health Protection Agency, Heart of England NHS Foundation Trust, Bordesley Green East, Birmingham B9 5SS, UK. Tel: þ44 121 4240250; fax: þ44 121 7726229. E-mail address: [email protected] (K.J. Hardy).
service (C. difficile ribotyping network: CDRN) and has been associated with outbreaks.3 Control of CDI is multifactorial and the reduction in CDI in the UK since 2007 has been due to the introduction of numerous infection control measures. Successful strategies have involved antimicrobial stewardship, isolation of symptomatic patients, improving cleanliness, and improvements in hand hygiene.4 Muto et al. have used a ‘bundle’ approach combining the abovementioned measures along with education and management teams to control an outbreak of ribotype 027.5 As in the study by Muto et al., the majority of published studies have described the control of CDI during outbreak situations.6,7 Ideally measures should be instituted before a large outbreak develops, and the trigger for these interventions could be early localised detection of periods of increased incidence. Recently published guidance from the Department of Health describes measures to investigate periods of increased incidence of CDI on a ward (which were developed at The Heart of England NHS Foundation Trust).8 We now describe how this standardised set of interventions can be used to control the incidence of CDI within an acute hospital setting.
0195-6701/$ – see front matter Ó 2009 The Hospital Infection Society. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.jhin.2009.12.004
278
K.J. Hardy et al. / Journal of Hospital Infection 75 (2010) 277–281
Methods Setting and definition of a period of increased incidence The Heart of England NHS Foundation Trust (HoEFT) is a large teaching hospital with a total of 1800 beds at three different sites. All cases of CDI occurring at HoEFT were defined as pre or post 48 h (those occurring <48 or >48 h after admission as defined by the Department of Health criteria). On a weekly basis all wards that had two or more post 48 h cases of C. difficile within a 28 day period were defined as having a period of increased incidence (PII). The Trust has antimicrobial guidelines for the empirical treatment of most common infections, with an emphasis on the use of narrow spectrum agents consistent with national guidance. Within the policies there is restriction on the use of cephalosporins and quinolones. There were no significant changes to antibiotic prescribing guidelines or compliance with these guidelines during the study period. The Trust has a C. difficile cohort ward which was opened in July 2007. Actions taken upon identification of a period of increased incidence Cases of CDI notified to the infection control service were reviewed at an infection control weekly operational meeting attended by the infection control nurses (ICNs), medical microbiologists, data analyst and clinical scientist. Following the confirmation of a PII, a standardised set of interventions was introduced. The first of these was notification of the PII to the matron, ward manager, directorate manager and clinical lead for the ward. A weekly C. difficile audit was carried out by the ICNs on each ward with a PII. The audit aimed to identify any areas of poor practice and covered the following: (i) compliance with hand hygiene; (ii) adequacy of environmental decontamination in sluice, bathrooms and toilets, manual handling equipment; (iii) compliance with isolation and clinical management of patients with confirmed or suspected CDI, including reviewing and stopping or changing antibiotic therapy as necessary. If the audit identified failure to adequately clean commodes or other toileting equipment, this automatically resulted in a 10% deduction from their audit score along with rectification. The wards with a PII were audited weekly until they had achieved three consecutive passes (audit score >90%) and there had been no further cases of post 48 h CDI. All audit results were discussed with the matron and ward manager for the area with training given where necessary to facilitate rectification of the issues. During the period in which a ward remained on weekly audits, the whole ward was cleaned with Chlorclean (1000 ppm available chlorine), with each bed space being cleaned with a separate cloth. All symptomatic C. difficile patients were either nursed within a side room or were transferred to the CDI cohort ward. Total antibiotic usage, measured in defined daily dose per 100 bed-days, within the Trust