Search

To evaluate the relationship between antipseudomonal antibiotic consumption and each individual drug resistance rate in Pseudomonas aeruginosa strains causing ICU acquired invasive device-related infections (IDRI).A post hoc analysis was made of the data collected prospectively from the ENVIN-HELICS registry.Intensive Care Units participating in the ENVIN-UCI registry between the years 2007 and 2016 (3-month registry each year).Patients admitted for over 24h.Annual linear and nonlinear trends of resistance rates of P. aeruginosa strains identified in IDRI and days of treatment of each antipseudomonal antibiotic family per 1000 occupied ICU bed days (DOT) were calculated.A total of 15,095 episodes of IDRI were diagnosed in 11,652 patients (6.2% out of a total of 187,100). Pseudomonas aeruginosa was identified in 2095 (13.6%) of 15,432 pathogens causing IDRI. Resistance increased significantly over the study period for piperacillin-tazobactam (P0.001), imipenem (P=0.016), meropenem (P=0.004), ceftazidime (P=0.005) and cefepime (P=0.015), while variations in resistance rates for amikacin, ciprofloxacin, levofloxacin and colistin proved nonsignificant. A significant DOT decrease was observed for aminoglycosides (P0.001), cephalosporins (P0.001), quinolones (P0.001) and carbapenems (P0.001).No significant association was observed between consumption of each antipseudomonal antibiotic family and the respective resistance rates for P. aeruginosa strains identified in IDRI.

Length of stay is one of the key determinants for the risk of nosocomial infections. The distribution of this at-risk time is heavily skewed and depends on discharge or death. This study applied landmark competing risk prediction models to account for a large proportion of short-stay patients and a small proportion of long-stay patients.

Objective:To study the impact of duration of mechanical ventilation, hospitalization and multiple ventilation episodes on the development of pneumonia while accounting for extubation as a competing event.Design:A multicenter data base from a Spanish surveillance network was used to conduct a retrospective analysis of prospectively collected intensive care patients followed from admission to discharge.Setting:Spanish intensive care units (ICUs).Patients:Mechanically ventilated adult patients from 158 ICUs with 45,486 admissions, 48,705 ventilation episodes, and 314,196 ventilator days.Methods:Competing-risk models were applied to account for extubation plus 48 hours as a competing event for acquiring ventilator-associated pneumonia (VAP).Results:Time in the ICU before mechanical ventilation was associated with an increased VAP hazard rate and with longer intubation time. This indirect prolongation of intubation increased the cumulative risk to eventually acquire VAP. For instance, comparing 3–4 versus 0 days, the adjusted VAP hazard ratio was 1.40 (95% confidence interval [CI], 1.19–1.64) and the adjusted extubation hazard ratio was 0.64 (95% CI, 0.61–0.68), which leads to an adjusted VAP subdistribution hazard ratio (sHR) of 2.13 (95% CI, 1.83–2.50). Similarly, due to prolonged intubation, multiple ventilation episodes increase the risk for VAP; the adjusted sHR is 1.52 (95% CI, 1.35–1.72) for the second episode compared to the first episode, and the adjusted sHR is 1.54 (95% CI, 1.03–2.30) for the third episode compared to the first episode. The Kaplan-Meier method produced an upward biased estimated cumulative risk for VAP.Conclusions:A competing-risk analysis is necessary to receive unbiased risk estimates and to quantify the indirect effect of intubation time on the cumulative VAP risk. Our findings may guide physicians to improve medical decisions related to the harms and benefits of the duration of ventilation.

Analysing the determinants and consequences of hospital-acquired infections involves the evaluation of large cohorts. Infected patients in the cohort are often rare for specific pathogens, because most of the patients admitted to the hospital are discharged or die without such an infection. Death and discharge are competing events to acquiring an infection, because these individuals are no longer at risk of getting a hospital-acquired infection. Therefore, the data is best analysed with an extended survival model – the extended illness-death model. A common problem in cohort studies is the costly collection of covariate values. In order to provide efficient use of data from infected as well as uninfected patients, we propose a tailored case-cohort approach for the extended illness-death model. The basic idea of the case-cohort design is to only use a random sample of the full cohort, referred to as subcohort, and all cases, namely the infected patients. Thus, covariate values are only obtained for a small part of the full cohort. The method is based on existing and established methods and is used to perform regression analysis in adapted Cox proportional hazards models. We propose estimation of all cause-specific cumulative hazards and transition probabilities in an extended illness-death model based on case-cohort sampling. As an example, we apply the methodology to infection with a specific pathogen using a large cohort from Spanish hospital data. The obtained results of the case-cohort design are compared with the results in the full cohort to investigate the performance of the proposed method. Copyright © 2016 John Wiley & Sons, Ltd.

