Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 7-Day Trial for You or Your Team.

Learn More →

Increasing rates in Clostridium difficile infection (CDI) among hospitalised patients, Spain 1999-2007

Increasing rates in Clostridium difficile infection (CDI) among hospitalised patients, Spain... Resea rc h a r ti cl es I n c r e a s I n g r a t e s I n C l o s t r i d i u m d i f f i C i l e I n f e c t I o n ( c D I) a m o n g h o s p I ta l I s e D p a t I e n t s , s p a I n 1999-2007 1 2 3 4 5 A Asensio ([email protected]) , J Vaque-Rafart , F Calbo-Torrecillas , J J Gestal-Otero , F López-Fernández , 6 7 A Trilla-Garcia , R Canton , EPINE Working Group 1. Department of Preventive Medicine, Hospital Universitario Puerta de Hierro. Madrid, Spain 2. Department of Preventive Medicine, Hospital Vall d´Hebron. Barcelona, Spain 3. Department of Preventive Medicine, Hospital Carlos Haya. Málaga, Spain 4. Department of Preventive Medicine and Public Health, University of Santiago de Compostela, Spain 5. Department of Preventive Medicine, Hospital Universitario Puerto del Mar, Cádiz, Spain 6. Epidemiologi Unit. Hospital Clinic. Barcelona, Spain 7. Department of Microbiolog y. Hospital Universitario Ramón y Cajal. Madrid, Spain Limited information is available on the burden and epidemiology Each year in May, acute care hospitals in Spain are requested to of Clostridium difficile infection (CDI) in Spain. The present report voluntarily join the EPINE prevalence study. Participating hospitals communicates the secular trends in prevalence of CDI among fill a standardised questionnaire on each hospitalised patient as hospitalised patients in Spain from 1999 through 2007. Data well as overall data on the hospital and the hospital’s wards. were obtained through the EPINE study (Estudio de prevalencia de CDI diagnosis relies on CDC case-definitions for nosocomial las infecciones nosocomiales en los hospitales españoles), a point infections (note: the EU case definitions were not available at prevalence study series of nosocomial infections among patients the time of the start of the study), and includes cases of either clinical diarrhoea or toxic megacolon with laboratory evidence of admitted to hospital in Spain. A total of 378 cases with CDI were identified. Median age was 74 positive stool culture and/or toxin assay for C. difficile. Thus our analysis encompassed a symptomatic population with a positive years. Prevalence rates of CDI increased from 3.9 to 12.2 cases per 10,000 hospitalised patients and showed a significantly increasing microbiological confirmation of CDI by culture, toxin assay or both. secular trend from 1999 through 2007 (prevalence rate ratio per each year increment 1.09; 95% CI 1.05 – 1.14). Percentage of In addition to information on nosocomial infections, the patient hospitalised patients receiving antimicrobials increased linearly forms collected from the hospitals included demographic data (age from 36.0% in 1999 to 40.7% in 2007 (p <0.001) and was and gender); information on underlying clinical conditions such strongly correlated to CDI prevalence (R square = 0.73; regression as diabetes mellitus, renal failure, inmunosuppression, chronic coefficient =1.194, 95% CI= 1.192 – 1.196). pressure ulcers and hypoalbuminemia; healthcare exposures such as previous surgery, enteral feeding, immunosuppressive Introduction therapy, use of antibiotics (as the proportion of patients receiving Clostridium difficile is the most commonly diagnosed cause of any antimicrobial on the day of the survey); type of ward (general infectious hospital-acquired diarrhoea [1]. Since 2003, outbreaks medical as opposed to a surgical, intensive care, paediatric or of severe nosocomial diarrhoea, caused by a new virulent strain obstetric ward); and size of the hospital as measured by number of C. difficile Type 027, characterised as toxinotype III, North of beds (small: lesser than 200 beds; medium: 200-500 beds; American pulsed-field type 1 (NAP1), restriction-endonuclease large: greater than 500 beds). Hospital validated forms were sent analysis group type BI and PCR-ribotype 027 have been recognised to an independent central analysis unit for further validation and in Canada and the USA, and soon thereafter in several European analysis. A hospital report was sent back to every participating countries, as well as in Japan, evoking great concern among public hospital to avoid possible disagreements before final integration health authorities [2-5]. Limited information is available on the of the collected results in a centralized database. We focused our burden and epidemiology of C. difficile infection (CDI) in Spain. analysis on the period 1999-2007. The present report communicates the secular trends in prevalence of CDI among hospitalised patients in Spain from 1999 through Prevalence rates were expressed as the number of patients with 2007 and factors associated with CDI prevalent cases. CDI per 10,000 hospitalised patients. Comparisons of facilities, clinical conditions, exposures and demographic features were Methods made by chi-square test, likelihood ratio test, Student’s t test or Since 1990, a point prevalence study series of nosocomial Mann-Whitney test if appropriate. Secular trends were evaluated infections among patients hospitalised in acute care facilities have by Poisson regression. For factors associated with CDI, prevalence been conducted in Spain (Estudio de prevalencia de las infecciones rate ratios and 95% confidence intervals were computed. For nosocomiales en los hospitales españoles – EPINE study). correlation of the use of antimicrobial and the annual prevalence EUROSURVEILL ANCE Vol . 