was monitored. For the period January 2008 to September 2008 all of the above measures were implemented, with PCR ribotyping of isolates being carried out on those PII with more than 10 cases. From October 2008 to July 2009 in addition to the measures detailed, isolates from all PII were ribotyped (dependent on the isolate being grown). Isolates were typed using PCR ribotyping as previously described.9 A PII was classified as an outbreak of CDI if there were two or more cases of the same PCR ribotype within a 28 day period. All confirmed outbreaks were reported as serious untoward incidents to the Primary Care Trust and Strategic Health Authority. All PII were reviewed on a weekly basis at the infection control operational meeting and incident meetings held as appropriate. For those wards that repeatedly failed the C. difficile audit and had additional post 48 h cases, an additional, more detailed
environmental audit was completed by one of the two head nurses responsible for nursing across the whole Trust. This audit concentrated on the built environment and equipment (e.g. state of repair of the ward, beds and medical equipment). Repair or replacement of equipment/wards arose from this audit. Data analysis For data analysis the PII was assigned to the month in which the second post 48 h case on that ward occurred. All of the C. difficile cases that occurred within 28 days of the previous case were classified as belonging to the same PII. If additional C. difficile cases occurred more than 28 days after the previous case, this was defined as a new PII. Results Overall From January 2008 to June 2009 the Trust had 584 post 48 h C. difficile cases, of which 439 (75%) were associated with PII. A total of 102 PII were investigated. The number of PII investigated per month decreased from a peak of 14 per month in February 2008 to one in June 2009 (Figure 1). Seven PII from January 2008 to September 2008 had more than 10 cases and five of these were confirmed as outbreaks of ribotype 027 and one as an outbreak of ribotype 106. The isolates from the seventh PII had been inadvertently discarded before typing was complete. October 2008 to June 2009 During the period October 2008 to June 2009 the Trust had 157 post 48 h cases of C. difficile, of which 79 (50.3%) were associated with PII. Twenty-eight PII were investigated on 21 different wards, with seven wards having two PII. Ribotyping of the isolates confirmed nine (32%) of these PII to be outbreaks, with three being due to ribotype 027, two ribotype 078 and all the others distinct ribotypes (Figure 2). A total of 11 different ribotypes were identified on wards that were not confirmed as outbreaks, with the most predominant types being 027 (seven), 020 (four), 014 (three) and 015 (three) and 023 (three). The mean number of patients involved in the PII confirmed as outbreaks was 3.3 with a mean of 2.6 in the non-outbreak PII. The mean number of weeks that wards remained on the audit was greater for the wards where outbreaks were confirmed; 10 weeks as opposed to 8 weeks for those PII which were not confirmed as outbreaks. The longest duration a ward remained on the audit was 16 weeks. Only two wards, one of which was confirmed as having an outbreak, passed the weekly audit in the minimum of three weeks. The most common reason for failure of the audit was dirty commodes. Nine out of the 28 wards (32%) had further C. difficile cases once a PII had been identified and measures implemented. A total of 57 of the 157 (36%) C. difficile positive patients were transferred to the cohort ward, the remaining patients either being nursed within a side room or being asymptomatic by the time the result was received. The overall antibiotic usage from April 2008 to June 2009 did not show any marked reductions (Figure 3). Discussion Worldwide, the rates of C. difficile have been increasing over the last decade, with numerous large scale outbreaks occurring. This study demonstrates that by identifying local (ward-based) increased incidences of C. difficile early and instituting a standardised
K.J. Hardy et al. / Journal of Hospital Infection 75 (2010) 277–281
279
16
120
14
100
80
10 8
60
6
Number of PIIs
Number of patients
12
40 4 20
Jun 09
May 09
Apr 09
Mar 09
Feb 09
Jan 09
Dec 08
Nov 08
Oct 08
Sep 08
Aug 08
Jul 08
Jun 08
May 08
Apr 08
Mar 08
Feb 08
Jan 08
0
2 0
Month Figure 1. Total number of periods of increased incidence (PII) recognised per month (solid line) and the corresponding number of patients with C. difficile (dashed line).