ObjectiveCompeting risks are a necessary consideration when analyzing risk factors for nosocomial infections (NIs). In this article, we identify additional information that a competing risks analysis provides in a hospital setting. Furthermore, we improve on established methods for nested case-control designs to acquire this information.MethodsUsing data from 2 Spanish intensive care units and model simulations, we show how controls selected by time-dynamic sampling for NI can be weighted to perform risk-factor analysis for death or discharge without infection. This extension not only enables hazard rate analysis for the competing risk, it also enables prediction analysis for NI.ResultsThe estimates acquired from the extension were in good agreement with the results from the full (real and simulated) cohort dataset. The reduced dataset results averted any false interpretation common in a competing-risks setting.ConclusionsUsing additional information that is routinely collected in a hospital setting, a nested case-control design can be successfully adapted to avoid a competing risks bias. Furthermore, this adapted method can be used to reanalyze past nested case-control studies to enhance their findings.

Predicting methicillin-resistant Staphylococcus aureus (MRSA) in intensive care units (ICUs) avoids inappropriate antimicrobial empirical treatment and enhances infection control. We describe risk factors for colonisation/infection related to MRSA (MRSA-C/I) in critically ill patients once in the ICU and on ICU admission, and search for an easy-to-use predictive model for MRSA colonisation/infection on ICU admission. This multicentre cohort study included 69,894 patients admitted consecutively (stay24h) in April-June in the five-year period 2006-2010 from 147 Spanish ICUs participating in the National Surveillance Study of Nosocomial Infections in ICUs (ENVIN-HELICS). Data from all patients included were used to identify risk factors for MRSA-C/I during ICU stays, from admission to discharge, using uni- and multivariable analysis (Poisson regression) to check that the sample to be used to develop the predictive models was representative of standard critical care population. To identify risk factors for MRSA-C/I on ICU admission and to develop prediction models, multivariable logistic regression analysis were then performed only on those admitted in 2010 (n=16950, 2/3 for analysis and 1/3 for subsequent validation). We found that, in the period 2006-2010, 1046 patients were MRSA-C/I. Independent risk factors for MRSA-C/I in ICU were: age65, trauma or medical patient, high APACHE-II score, admitted from a long-term care facility, urinary catheter, previous antibiotic treatment and skin-soft tissue or post-surgical superficial skin infections. Colonisation with several different MDRs significantly increased the risk of MRSA-C/I. Risk factors on ICU admission were: male gender, trauma critical patient, urgent surgery, admitted from other ICUs, hospital ward or long-term facility, immunosuppression and skin-soft tissue infection. Although the best model to identify carriers of MRSA had a good discrimination (AUC-ROC, 0.77; 95% CI, 0.72-0.82), sensitivity was 67% and specificity 76.5%. Including more complex variables did not improve prediction capability. Our conclusion is that clinical-demographic risk factors for colonisation/infection related to MRSA should not be used to accurately identify patients who would benefit from empirical anti-MRSA treatment or from specific preventive measures. Independent risk factors for MRSA colonisation/infection during ICU stay and on ICU admission are described. The latter should be considered in future studies for MRSA prediction.

Analysing the determinants and consequences of hospital-acquired infections involves the evaluation of large cohorts. Infected patients in the cohort are often rare for specific pathogens, because most of the patients admitted to the hospital are discharged or die without such an infection. Death and discharge are competing events to acquiring an infection, because these individuals are no longer at risk of getting a hospital-acquired infection. Therefore, the data is best analysed with an extended survival model - the extended illness-death model. A common problem in cohort studies is the costly collection of covariate values. In order to provide efficient use of data from infected as well as uninfected patients, we propose a tailored case-cohort approach for the extended illness-death model. The basic idea of the case-cohort design is to only use a random sample of the full cohort, referred to as subcohort, and all cases, namely the infected patients. Thus, covariate values are only obtained for a small part of the full cohort. The method is based on existing and established methods and is used to perform regression analysis in adapted Cox proportional hazards models. We propose estimation of all cause-specific cumulative hazards and transition probabilities in an extended illness-death model based on case-cohort sampling. As an example, we apply the methodology to infection with a specific pathogen using a large cohort from Spanish hospital data. The obtained results of the case-cohort design are compared with the results in the full cohort to investigate the performance of the proposed method. Copyright © 2016 John WileySons, Ltd.

Introduction: There are limited data about the use of antifungal agents (AF) in critically ill patients and treatment trends since the inclusion of the new generation AF. The use of these agents may have a significant influence on the development of new resistances. Methods: Observational prospective study of the systemic use of AF in patients admitted to Spanish intensive care units (ICU) participating in the ENVIN-HELICS register, from 2006 to 2010. The annual

Purpose To explore the impact of length-biased sampling on the evaluation of risk factors of nosocomial infections (NIs) in point-prevalence studies. Methods We used cohort data with full information including the exact date of the NI and mimicked an artificial 1-day prevalence study by picking a sample from this cohort study. Based on the cohort data, we studied the underlying multistate model which accounts for NI as an intermediate and discharge/death as competing events. Simple formulas are derived to display relationships between risk, hazard, and prevalence odds ratios. Results Due to length-biased sampling, long stay and thus sicker patients are more likely to be sampled. In addition, patients with NIs usually stay longer in hospital. We explored mechanisms that are—due to the design—hidden in prevalence data. In our example, we showed that prevalence odds ratios were usually less pronounced than risk odds ratios but more pronounced than hazard ratios. Conclusions Thus, to avoid misinterpretation, knowledge of the mechanisms from the underlying multistate model is essential for the interpretation of risk factors derived from point-prevalence data.