13 · Issues 7–9 · Jul–Sep 2008 · www.eurosurveill an ce .org 1 rates Spearman correlation coefficient and regression coefficient (82-85%) participating in the survey at any given year took part along 95% confidence intervals were calculated. All calculations in the entire nine-year series. The mean age of patients increased were performed with Stata/SE 9.0 statistical software. from 56.2 years in 1999 to 58.7 years in 2007. A total of 378 CDI cases were identified. Prevalence rates of CDI ranged from 3.9 Results cases/10,000 patients in 1999 to 12.2 cases/10,000 patients Between 1999 and 2007 on average 249 hospitals per year in 2007, and showed a significantly increasing trend from 1999 participated in the EPINE survey yielding a representative sample of through 2007 (prevalence rate ratio for one year increment 1.09; almost 57,000 hospitalised patients per year. Most of the hospitals 95% CI 1.05 – 1.14) (Table 1). Prevalence rates were consistently T able 1 Prevalence rates of Clostridium difficile infection (CDI) and use of antimicrobials in hospitals in Spain, by year of the survey 1999 2000 2001 2002 2003 2004 2005 2006 2007 Prevalence 95% CI (N=233) (N=243) (N=243) (N=246) (N=241) (N=258) (N=257) (N=253) (N=266) ratio* Age group 18-64 years Cases 6 9 11 11 12 11 14 8 26 Patients 23,077 23,357 23,369 22,690 22,565 24,130 23,823 23,857 25,042 Prevalence rate 2.6 3.9 4.7 4.8 5.3 4.6 5.9 3.4 10.4 1.12 (1.04-1.21) 65-79 years Cases 8 11 14 16 10 15 19 14 24 Patients 18,569 19,164 19,718 18,752 18,542 19,488 19,256 18,513 19,466 Prevalence rate 4.3 5.7 7.1 8.5 5.4 7.7 9.9 7.6 12.3 1.09 (1.03-1.18) > 80 years Cases 7 17 10 13 9 17 13 17 23 Patients 7,170 7,649 8,342 8,493 8,738 9,468 9,975 10,624 11,786 Prevalence rate 9.8 22.2 12.0 15.3 10.3 18.0 13.0 16.0 19.5 1.03 (0.96-1.10) All age groups Cases 21 39 35 40 33 45 50 40 75 Patients 53,689 55,323 56,321 54,882 54,864 58,672 58,379 57,989 61,496 1.09 (1.05-1.14) Prevalence rate 3.9 7.0 6.2 7.3 6.0 7.7 8.6 6.9 12.2 Patients receiving antimicrobials (%) 36.0 36.7 36.4 37.0 36.9 38.6 39.4 39.4 40.7 Prevalence rates are given per 10,000 hospitalised patients N = number of participating hospitals *Prevalence ratio for one year increment, estimated by Poisson regression T able 2 Clinical and demographic characteristics of patients with Clostridium difficile infection (CDI) in comparison with non-CDI patients, hospitals in Spain 1999-2007 CDI patients non-CDI patients Prevalence ratio p value N = 378 N= 511,237 (95% CI) Age in years, median (range) 74 (4-97) 64 (1-99) 10* <0.001 Age > 65 years 207 (68.3%) 216,361 (48.1%) 2.3 (1.8-3.0) <0.001 Male gender 203 (54.4%) 259,068 (51.6%) 1.1 (0.9-1.4) 0.273 Renal failure 80 (22.2%) 41,787 (8.5%) 3.1 (2.4-3.9) <0.001 Diabetes mellitus 95 (26.1%) 97,475 (19.8%) 1.4 (1.1-1.8) <0.003 Immunodeficiency 40 (11-1%) 18,305 (3.7%) 3.2 (2.3-4.5) <0.001 Hypoalbuminemia 107 (30.3%) 29,093 (6.1%) 6.7 (5.3-8.4) <0.001 Pressure ulcers 69 (19.1%) 24,051 (5.0%) 4.5 (3.5-5.9) <0.001 Previous surgery 57 (15.5%) 151,639 (30.2%) 0.4 (0.3-0.6) <0.001 Enteral feeding 60 (16.4%) 31,990 (6.5%) 2.8 (2.2-3.7) <0.001 Immunosuppressive therapy 62 (17.2%) 41,468 (8.4%) 2.3 (1.7-3.0) <0.001 Hospital size** < 200 beds 75 (19.8%) 147,521 (28.9%) Reference < 0.001 200-500 beds 145 (38.4%) 213,673 (41.8%) 1.3 (1.0-1.8) > 500 beds 158 (41.8%) 149,930 (29.3%) 2.1 (1.6-2.7) Medical wards*** 266 (70.4%) 209,276 (40.9%) 3.4 (2.7-4.3) <0.001 * Median difference ** Likelihood ratio test=30.7 *** In this study, we use the term “medical ward” to indicate internal medicine (and its subspecialties) wards as opposed to “non-medical wards” including surgical, intensive care, paediatric and obstetric wards. 2 EUROSURVEILL ANCE Vol . 13 · Issues 7–9 · Jul–Sep 2008 · www.eurosurveill an ce .org higher in older age groups for every year. Furthermore, for adults, It has been previously shown that the older and the sicker the prevalence rates showed a statistically significant increasing time patients the more prone they are to CDI. In fact, in our study, CDI trend for every age group except for the group of patients aged 80 patients were older than the patients not infected, and CDI rates years and older (Table 1). were consistently higher for older age groups during the entire study period. Furthermore, the mean age of patients increased by almost The prevalence of use of antimicrobials in the hospitalised 2.5 years from 1999 to 2007. The severity of the main underlying population (given as the number of patients on antimicrobials disease and/or the number of comorbidities also increased during per 100 hospitalised patients) increased linearly from 36.0% in the period [7] and could be another factor accounting for the 1999 through 40.7% in 2007 (p <0.001) (Table 1) and showed a increase in CDI. strong correlation with CDI prevalence rates (R = 0.73; regression coefficient for percentage of use of antimicrobials 1.194, 95% The possibility that the new virulent strain of C. difficile Type confidence interval 1.192 – 1.196). 