Documents

MEDICINA BASADA EN EVIDENCIA.pdf

Categories
Published
of 9
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Share
Description
M A J O R A R T I C L E Impact of an Evidence-Based Bundle Intervention in the Quality-of-Care Management and Outcome of Staphylococcus aureus Bacteremia Luis E. López-Cortés, 1,a Maria Dolores del Toro, 1,2 Juan Gálvez-Acebal, 1,2 Elena Bereciartua-Bastarrica, 3 María Carmen Fariñas, 4 Mercedes Sanz-Franco, 5 Clara Natera, 6 Juan E. Corzo, 7 José Manuel Lomas, 8 Juan Pasquau, 9 Alfonso del Arco, 10 María Paz Martínez, 11 Alberto Romero, 12 Miguel A. Muniain, 1,2,14 Marina de Cueto, 1,2 Álvaro
Transcript
  M A J O R A R T I C L E Impact of an Evidence-Based BundleIntervention in the Quality-of-Care Managementand Outcome of   Staphylococcus aureus Bacteremia Luis E. López-Cortés, 1,a Maria Dolores del Toro, 1,2 Juan Gálvez-Acebal, 1,2 Elena Bereciartua-Bastarrica, 3 María Carmen Fariñas, 4 Mercedes Sanz-Franco, 5 Clara Natera, 6 Juan E. Corzo, 7 José Manuel Lomas, 8 Juan Pasquau, 9 Alfonso del Arco, 10 María Paz Martínez, 11 Alberto Romero, 12 Miguel A. Muniain, 1,2,14 Marina de Cueto, 1,2 Álvaro Pascual, 1,2,13 and Jesús Rodríguez-Baño; 1,2,14 for the REIPI/SAB group b 1 Unidad Clínica de Enfermedades Infecciosas y Microbiología, Hospital Universitario Virgen Macarena, Sevilla;  2 Spanish Network for Research inInfectious Diseases, Instituto de Salud Carlos III, Madrid;  3 Unidad de Enfermedades Infecciosas, Hospital de Cruces, Baracaldo;  4 Unidad deEnfermedades Infecciosas, Hospital Universitairo Marqués de Valdecilla, Universidad de Cantabria – IFIMAV, Santander, Cantabria;  5 Unidad deEnfermedades Infecciosas, Hospital de San Pedro, La Rioja;  6 Unidad de Enfermedades Infecciosas, Hospital Universitario Reina Sofía, Córdoba; 7 Unidad de Gestión Clínica de Enfermedades Infecciosas y Microbiología, Hospital Virgen de Valme, Sevilla;  8 Unidad de Enfermedades Infecciosas,Hospital Juan Ramón Jiménez, Huelva;  9 Unidad de Enfermedades Infecciosas y Microbiología Clínica, Hospital Virgen de las Nieves, Granada; 10 Unidad de Enfermedades Infecciosas, Hospital Costa del Sol, Málaga;  11 Unidad de Enfermedades Infecciosas, Hospital de Torrecárdenas, Almería; 12 Unidad de Enfermedades Infecciosas, Hospital Universitario de Puerto Real, Cádiz; and  13 Departamento de Microbiología and  14 Departamento deMedicina, Universidad de Sevilla, Spain (See the Editorial Commentary by Liu on pages 1234 – 6.) Background  .  Staphylococcus aureus  bacteremia (SAB) is associated with signi 󿬁 cant morbidity and mortality.Several aspects of clinical management have been shown to have signi 󿬁 cant impact on prognosis. The objective of the study was to identify evidence-based quality-of-care indicators (QCIs) for the management of SAB, and toevaluate the impact of a QCI-based bundle on the management and prognosis of SAB.  Methods .  A systematic review of the literature to identify QCIs in the management of SAB was performed. Then,the impact of a bundle including selected QCIs was evaluated in a quasi-experimental study in 12 tertiary Spanish hos-pitals. The main and secondary outcome variables were adherence to QCIs and mortality. Speci 󿬁 c structured individu-alized written recommendations on 6 selected evidence-based QCIs for the management of SAB were provided. Results .  A total of 287 and 221 patients were included in the preintervention and intervention periods, respective-ly. After controlling for potential confounders, the intervention was independently associated with improved adher-ence to follow-up blood cultures (odds ratio [OR], 2.83; 95% con 󿬁 dence interval [CI], 1.78 – 4.49), early source control(OR, 4.56; 95% CI, 2.12 – 9.79), early intravenous cloxacillin for methicillin-susceptible isolates (OR, 1.79; 95% CI,1.15 – 2.78), and appropriate duration of therapy (OR, 2.13; 95% CI, 1.24 – 3.64). The intervention was independentlyas-sociated with a decrease in 14-day and 30-day mortality (OR, 0.47; 95% CI, .26 – .85 and OR, 0.56; 95% CI, .34 – .93, re-spectively). Conclusions .  A bundle orientated to improving adherence to evidence-based QCIs improved the management of patients with SAB and was associated with reduced mortality. Keywords .  