SummaryBackground: Sampling from a large cohort in order to derive a subsample that would be sufficient for statistical analysis is a frequently used method for handling large data sets in epidemiological studies with limited resources for exposure measurement. For clinical studies however, when interest is in the influence of a potential risk factor, cohort studies are often the first choice with all individuals entering the analysis.Objectives: Our aim is to close the gap between epidemiological and clinical studies with respect to design and power considerations. Schoenfeld’s formula for the number of events required for a Cox’ proportional hazards model is fundamental. Our objective is to compare the power of analyzing the full cohort and the power of a nested case- control and a case-cohort design.Methods: We compare formulas for power for sampling designs and cohort studies. In our data example we simultaneously apply a nested case-control design with a varying number of controls matched to each case, a case cohort design with varying subcohort size, a random subsample and a full cohort analysis. For each design we calculate the standard error for estimated regression coefficients and the mean number of distinct persons, for whom covariate information is required.Results: The formula for the power of a nested case-control design and the power of a case-cohort design is directly connected to the power of a cohort study using the well known Schoenfeld formula. The loss in precision of parameter estimates is relatively small compared to the saving in resources.Conclusions: Nested case-control and case-cohort studies, but not random subsamples yield an attractive alternative for analyzing clinical studies in the situation of a low event rate. Power calculations can be conducted straightforwardly to quantify the loss of power compared to the savings in the number of patients using a sampling design instead of analyzing the full cohort.

Summary Background Data validation is an essential aspect for the accuracy of a nosocomial infection surveillance registry. Aim To report the results of the first quality control programme in the national surveillance programme of intensive care unit (ICU)-acquired infection in Spain (ENVIN–HELICS registry). Method During 2008, of 13,824 records included in the database, 1500 (10.8%) registries from 20 ICUs were reviewed. These ICUs were selected at random and stratified according to the number of cases included in the registry. The proportion of infected patients, which was 9.6% [95% confidence interval (CI) 8.09–11.16], was maintained during the selection of cases for review. Two physicians were trained for the purpose of the study and undertook the review. Results Overall sensitivity, specificity and positive and negative predictive values of the ENVIN–HELICS registry for the identification of patients with any device-related infection acquired during their ICU stay were 86.0% (95% CI 80.0–92.0), 98.7% (95% CI 82.19–93.6), 87.9% (95% CI 82.19–93.6) and 98.5% (95% CI 97.8–99.2), respectively, with a kappa index of 0.85 (95% CI 0.79–0.92). Secondary bloodstream infection had the lowest sensitivity (59.3%), and intubation-associated pneumonia had the highest sensitivity (86.3%). Conclusion There was good correlation between data reported by the registrars and data validated by auditors, confirming the reliability of the ENVIN–HELICS registry.

The appearance of new antimicrobials with activity against Gram-positive multiresistant cocci and knowledge of the limitations of glycopeptides has represented an important change in the use of these antibiotics.To analyze at the national level changes in the use of antibiotics with specific activity against Gram-positive multiresistant cocci in critically ill patients admitted to the ICU as well as the characteristics of patients treated with these agents and the forms of administration.Retrospective cohort study of patients admitted to the ICU for more than 24 hours between 2008 and 2010 in the ENVIN-HELICS national registry. Cases were defined as patients who had received one or more of the following antibiotics: vancomycin, teicoplanin, linezolid or daptomycin. The characteristics of patients who used one or more of these agents were compared with those treated with other antibiotics. Indications and forms of use of each antibiotic were assessed. Descriptive results are presented.A total of 45,757 patients, 27,982 (61.2%) of whom received 63,823 antimicrobials were included in the study. In 6,368 (13.9%) patients, one or more antibiotics specifically active against Gram-positive multiresistant cocci were given. There was a predominance of the use of vancomycin and linezolid and an important increase in the prescription of daptomycin (+320%) and linezolid (+22.4%). In more than 95% of cases, linezolid and daptomycin were prescribed for the treatment of infections, whereas vancomycin and teicoplanin were used for prophylaxis in 20-25% of cases. Between 75% and 80% of indications for treating infections, antibiotics were used empirically except for daptomycin which was used as a directed treatment in 43% of the cases. Only in one third of the indications for empirical treatment, susceptible microorganisms were identified (appropriate treatment).The use of antibiotics with activity against Gram-positive multiresistant cocci remained stable around 14% of all indications. The use of vancomycin and linezolid predominated and there was a clear trend towards an increase in the use of daptomycin and linezolid and a decrease in the use of glycopeptides. Empirical treatments were considered appropriate in only one third of cases.