027 could account for the increasing trend observed is extremely remote. This strain has been identified in Spain in two cases only: Comparison of CDI and non-CDI patients by main characteristics an imported case of CDI in a patient transferred from a hospital in is displayed in Table 2. No differences were found for gender. the United Kingdom, and another one in a laboratory technician who However, CDI patients were older, presented more frequently had worked with C. difficile isolates and subsequently developed underlying conditions such as renal failure, diabetes mellitus, CDI. However, no outbreaks associated with this strain have been immunodeficiency, pressure ulcers or hypoalbuminemia. CDI communicated to date [8]. patients were also more frequently exposed to enteral feeding, and to immunosuppressive therapy, but signic fi antly less often exposed As previously reported, the underlying diseases and certain to surgical procedures. Furthermore, being admitted to a general clinical characteristics were associated with a higher risk of CDI. medical ward (such as internal medicine or its subspecialties: We found diabetes mellitus, renal failure, immunodeficiency or cardiology, pulmonology, etc.), as opposed to a surgical, intensive hypoalbuminemia as well as being subjected to enteral feeding or care, paediatric or obstetric ward was associated with a higher immunosuppressive therapy to be associated with CDI. Furthermore, prevalence of CDI, and so was the size of the hospital (rate ratio being admitted to a general medical ward and a large hospital, were 1.3 and 2.1 for medium and large size hospitals, respectively, both associated with a higher rate of CDI, whereas a history of compared with small size hospitals) (Table 2). previous surgery was associated with a lower rate of CDI. However, higher rates of CDI in larger hospitals could also be related to the Discussion more complex case-mix and to better awareness of CDI by clinicians in third care, including many referral, centres. One of the strengths of this prevalence series is that it represents more than half of the population hospitalised in acute care centres Our study has several limitations. Prevalence rates are not in Spain in a given day, and most data come from hospitals that have regularly participated in the survey every year. These data directly comparable to incidence rates that have been proposed for surveillance of CDI. Estimation of incidence rates from indicate that prevalence rates of CDI per 10,000 hospitalised patients over the period 1999-2007 increased significantly from prevalence rates in the hospital framework is risky and has not been recommended [9]. A calculation from the formula proposed 3.9 to 12.2, at an annual rate of 9%. Furthermore, this increase could also be demonstrated for patients pertaining to the age group by Rhame and Sudderth [10] yielded an average incidence rate of 18-79 years (the average annual increases for 18-64-year-olds of 9.8 cases/10,000 patient-days for the whole period studied and 65-79-year-olds were 12% and 9% respectively). (1999-2007). This estimate would be within the range of other incidence estimates [11-14] before the emergence of the new Several factors could explain this increase in CDI rates. When virulent C. difficile Type 027. looking for potential outbreaks that could account for the differences between the various years, we were able to identify one hospital in It is also likely that the figures we obtained underestimate the 2002, two hospitals in 2004, another hospital in 2006 and three actual prevalence, since testing for C. difficile is not a routine hospitals in 2007 showing point prevalence rates higher than 40 clinical practice in less severe cases, and is performed at the per 10,000 patients. Thus, even if prevalence surveys are not a discretion of the attending physician. On the other hand, in recent powerful tool to detect outbreaks, the hypothesis of increasing years, clinicians have shown increased awareness of CDI in endemic trends related to more frequent hospital outbreaks in most recent situations and have more frequently tested for C. difficile toxins years cannot be ruled out on the basis of our data. thus yielding a higher number of CDI diagnoses. Exposure to several classes of antimicrobials has been Further limitation of our study is that we lack information on consistently found to be associated with CDI [6]. During the strain identification therefore the importance of C. difficile Type 027 study period the proportion of patients receiving antimicrobials can not be definitely ruled out. Both cross-sectional