set of measures, control can be achieved. The measures introduced resulted in a reduction in both the number of PII and the number of patients involved in each PII. It is well recognised that control of CDI requires a multifactorial approach. There is the need first to ensure that patients are not predisposed to CDI by the use of antibiotic stewardship programmes and second to interrupt the transmission process using good infection control procedures.10 The relative importance of each of these factors in the control of C. difficile has been widely debated, with some claiming control of CDI with antibiotic stewardship alone, whereas others state that infection control measures alone can reduce the incidence.11,12
Many of the studies claiming to achieve a reduction in CDI through good antibiotic stewardship alone have been retrospective and have not controlled for changes in infection control procedures that were implemented during the outbreak.6,13 One prospective study did demonstrate that the introduction of a narrow spectrum antibiotic policy and feedback led to a reduction in the incidence of CDI, but this was against a background of good isolation and infection control practice.14 A time-interrupted study by Valiquette et al. during an outbreak of ribotype 027 saw no changes in the rate of CDI after the implementation of infection control procedures, but saw a reduction after a modification in
6
5 078
078
Number of PII
4 027
027
3 003 2 171
027
029
1 052 0
Oct 08
Nov 08
Dec 08
Jan 09
Feb 09 Month
Mar 09
Apr 09
May 09
Jun 09
Figure 2. Number of periods of increased incidence (PII) identified per month. White bars: PII, confirmed not to be an outbreak; black bars: PII, confirmed to be an outbreak with corresponding ribotype.
280
K.J. Hardy et al. / Journal of Hospital Infection 75 (2010) 277–281
100
DDD/100 bed days
80
60
40
Jun 09
May 09
Apr 09
Mar 09
Feb 09
Jan 09
Dec 08
Nov 08
Oct 08
Sep 08
Aug 08
Jul 08
Jun 08
May 08
0
Apr 08
20
Figure 3. The total antibiotic consumption as defined daily dose (DDD) for the Trust from April 2008 to June 2009. Total DDD per 100 bed-days (dashed line: mean).
antibiotic prescribing.15 However, the authors showed no data on compliance with infection control precautions during the study and the decrease may have partly been due to either improved infection control measures or an accumulative effect of the introduction of improved control. The majority of studies describing control of CDI have used a multifactorial approach combining both prudent antimicrobial prescribing and infection control measures, with the majority in recent years describing the control of outbreaks due to ribotype 027.3,5 Unlike the current study which used a set of standardised interventions to prevent increased incidences of CDI becoming outbreaks, Muto et al. used a ‘bundle’ consisting of early casefinding, infection control measures, a CDI management team and monitoring of antimicrobial usage to control an outbreak of ribotype 027. Although the authors were unable to ascertain which part of the ‘bundle’ contributed to the decrease in CDI, they felt that control was not achieved due to the control of antimicrobial usage alone. Salgado et al. have similarly described sustained control of an outbreak with only a change in infection control measures.7 As in the current study, the measures were aimed at reducing the transmission of C. difficile through cleaning with a chlorine-based product, soap and water for hand washing and cohort or isolation nursing of patients. One of the major differences between previously described studies and the current study is that we aimed to pre-empt and prevent outbreaks of CDI from becoming established, as opposed to being reactive and trying to control CDI once an outbreak was evident. The early identification and notification of PII enabled actions to be prompt and targeted. The initiation of cleaning with a chlorine-based agent ensured that C. difficile was removed from the environment, thereby preventing onward transmission. The audits were invaluable in establishing areas of poor practice that may have contributed to transmission, and continuous monitoring ensured that these practices were rectified and improvements sustained. The wards that were confirmed as having outbreaks of CDI had a greater number of failed audits, and this may therefore imply that transmission of CDI occurred on these wards due to poor infection control practices.
This study focused on enhanced interventions on localised wards, in addition to the hospitalwide infection control measures. By concentrating on selected PII wards the potential environmental sources of CDI transmission to the rest of the hospital was reduced. Ribotype 027 is widespread throughout North America and Europe and was the predominant cause of outbreaks in this study. This study has used PCR ribotyping for defining and controlling outbreaks, in contrast to previous studies that have described management of outbreaks due to ribotype 027.5,6 Although ribotype 027 was predominant during the initial phase of the study and remained within our hospital for the entire period, it was not the only cause of CDI outbreaks, with six other ribotypes also causing outbreaks. This is similar to findings by Belmare et al. who demonstrated that during a period of 10 years several different restriction endonuclease analysis (REA) types were responsible for large clusters.16 This study demonstrates that if early action is taken using a standardised set of measures in accordance with the Department of Health (UK) guidance, periods of increased incidence of C. difficile can be controlled. CDI is a growing problem in healthcare facilities in both Europe and North America, and the application of the simple measures described in this paper could have a significant impact on the mortality and morbidity associated with CDI. Acknowledgements We thank all members of the infection control nursing team who completed the weekly audits and the laboratory staff for PCR ribotyping of isolates. Conflict of interest statement None declared. Funding source The study was funded by the West Midlands Public Health Laboratory.