Staphylococcus aureus ; intervention; bacteremia; bloodstream infections; clinical management. Received 14 March 2013; accepted 17 June 2013; electronically published 8August 2013. a Present af 󿬁 liation: Unidad Clínica de Enfermedades Infecciosas, MicrobiologíaClínica y Medicina Preventiva, Hospitales Universitarios Virgen del Rocío, Sevilla, Spain. b Other authors from the REIPI/SAB group are listed in the Acknowledgments.Correspondence: Jesús Rodríguez-Baño, PhD, MD, Unidad Clínica de Enferme-dades Infecciosas y Microbiología, Avda Dr Fedriani 3, 41009 Sevilla, Spain(jesusrb@us.es). Clinical Infectious Diseases 2013;57(9):1225 – 33 © The Author 2013. Published by Oxford University Press on behalf of the InfectiousDiseases Society of America. All rights reserved. For Permissions, please e-mail:journals.permissions@oup.com.DOI: 10.1093/cid/cit499 Evidence-Based Bundle on SAB  ã  CID 2013:57 (1 November)  ã  1225   b  y g u e  s  t   on O c  t   o b  e r 2  3  ,2  0 1 4 h  t   t   p :  /   /   c i   d  . oxf   or  d  j   o ur n a l   s  . or  g /  D o wnl   o a  d  e  d f  r  om   Staphylococcus aureus  is an important human pathogen andone of the leading causes of both nosocomial and community-onset bloodstream infections worldwide [1].  Staphylococcusaureus  bacteremia (SAB) causes signi 󿬁 cant morbidity, mortali-ty, and healthcare costs; complications are frequent, and mor-tality ranges from 20% to 40% [2 – 5].Importantly, some aspectsof clinical management have been associated with better out-comes [6 – 8]. Thus, previous studies showed that adherence tospecialized advice is associated with improved managementand, in some of them, even reduced mortality [9 – 15]. In thesestudies, the management and outcomes of patients with SABwho were treated following the recommendations of infectiousdiseases specialists were compared with those of patients forwhom specialized consultation was not sought or where therecommendations provided were not followed. The recommen-dations provided by specialists in these studies were not struc-tured and/or had not been prioritized in accordance with anevidence-based procedure.At present, many tertiary hospitals develop active  “ bactere-mia programs, ”  in which infectious diseases specialists andclinical microbiologists provide early unsolicited advice for themanagement of patients with bacteremia. Despite this, the spe-ci 󿬁 c dif  󿬁 culties inherent to management of SAB may need ad-ditional interventions. The objectives of this study were (1) toidentify evidence-based quality-of-care indicators (QCIs) forthe management of SAB; and (2) to evaluate the impact of anintervention based on a bundle of selected QCIs aimed at im-proving the management and outcome of patients with SAB. METHODS Identi 󿬁 cation of Quality-of-Care Indicators for the Managementof SAB A systematic review of the literature was performed to identify the best evidence on aspects related to the clinical managementof SAB that had a signi 󿬁 cant in 󿬂 uence in prognosis. Studieswere retrieved from the PubMed database using the following search terms:  Staphylococcus aureus  OR   S. aureus , AND  bacter-emia  OR   bloodstream infection  OR   sepsis , AND  outcome  OR  complication  OR   mortality   OR   death  OR   recurrence . Observa-tional and randomized studies were selected if the 2 following criteria were ful 󿬁 lled: the predictors or risk factors for outcomedeterminants (including rates of clinical cure, microbiologicalcure, mortality, complications or recurrence) were studied, andaccepted methods for control of confounding were used in thecase of observational studies (including multivariate or strati- 󿬁 ed analysis or matching). The studies were reviewed by 2 in- vestigators (L.E.L.-C. and J.R.-B.). The variables independently and consistently (eg, they were found in at least 2 studies)related to outcome and amenable to clinical intervention wereselected as QCIs; for each of them a formula to measure thelevel of adherence to the indicator was de 󿬁 ned. Intervention: Study Design and Setting The intervention study was performed in 12 tertiary hospitalsin Spain; 8 of them are teaching hospitals, and 10 have >500beds. There are infectious diseases services or units in all 12,and active transplantation programs in 4. A quasi-experimentaldesign, before (from January through June 2010) and during the implementation of the intervention (from July to December2010), was