and ecological increased significantly and was found strongly correlated to the studies are not a valid study design for risk factor research as they CDI prevalence rates. This increase in the use of antimicrobials do not allow for establishing causal inferences, but they can point suggests it could be one of the causes of the observed increase in out potential risk factors for further evaluation. Another concern CDI rates. Nevertheless, this hypothesis can not be proven from the is seasonality. As the survey was performed every year during May, ecological trend presented in our study since individual exposition seasonal variations in time could not be assessed. Other studies should be taken into account for a causal association and we lacked have observed seasonality with rates peaking in winter months and data on individual patients’ exposures to antimicrobials before their lower rates in summer [15]. However, the fact that we performed developing CDI. the survey in the same month each year, although precluding a study of seasonality, allowed us to measure trends. EUROSURVEILL ANCE Vol . 13 · Issues 7–9 · Jul–Sep 2008 · www.eurosurveill an ce .org 3 To conclude: over the 1999-2007 period prevalence rates of CDI increased significantly in Spanish hospitals. On-going surveillance systems are needed to closely monitor incidence, C. difficile strains characteristics, as well as the changing epidemiology of CDI in Spain. Acknowledgment Supported by the Fondo para la investigación, Spanish Ministry of Health, grants BA06/90053 and PI07/90255. Refe re nces 1. Riley TV. The epidemiolog y of Clostridium difc fi ile-associated diarrhoea. Rev Med Microbiol. 1994;5:117–122. 2. Warny M, Pepin J, Fang A, Killgore G, Thompson A, Brazier J, et al. Toxin production by an emerging strain of Clostridium difficile associated with outbreaks of severe disease in North America and Europe. Lancet. 2005;366:1079–84. 3. Kuijper EJ, Coignard B, Tüll P. Emergence of Clostridium difficile-associated disease in North America and Europe. Clin Microbiol Infect. 2006;12(suppl 6):2-18. 4. Kuijper EJ, Coignard B, Brazier JS, Suetens C, Drudy D, Wiuff C, et al. Update of Clostridium difc fi ile-associated disease due to PCR ribotype 027 in Europe. Euro Surveill. 2007;12(6):pii=714. Available from: http://www.eurosurveillance. org/ViewArticle.aspx?ArticleId=714. 5. Kato H, Ito Y, van den Berg R, Kuijper EJ, Arakawa Y. First isolation of Clostridium difc fi ile 027 in Japan. Euro Surveill. 2007;12(2):pii=3110. Available from: http://www.eurosurveillance.org/ViewArticle.aspx?ArticleId=3110 6. Bartlett JG. Clostridium difficile. Old and New Observations. J Clin Gastroenterol 2007;41:S24–S29. 7. Vaqué J, Rodrigo JA y Grupo de Trabajo EPINE, editors. Prevalencia de las infecciones en los hospitales españoles. Estudio EPINE. Resultados de los estudios de 2004, 2005, 2006 y 2007, y evolución 1990-2207: 18 años. Medicina Preventiva 2008;XIV: 1-73. 8. Marín M, Martín A, Alcalá L, Peláez T, Sánchez-Somolinos M, Cercenado E, et al. Descripción de ribotipos y perl fi toxigénico de cepas de cepas aisladas de pacientes con diarrea asociada a Clostridium difficile (DACD). Enferm Infecc Microbiol Clin 2008;26 (Espec Congr):1-221. Resumen 266. 9. French G. Repeated prevalence surveys. In: Emmerson A, Ayliffe G, eds. Clinical Infectious Diseases. London, UK: Baillière; 1996:179-195. 10. Rhame F, Sudderth W. Incidence and prevalence as used in the analysis of the occurrence of nosocomial infection rates. Am J Epidemiol. 1981;113(1):1-11. 11. Olson MM, Shanholtzer CJ, Lee JT Jr, Gerding DN. Ten years of prospective Clostridium difficile-associated disease surveillance and treatment at the Minneapolis VA Medical Center, 1982-1991. Infect Control Hosp Epidemiol. 1994;15(6):371-81. 12. Alfa MJ, Du T, Beda G. Survey of incidence of Clostridium difficile infection in Canadian hospitals and diagnostic approaches. J Clin Microbiol. 1998;36(7):2076-80. 13. Barbut F, Delmée M, Brazier JS, Petit JC, Poxton IR, Rupnik M, et al. A European survey of diagnostic methods and testing protocols for Clostridium difc fi ile. Clin Microbiol Infect. 2003;9(10):989-96. 14. McDonald LC, Owings M, Jernigan DB. Clostridium difficile infection in patients discharged from US short-stay hospitals, 1996–2003. Emerg Infect Dis. 2006;12(3):409-15. 15. Archibald LK, Banerjee SN, Jarvis WR. Secular trends in hospital acquired Clostridium difficile disease in the United States, 1987–2001. J Infect Dis. 2004;189(9):1585–9. This article was published on 31 July 2008. Citation style for this article: Asensio A, Vaque-Rafart J, Calbo-Torrecillas F, Gestal-Otero JJ, López-Fernández F, Trilla-Garcia A, Canton R, EPINE Working Group. Increasing rates in Clostridium difficile infection (CDI) among hospitalised patients, Spain 1999-2007. Euro Surveill. 2008;13(31):pii=18943. Available online: http://www. eurosurveillance.org/ViewArticle.aspx?ArticleId=18943 4 EUROSURVEILL ANCE Vol . 13 · Issues 7–9 · Jul–Sep 2008 · www.eurosurveill an ce .org http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Eurosurveillance Unpaywall

