INTRODUCTION The role of methicillin-resistant Staphylococcus aureus (MRSA) colonization as predictor of invasive disease in ICU patients has been established many years ago.(1-3) The role of methicillin-sensitive Staphylococcus aureus (MSSA) is more debated, although in a recent report, both MRSA and MSSA carriers at admission have been found at increased risk.(4) Whether carriage at ICU admission involves a higher risk of invasive infection than carriage acquired during ICU stay has not been established. We report the results of a study aimed at estimating the frequency of S. aureus (MRSA and MSSA) colonization at admission and at discharge in patients admitted to several ICUs in Italy and at estimating the relationship between colonization status and infection by S. aureus. METHODS The ISABEL (Italian Staphylococcus aureus Benchmarked EpidemioLogy) is a prospective study conducted in 28 Italian ICUs: 13 (46%) belonged to hospitals with 500-999 beds; 23 (82.1%) were medical/surgical ICU and the remaining were general/transplant ICUs. Ethical Committee approval was obtained. All patients admitted from 23/09/2004 to 26/06/2005 were enrolled. On the days of admission and discharge, a nasal swab was collected. All samples were cultured at local laboratories; if positive for S. aureus, they were shipped to a central microbiological laboratory (Laboratorio di Microbiologia, Ospedale Ca’ Grande, Milano; Italy) for confirmation. For each patient, demographic, history and clinical data were collected at admission and throughout ICU stay, including infection caused by S. aureus during ICU stay as defined by standard criteria.(5) End-points were: proportion of patients with S. aureus/MRSA/MSSA colonization at the time of admission in ICU; proportion of non-colonized patients that acquired colonization by the end of ICU stay; proportion of S. aureus/MRSA/MSSA colonized patients (at admission or during ICU stay) who developed S. aureus infection; rate of infection. Cox proportional hazards regression was applied to assess risk factors for S. aureus infection; patients not developing infection were censored at ICU discharge. Variables were entered in multivariate models if p<0.1 at univariate analysis, with no further selection. RESULTS Enrolled patients were 3285; those colonized at admission by S. aureus were 511 (15.5%; 95%CI 14.3 – 16.8): 4.2% by MRSA, 11.3% by MSSA. Nasal swab at discharge from ICU was taken in 2885 of enrolled patients (87.8%); of these, it was positive for MRSA in 150 (5.2%) and for MSSA in 220 (7.6%): respectively 78 and 68 of these colonizations were acquired during ICU stay (incidence risk of acquiring MRSA or MSSA colonization, i.e. excluding patients already colonized at admission, during ICU stay was 2.8% [95%CI: 2.2 – 3.4] and 2.6% [95%CI: 2.0 - 3.3] respectively). S. aureus infections during stay were 119 (incidence risk 3.6%; 95CI 3.0 – 4.3), of which lower respiratory tract were 81 (68.1%) and bloodstream infections 24 (20.2%), for an overall rate of 0.5 per 100 patient-days. S. aureus infection rate was 1.6 and 0.7 per 100 patient-days among MRSA and MSSA colonized patients respectively, and was 1.3 per 100 patient-days among those acquiring MRSA colonization during stay (no patient with acquired MSSA colonization during stay developed infection). Variables associated to S. aureus infection at multivariate analysis are reported in Table 1. DISCUSSION The role of the nasal colonization by S. aureus as risk factor for subsequent bacteremia in ICU patients has been demonstrated since 2001 (6). A very recent study demonstrated that S. aureus nasal colonization is independently associated with ICU-acquired S. aureus infection and that MRSA-colonized ICU patients are more likely to become infected even after adjusting for patient comorbidities and ICU process of care.(4) Our results confirm these findings, in a multicenter setting and in a Country, Italy, where MRSA prevalence is at its highest in Europe. Additionally, our results indicate that the increase in risk applies equally to patients colonized at admission and to those acquiring colonization during ICU stay, but for MRSA only. Though this might be due to “reverse causation” (i.e. patients with S. aureus infection during ICU stay are also more likely to be carriers at discharge (despite treatment), carriage acquisition during ICU stay should perhaps be monitored, particularly in high incidence settings. In fact, MRSA may colonize the nostrils for more than 1 year and MRSA screening at ICU discharge might prove an important hospital-wide control measure.(7) Strengths of this study are the large number of patients enrolled in many centres and the prospective data collection, limiting missing data and outcome assessment bias. Additionally, this is one of the few studies assessing incidence rates of S. aureus infection (i.e. in terms of patient-days) and not simply incidence risk (proportion of admitted patients developing infection during ICU stay).(4, 8, 9) However, some limitations have to be stressed upon. First, due to economic constraints, strains of S. aureus infection were not collected for molecular biology; therefore, it has not been possible to assess homology between colonizing and infecting strains. Secondly, only nasal swabs were taken, possibly underestimating the frequency of S. aureus colonization.(10) Also, swabs were only taken at admission and at discharge, (i.e. the actual date of acquisition of carriage was not known) and therefore the denominator (i.e. the time-at risk) has been artificially increased, thus somewhat “diluting” the true incidence. Finally, the discharge swab was not taken in 400 (12.2%) patients, in line with other published studies, but possibly underestimating the rate of carriage acquisition.In conclusion, our findings confirms the tight link between S. aureus colonization and invasive disease, irrespectively from site of carriage acquisition.