K.J. Hardy et al. / Journal of Hospital Infection 75 (2010) 277–281
References 1. Redelings MD, Sorvillo F, Mascola L. Increase in Clostridium difficile-related mortality rates, United States, 1999–2004. Emerg Infect Dis 2007;13: 1417–1419. 2. Deaths involving Clostridium difficile; England and Wales, 2003–07. Health Stat Q 2008;39:67–76. 3. Weiss K, Boisvert A, Chagnon M, et al. Multipronged intervention strategy to control an outbreak of Clostridium difficile infection (CDI) and its impact on the rates of CDI from 2002 to 2007. Infect Control Hosp Epidemiol 2009; 30:156–162. 4. Gerding DN, Muto CA, Owens Jr RC. Measures to control and prevent Clostridium difficile infection. Clin Infect Dis 2008;46(Suppl. 1):S43–S49. 5. Muto CA, Blank MK, Marsh JW, et al. Control of an outbreak of infection with the hypervirulent Clostridium difficile BI strain in a university hospital using a comprehensive ‘‘bundle’’ approach. Clin Infect Dis 2007;45:1266–1273. 6. Debast SB, Vaessen N, Choudry A, Wiegers-Ligtvoet EA, van den Berg RJ, Kuijper EJ. Successful combat of an outbreak due to Clostridium difficile PCR ribotype 027 and recognition of specific risk factors. Clin Microbiol Infect 2009;15:427–434. 7. Salgado CD, Mauldin PD, Fogle PJ, Bosso JA. Analysis of an outbreak of Clostridium difficile infection controlled with enhanced infection control measures. Am J Infect Control 2009;37:458–464. 8. Department of Health Working Party. Clostridium difficile infection: how to deal with the problem. London: DoH; 2009.
281
9. Stubbs SL, Brazier JS, O’Neill GL, Duerden BI. PCR targeted to the 16S–23S rRNA gene intergenic spacer region of Clostridium difficile and construction of a library consisting of 116 different PCR ribotypes. J Clin Microbiol 1999;37:461–463. 10. Vonberg RP, Kuijper EJ, Wilcox MH, et al. Infection control measures to limit the spread of Clostridium difficile. Clin Microbiol Infect 2008;14(Suppl. 5):2–20. 11. Beaulieu M, Thirion DJ, Williamson D, Pichette G. Clostridium difficile-associated diarrhea outbreaks: the name of the game is isolation and cleaning. Clin Infect Dis 2006;42:725–729. 12. Pear SM, Williamson TH, Bettin KM, Gerding DN, Galgiani JN. Decrease in nosocomial Clostridium difficile-associated diarrhea by restricting clindamycin use. Ann Intern Med 1994;120:272–277. 13. Pepin J, Saheb N, Coulombe MA, et al. Emergence of fluoroquinolones as the predominant risk factor for Clostridium difficile-associated diarrhea: a cohort study during an epidemic in Quebec. Clin Infect Dis 2005;41: 1254–1260. 14. Fowler S, Webber A, Cooper BS, et al. Successful use of feedback to improve antibiotic prescribing and reduce Clostridium difficile infection: a controlled interrupted time series. J Antimicrob Chemother 2007;59:990–995. 15. Valiquette L, Cossette B, Garant MP, Diab H, Pepin J. Impact of a reduction in the use of high-risk antibiotics on the course of an epidemic of Clostridium difficileassociated disease caused by the hypervirulent NAP1/027 strain. Clin Infect Dis 2007;45(Suppl. 2):S112–S121. 16. Belmares J, Johnson S, Parada JP, et al. Molecular epidemiology of Clostridium difficile over the course of 10 years in a tertiary care hospital. Clin Infect Dis 2009;49:1141–1147.