Increasing rates in Clostridium difficile infection (CDI) among hospitalised patients, Spain 1999-2007

Download PDF
4 pages

Loading next page...
 
/lp/unpaywall/increasing-rates-in-clostridium-difficile-infection-cdi-among-IFiXj4tJ04

References

References for this paper are not available at this time. We will be adding them shortly, thank you for your patience.

Publisher
Unpaywall
ISSN
1025-496X
DOI
10.2807/ese.13.31.18943-en
Publisher site
See Article on Publisher Site

Abstract

Resea rc h a r ti cl es I n c r e a s I n g r a t e s I n C l o s t r i d i u m d i f f i C i l e I n f e c t I o n ( c D I) a m o n g h o s p I ta l I s e D p a t I e n t s , s p a I n 1999-2007 1 2 3 4 5 A Asensio ([email protected]) , J Vaque-Rafart , F Calbo-Torrecillas , J J Gestal-Otero , F López-Fernández , 6 7 A Trilla-Garcia , R Canton , EPINE Working Group 1. Department of Preventive Medicine, Hospital Universitario Puerta de Hierro. Madrid, Spain 2. Department of Preventive Medicine, Hospital Vall d´Hebron. Barcelona, Spain 3. Department of Preventive Medicine, Hospital Carlos Haya. Málaga, Spain 4. Department of Preventive Medicine and Public Health, University of Santiago de Compostela, Spain 5. Department of Preventive Medicine, Hospital Universitario Puerto del Mar, Cádiz, Spain 6. Epidemiologi Unit. Hospital Clinic. Barcelona, Spain 7. Department of Microbiolog y. Hospital Universitario Ramón y Cajal. Madrid, Spain Limited information is available on the burden and epidemiology Each year in May, acute care hospitals in Spain are requested to of Clostridium difficile infection (CDI) in Spain. The present report voluntarily join the EPINE prevalence study. Participating hospitals communicates the secular trends in prevalence of CDI among fill a standardised questionnaire on each hospitalised patient as hospitalised patients in Spain from 1999 through 2007. Data well as overall data on the hospital and the hospital’s wards. were obtained through the EPINE study (Estudio de prevalencia de CDI diagnosis relies on CDC case-definitions for nosocomial las infecciones nosocomiales en los hospitales españoles), a point infections (note: the EU case definitions were not available at prevalence study series of nosocomial infections among patients the time of the start of the study), and includes cases of either clinical diarrhoea or toxic megacolon with laboratory evidence of admitted to hospital in Spain. A total of 378 cases with CDI were identified. Median age was 74 positive stool culture and/or toxin assay for C. difficile. Thus our analysis encompassed a symptomatic population with a positive years. Prevalence rates of CDI increased from 3.9 to 12.2 cases per 10,000 hospitalised patients and showed a significantly increasing microbiological confirmation of CDI by culture, toxin assay or both. secular trend from 1999 through 2007 (prevalence rate ratio per each year increment 1.09; 95% CI 1.05 – 1.14). Percentage of In addition to information on nosocomial infections, the patient hospitalised patients receiving antimicrobials increased linearly forms collected from the hospitals included demographic data (age from 36.0% in 1999 to 40.7% in 2007 (p <0.001) and was and gender); information on underlying clinical conditions such strongly correlated to CDI prevalence (R square = 0.73; regression as diabetes mellitus, renal failure, inmunosuppression, chronic coefficient =1.194, 95% CI= 1.192 – 1.196). pressure ulcers and hypoalbuminemia; healthcare exposures such as previous surgery, enteral feeding, immunosuppressive Introduction therapy, use of antibiotics (as the proportion of patients receiving Clostridium difficile is the most commonly diagnosed cause of any antimicrobial on the day of the survey); type of ward (general infectious hospital-acquired diarrhoea [1]. Since 2003, outbreaks medical as opposed to a surgical, intensive care, paediatric or of severe nosocomial diarrhoea, caused by a new virulent strain obstetric ward); and size of the hospital as measured by number of C. difficile Type 027, characterised as toxinotype III, North of beds (small: lesser than 200 beds; medium: 200-500 beds; American pulsed-field type 1 (NAP1), restriction-endonuclease large: greater than 500 beds). Hospital validated forms were sent analysis group type BI and PCR-ribotype 027 have been recognised to an independent central analysis unit for further validation and in Canada and the USA, and soon thereafter in several European analysis. A hospital report was sent back to every participating countries, as well as in Japan, evoking great concern among public hospital to avoid possible disagreements before final integration health authorities [2-5]. Limited information is available on the of the collected results in a centralized database. We focused our burden and epidemiology of C. difficile infection (CDI) in Spain. analysis on the period 1999-2007. The present report communicates the secular trends in prevalence of CDI among hospitalised patients in Spain from 1999 through Prevalence rates were expressed as the number of patients with 2007 and factors associated with CDI prevalent cases. CDI per 10,000 hospitalised patients. Comparisons of facilities, clinical conditions, exposures and demographic features were Methods made by chi-square test, likelihood ratio test, Student’s t test or Since 1990, a point prevalence study series of nosocomial Mann-Whitney test if appropriate. Secular trends were evaluated infections among patients hospitalised in acute care facilities have by Poisson regression. For factors associated with CDI, prevalence been conducted in Spain (Estudio de prevalencia de las infecciones rate ratios and 95% confidence intervals were computed. For nosocomiales en los hospitales españoles – EPINE study). correlation of the use of antimicrobial and the annual prevalence EUROSURVEILL ANCE Vol . 