Multicenter, Prospective Surveillance Study of Staphylococcus aureus Nasal Colonization in 28 Italian Intensive Care Units: The ISABEL Study
PRIVITERA, GAETANO PIERPAOLO;
2011-01-01
Abstract
INTRODUCTION The role of methicillin-resistant Staphylococcus aureus (MRSA) colonization as predictor of invasive disease in ICU patients has been established many years ago.(1-3) The role of methicillin-sensitive Staphylococcus aureus (MSSA) is more debated, although in a recent report, both MRSA and MSSA carriers at admission have been found at increased risk.(4) Whether carriage at ICU admission involves a higher risk of invasive infection than carriage acquired during ICU stay has not been established. We report the results of a study aimed at estimating the frequency of S. aureus (MRSA and MSSA) colonization at admission and at discharge in patients admitted to several ICUs in Italy and at estimating the relationship between colonization status and infection by S. aureus. METHODS The ISABEL (Italian Staphylococcus aureus Benchmarked EpidemioLogy) is a prospective study conducted in 28 Italian ICUs: 13 (46%) belonged to hospitals with 500-999 beds; 23 (82.1%) were medical/surgical ICU and the remaining were general/transplant ICUs. Ethical Committee approval was obtained. All patients admitted from 23/09/2004 to 26/06/2005 were enrolled. On the days of admission and discharge, a nasal swab was collected. All samples were cultured at local laboratories; if positive for S. aureus, they were shipped to a central microbiological laboratory (Laboratorio di Microbiologia, Ospedale Ca’ Grande, Milano; Italy) for confirmation. For each patient, demographic, history and clinical data were collected at admission and throughout ICU stay, including infection caused by S. aureus during ICU stay as defined by standard criteria.(5) End-points were: proportion of patients with S. aureus/MRSA/MSSA colonization at the time of admission in ICU; proportion of non-colonized patients that acquired colonization by the end of ICU stay; proportion of S. aureus/MRSA/MSSA colonized patients (at admission or during ICU stay) who developed S. aureus infection; rate of infection. Cox proportional hazards regression was applied to assess risk factors for S. aureus infection; patients not developing infection were censored at ICU discharge. Variables were entered in multivariate models if p<0.1 at univariate analysis, with no further selection. RESULTS Enrolled patients were 3285; those colonized at admission by S. aureus were 511 (15.5%; 95%CI 14.3 – 16.8): 4.2% by MRSA, 11.3% by MSSA. Nasal swab at discharge from ICU was taken in 2885 of enrolled patients (87.8%); of these, it was positive for MRSA in 150 (5.2%) and for MSSA in 220 (7.6%): respectively 78 and 68 of these colonizations were acquired during ICU stay (incidence risk of acquiring MRSA or MSSA colonization, i.e. excluding patients already colonized at admission, during ICU stay was 2.8% [95%CI: 2.2 – 3.4] and 2.6% [95%CI: 2.0 - 3.3] respectively). S. aureus infections during stay were 119 (incidence risk 3.6%; 95CI 3.0 – 4.3), of which lower respiratory tract were 81 (68.1%) and bloodstream infections 24 (20.2%), for an overall rate of 0.5 per 100 patient-days. S. aureus infection rate was 1.6 and 0.7 per 100 patient-days among MRSA and MSSA colonized patients respectively, and was 1.3 per 100 patient-days among those acquiring MRSA colonization during stay (no patient with acquired MSSA colonization during stay developed infection). Variables associated to S. aureus infection at multivariate analysis are reported in Table 1. DISCUSSION The role of the nasal colonization by S. aureus as risk factor for subsequent bacteremia in ICU patients has been demonstrated since 2001 (6). A very recent study demonstrated that S. aureus nasal colonization is independently associated with ICU-acquired S. aureus infection and that MRSA-colonized ICU patients are more likely to become infected even after adjusting for patient comorbidities and ICU process of care.(4) Our results confirm these findings, in a multicenter setting and in a Country, Italy, where MRSA prevalence is at its highest in Europe. Additionally, our results indicate that the increase in risk applies equally to patients colonized at admission and to those acquiring colonization during ICU stay, but for MRSA only. Though this might be due to “reverse causation” (i.e. patients with S. aureus infection during ICU stay are also more likely to be carriers at discharge (despite treatment), carriage acquisition during ICU stay should perhaps be monitored, particularly in high incidence settings. In fact, MRSA may colonize the nostrils for more than 1 year and MRSA screening at ICU discharge might prove an important hospital-wide control measure.(7) Strengths of this study are the large number of patients enrolled in many centres and the prospective data collection, limiting missing data and outcome assessment bias. Additionally, this is one of the few studies assessing incidence rates of S. aureus infection (i.e. in terms of patient-days) and not simply incidence risk (proportion of admitted patients developing infection during ICU stay).(4, 8, 9) However, some limitations have to be stressed upon. First, due to economic constraints, strains of S. aureus infection were not collected for molecular biology; therefore, it has not been possible to assess homology between colonizing and infecting strains. Secondly, only nasal swabs were taken, possibly underestimating the frequency of S. aureus colonization.(10) Also, swabs were only taken at admission and at discharge, (i.e. the actual date of acquisition of carriage was not known) and therefore the denominator (i.e. the time-at risk) has been artificially increased, thus somewhat “diluting” the true incidence. Finally, the discharge swab was not taken in 400 (12.2%) patients, in line with other published studies, but possibly underestimating the rate of carriage acquisition.In conclusion, our findings confirms the tight link between S. aureus colonization and invasive disease, irrespectively from site of carriage acquisition.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.