13 · Issues 7–9 · Jul–Sep 2008 · www.eurosurveill an ce .org 1 rates Spearman correlation coefficient and regression coefficient (82-85%) participating in the survey at any given year took part along 95% confidence intervals were calculated. All calculations in the entire nine-year series. The mean age of patients increased were performed with Stata/SE 9.0 statistical software. from 56.2 years in 1999 to 58.7 years in 2007. A total of 378 CDI cases were identified. Prevalence rates of CDI ranged from 3.9 Results cases/10,000 patients in 1999 to 12.2 cases/10,000 patients Between 1999 and 2007 on average 249 hospitals per year in 2007, and showed a significantly increasing trend from 1999 participated in the EPINE survey yielding a representative sample of through 2007 (prevalence rate ratio for one year increment 1.09; almost 57,000 hospitalised patients per year. Most of the hospitals 95% CI 1.05 – 1.14) (Table 1). Prevalence rates were consistently T able 1 Prevalence rates of Clostridium difficile infection (CDI) and use of antimicrobials in hospitals in Spain, by year of the survey 1999 2000 2001 2002 2003 2004 2005 2006 2007 Prevalence 95% CI (N=233) (N=243) (N=243) (N=246) (N=241) (N=258) (N=257) (N=253) (N=266) ratio* Age group 18-64 years Cases 6 9 11 11 12 11 14 8 26 Patients 23,077 23,357 23,369 22,690 22,565 24,130 23,823 23,857 25,042 Prevalence rate 2.6 3.9 4.7 4.8 5.3 4.6 5.9 3.4 10.4 1.12 (1.04-1.21) 65-79 years Cases 8 11 14 16 10 15 19 14 24 Patients 18,569 19,164 19,718 18,752 18,542 19,488 19,256 18,513 19,466 Prevalence rate 4.3 5.7 7.1 8.5 5.4 7.7 9.9 7.6 12.3 1.09 (1.03-1.18) > 80 years Cases 7 17 10 13 9 17 13 17 23 Patients 7,170 7,649 8,342 8,493 8,738 9,468 9,975 10,624 11,786 Prevalence rate 9.8 22.2 12.0 15.3 10.3 18.0 13.0 16.0 19.5 1.03 (0.96-1.10) All age groups Cases 21 39 35 40 33 45 50 40 75 Patients 53,689 55,323 56,321 54,882 54,864 58,672 58,379 57,989 61,496 1.09 (1.05-1.14) Prevalence rate 3.9 7.0 6.2 7.3 6.0 7.7 8.6 6.9 12.2 Patients receiving antimicrobials (%) 36.0 36.7 36.4 37.0 36.9 38.6 39.4 39.4 40.7 Prevalence rates are given per 10,000 hospitalised patients N = number of participating hospitals *Prevalence ratio for one year increment, estimated by Poisson regression T able 2 Clinical and demographic characteristics of patients with Clostridium difficile infection (CDI) in comparison with non-CDI patients, hospitals in Spain 1999-2007 CDI patients non-CDI patients Prevalence ratio p value N = 378 N= 511,237 (95% CI) Age in years, median (range) 74 (4-97) 64 (1-99) 10* <0.001 Age > 65 years 207 (68.3%) 216,361 (48.1%) 2.3 (1.8-3.0) <0.001 Male gender 203 (54.4%) 259,068 (51.6%) 1.1 (0.9-1.4) 0.273 Renal failure 80 (22.2%) 41,787 (8.5%) 3.1 (2.4-3.9) <0.001 Diabetes mellitus 95 (26.1%) 97,475 (19.8%) 1.4 (1.1-1.8) <0.003 Immunodeficiency 40 (11-1%) 18,305 (3.7%) 3.2 (2.3-4.5) <0.001 Hypoalbuminemia 107 (30.3%) 29,093 (6.1%) 6.7 (5.3-8.4) <0.001 Pressure ulcers 69 (19.1%) 24,051 (5.0%) 4.5 (3.5-5.9) <0.001 Previous surgery 57 (15.5%) 151,639 (30.2%) 0.4 (0.3-0.6) <0.001 Enteral feeding 60 (16.4%) 31,990 (6.5%) 2.8 (2.2-3.7) <0.001 Immunosuppressive therapy 62 (17.2%) 41,468 (8.4%) 2.3 (1.7-3.0) <0.001 Hospital size** < 200 beds 75 (19.8%) 147,521 (28.9%) Reference < 0.001 200-500 beds 145 (38.4%) 213,673 (41.8%) 1.3 (1.0-1.8) > 500 beds 158 (41.8%) 149,930 (29.3%) 2.1 (1.6-2.7) Medical wards*** 266 (70.4%) 209,276 (40.9%) 3.4 (2.7-4.3) <0.001 * Median difference ** Likelihood ratio test=30.7 *** In this study, we use the term “medical ward” to indicate internal medicine (and its subspecialties) wards as opposed to “non-medical wards” including surgical, intensive care, paediatric and obstetric wards. 2 EUROSURVEILL ANCE Vol . 13 · Issues 7–9 · Jul–Sep 2008 · www.eurosurveill an ce .org higher in older age groups for every year. Furthermore, for adults, It has been previously shown that the older and the sicker the prevalence rates showed a statistically significant increasing time patients the more prone they are to CDI. In fact, in our study, CDI trend for every age group except for the group of patients aged 80 patients were older than the patients not infected, and CDI rates years and older (Table 1). were consistently higher for older age groups during the entire study period. Furthermore, the mean age of patients increased by almost The prevalence of use of antimicrobials in the hospitalised 2.5 years from 1999 to 2007. The severity of the main underlying population (given as the number of patients on antimicrobials disease and/or the number of comorbidities also increased during per 100 hospitalised patients) increased linearly from 36.0% in the period [7] and could be another factor accounting for the 1999 through 40.7% in 2007 (p <0.001) (Table 1) and showed a increase in CDI. strong correlation with CDI prevalence rates (R = 0.73; regression coefficient for percentage of use of antimicrobials 1.194, 95% The possibility that the new virulent strain of C. difficile Type confidence interval 1.192 – 1.196). 027 could account for the increasing trend observed is extremely remote. This strain has been identified in Spain in two cases only: Comparison of CDI and non-CDI patients by main characteristics an imported case of CDI in a patient transferred from a hospital in is displayed in Table 2. No differences were found for gender. the United Kingdom, and another one in a laboratory technician who However, CDI patients were older, presented more frequently had worked with C. difficile isolates and subsequently developed underlying conditions such as renal failure, diabetes mellitus, CDI. However, no outbreaks associated with this strain have been immunodeficiency, pressure ulcers or hypoalbuminemia. CDI communicated to date [8]. patients were also more frequently exposed to enteral feeding, and to immunosuppressive therapy, but signic fi antly less often exposed As previously reported, the underlying diseases and certain to surgical procedures. Furthermore, being admitted to a general clinical characteristics were associated with a higher risk of CDI. medical ward (such as internal medicine or its subspecialties: We found diabetes mellitus, renal failure, immunodeficiency or cardiology, pulmonology, etc.), as opposed to a surgical, intensive hypoalbuminemia as well as being subjected to enteral feeding or care, paediatric or obstetric ward was associated with a higher immunosuppressive therapy to be associated with CDI. Furthermore, prevalence of CDI, and so was the size of the hospital (rate ratio being admitted to a general medical ward and a large hospital, were 1.3 and 2.1 for medium and large size hospitals, respectively, both associated with a higher rate of CDI, whereas a history of compared with small size hospitals) (Table 2). previous surgery was associated with a lower rate of CDI. However, higher rates of CDI in larger hospitals could also be related to the Discussion more complex case-mix and to better awareness of CDI by clinicians in third care, including many referral, centres. One of the strengths of this prevalence series is that it represents more than half of the population hospitalised in acute care centres Our study has several limitations. Prevalence rates are not in Spain in a given day, and most data come from hospitals that have regularly participated in the survey every year. These data directly comparable to incidence rates that have been proposed for surveillance of CDI. Estimation of incidence rates from indicate that prevalence rates of CDI per 10,000 hospitalised patients over the period 1999-2007 increased significantly from prevalence rates in the hospital framework is risky and has not been recommended [9]. A calculation from the formula proposed 3.9 to 12.2, at an annual rate of 9%. Furthermore, this increase could also be demonstrated for patients pertaining to the age group by Rhame and Sudderth [10] yielded an average incidence rate of 18-79 years (the average annual increases for 18-64-year-olds of 9.8 cases/10,000 patient-days for the whole period studied and 65-79-year-olds were 12% and 9% respectively). (1999-2007). This estimate would be within the range of other incidence estimates [11-14] before the emergence of the new Several factors could explain this increase in CDI rates. When virulent C. difficile Type 027. looking for potential outbreaks that could account for the differences between the various years, we were able to identify one hospital in It is also likely that the figures we obtained underestimate the 2002, two hospitals in 2004, another hospital in 2006 and three actual prevalence, since testing for C. difficile is not a routine hospitals in 2007 showing point prevalence rates higher than 40 clinical practice in less severe cases, and is performed at the per 10,000 patients. Thus, even if prevalence surveys are not a discretion of the attending physician. On the other hand, in recent powerful tool to detect outbreaks, the hypothesis of increasing years, clinicians have shown increased awareness of CDI in endemic trends related to more frequent hospital outbreaks in most recent situations and have more frequently tested for C. difficile toxins years cannot be ruled out on the basis of our data. thus yielding a higher number of CDI diagnoses. Exposure to several classes of antimicrobials has been Further limitation of our study is that we lack information on consistently found to be associated with CDI [6]. During the strain identification therefore the importance of C. difficile Type 027 study period the proportion of patients receiving antimicrobials can not be definitely ruled out. Both cross-sectional and ecological increased significantly and was found strongly correlated to the studies are not a valid study design for risk factor research as they CDI prevalence rates. This increase in the use of antimicrobials do not allow for establishing causal inferences, but they can point suggests it could be one of the causes of the observed increase in out potential risk factors for further evaluation. Another concern CDI rates. Nevertheless, this hypothesis can not be proven from the is seasonality. As the survey was performed every year during May, ecological trend presented in our study since individual exposition seasonal variations in time could not be assessed. Other studies should be taken into account for a causal association and we lacked have observed seasonality with rates peaking in winter months and data on individual patients’ exposures to antimicrobials before their lower rates in summer [15]. However, the fact that we performed developing CDI. the survey in the same month each year, although precluding a study of seasonality, allowed us to measure trends. EUROSURVEILL ANCE Vol . 13 · Issues 7–9 · Jul–Sep 2008 · www.eurosurveill an ce .org 3 To conclude: over the 1999-2007 period prevalence rates of CDI increased significantly in Spanish hospitals. On-going surveillance systems are needed to closely monitor incidence, C. difficile strains characteristics, as well as the changing epidemiology of CDI in Spain. Acknowledgment Supported by the Fondo para la investigación, Spanish Ministry of Health, grants BA06/90053 and PI07/90255. Refe re nces 1. Riley TV. The epidemiolog y of Clostridium difc fi ile-associated diarrhoea. Rev Med Microbiol. 1994;5:117–122. 2. Warny M, Pepin J, Fang A, Killgore G, Thompson A, Brazier J, et al. Toxin production by an emerging strain of Clostridium difficile associated with outbreaks of severe disease in North America and Europe. Lancet. 2005;366:1079–84. 3. Kuijper EJ, Coignard B, Tüll P. Emergence of Clostridium difficile-associated disease in North America and Europe. Clin Microbiol Infect. 2006;12(suppl 6):2-18. 4. Kuijper EJ, Coignard B, Brazier JS, Suetens C, Drudy D, Wiuff C, et al. Update of Clostridium difc fi ile-associated disease due to PCR ribotype 027 in Europe. Euro Surveill. 2007;12(6):pii=714. Available from: http://www.eurosurveillance. org/ViewArticle.aspx?ArticleId=714. 5. Kato H, Ito Y, van den Berg R, Kuijper EJ, Arakawa Y. First isolation of Clostridium difc fi ile 027 in Japan. Euro Surveill. 2007;12(2):pii=3110. Available from: http://www.eurosurveillance.org/ViewArticle.aspx?ArticleId=3110 6. Bartlett JG. Clostridium difficile. Old and New Observations. J Clin Gastroenterol 2007;41:S24–S29. 7. Vaqué J, Rodrigo JA y Grupo de Trabajo EPINE, editors. Prevalencia de las infecciones en los hospitales españoles. Estudio EPINE. Resultados de los estudios de 2004, 2005, 2006 y 2007, y evolución 1990-2207: 18 años. Medicina Preventiva 2008;XIV: 1-73. 8. Marín M, Martín A, Alcalá L, Peláez T, Sánchez-Somolinos M, Cercenado E, et al. Descripción de ribotipos y perl fi toxigénico de cepas de cepas aisladas de pacientes con diarrea asociada a Clostridium difficile (DACD). Enferm Infecc Microbiol Clin 2008;26 (Espec Congr):1-221. Resumen 266. 9. French G. Repeated prevalence surveys. In: Emmerson A, Ayliffe G, eds. Clinical Infectious Diseases. London, UK: Baillière; 1996:179-195. 10. Rhame F, Sudderth W. Incidence and prevalence as used in the analysis of the occurrence of nosocomial infection rates. Am J Epidemiol. 1981;113(1):1-11. 11. Olson MM, Shanholtzer CJ, Lee JT Jr, Gerding DN. Ten years of prospective Clostridium difficile-associated disease surveillance and treatment at the Minneapolis VA Medical Center, 1982-1991. Infect Control Hosp Epidemiol. 1994;15(6):371-81. 12. Alfa MJ, Du T, Beda G. Survey of incidence of Clostridium difficile infection in Canadian hospitals and diagnostic approaches. J Clin Microbiol. 1998;36(7):2076-80. 13. Barbut F, Delmée M, Brazier JS, Petit JC, Poxton IR, Rupnik M, et al. A European survey of diagnostic methods and testing protocols for Clostridium difc fi ile. Clin Microbiol Infect. 2003;9(10):989-96. 14. McDonald LC, Owings M, Jernigan DB. Clostridium difficile infection in patients discharged from US short-stay hospitals, 1996–2003. Emerg Infect Dis. 2006;12(3):409-15. 15. Archibald LK, Banerjee SN, Jarvis WR. Secular trends in hospital acquired Clostridium difficile disease in the United States, 1987–2001. J Infect Dis. 2004;189(9):1585–9. This article was published on 31 July 2008. Citation style for this article: Asensio A, Vaque-Rafart J, Calbo-Torrecillas F, Gestal-Otero JJ, López-Fernández F, Trilla-Garcia A, Canton R, EPINE Working Group. Increasing rates in Clostridium difficile infection (CDI) among hospitalised patients, Spain 1999-2007. Euro Surveill. 2008;13(31):pii=18943. Available online: http://www. eurosurveillance.org/ViewArticle.aspx?ArticleId=18943 4 EUROSURVEILL ANCE Vol . 13 · Issues 7–9 · Jul–Sep 2008 · www.eurosurveill an ce .org

Journal

Eurosurveillance – Unpaywall

Published: Jul 31, 2008

There are no references for this article.