used; in one hospital where the intervention waspiloted, the preintervention and intervention periods werefrom March 2008 to August 2009, and from September 2009 toMay 2011, respectively. All episodes of SAB involving admittedpatients >17 years of age were considered eligible. Patients weredetected through daily review of microbiology reports. Only 1episode per patient (the  󿬁 rst) was included, unless a laterepisode was separated from the previous one by an interval of >3 months without evidence of recurrence from a deep-seated Table 1. Preintervention and Intervention Activities Performedon Patients With  Staphylococcus aureus   Bacteremia in the Partic-ipating Hospitals Period ActivitiesPreintervention Early report (verbal or written) of Gram stainresultswas provided forall patients withpositive blood cultures by clinicalmicrobiologists in 6 of the 12 participatingcenters. Seven centers had an active “ bacteremia program ”  in which unsolicitedconsultation forall SAB casesof BSI wereprovided by infectious diseases subspecialists;neither the recommendations providednor thefollow-upprocedures were structured, but weredone at the discretion of the infectious diseasessubspecialist. Adherence to recommendationswas not prospectively measured.Intervention 1. The intervention was explainedto the differentservicesin specific educational sessions. Aninformative letter was also sent to all heads ofservicesbefore the intervention period started.2. Specific recommendations, based on the 6selectedquality-of-care indicators, werespecifically provided at least 3 days per week byan infectious diseasesspecialists from the day S. aureus   was identified from blood culture untilthe patient was dischargedor died. Therecommendations were discussedwith theattendingphysicianand were also provided in astructured form which was addedto the charts(Supplementary Figure 1), and signedby theinfectiousdiseases specialist at each visit.Adherence to the recommendations was at thediscretionof attending physician.3. The form also includeda summaryof therationale for the intervention, whichserved aseducational material. Abbreviations: BSI, bloodstream infection; SAB,  Staphylococcus aureus  bacteremia. 1226  ã  CID 2013:57 (1 November)  ã  López-Cortés et al   b  y g u e  s  t   on O c  t   o b  e r 2  3  ,2  0 1 4 h  t   t   p :  /   /   c i   d  . oxf   or  d  j   o ur n a l   s  . or  g /  D o wnl   o a  d  e  d f  r  om   infection. Patients who died in the  󿬁 rst 48 hours (who were notsubject to intervention) and those patients receiving palliativecare for terminal conditions were excluded.The intervention and the activities performed during thepreintervention period are summarized in Table 1. The inter- vention consisted of a set of written recommendations accord-ing to the 6 aspects selected as QCIs provided in a structuredform by an infectious diseases specialist at each hospital. All pa-tients were followed until discharge or death and were assessedfor survival and recurrence on days 30 and 90 during a visit tothe outpatient clinic or by phone call. Patient data were collect-ed by a nonblinded investigator in each of the participating hospitals.The study was approved by the ethics committee of the Hos-pital Universitario Virgen Macarena, which waived the need toobtain written informed consent from the patients on the un-derstanding that the intervention was aimed at improving quality of care according to evidence-based standard of care. Variables and De 󿬁 nitions The main outcome variable of the quasi-experimental study was adherence to the 6 QCIs selected, measured as the propor-tion of cases in which the recommended action was performed.As secondary outcome variables, 14- and 30-day all-cause mor-tality and the 90-day recurrence rate were considered. Explana-tory variables included demographics, type and severity of underlying conditions, acquisition type of SAB, source of infec-tion, severity of systemic in 󿬂 ammatory response syndrome atpresentation [16], susceptibility to methicillin, antimicrobialtherapy, support therapy, and outcome [17]. We used theCharlson comorbidity index to measure the severity of chronicunderlying conditions [18], validated as predictive of mortality among patients with SAB [3]. Acute severity of illness was as-sessed using the Pitt bacteremia score, measured retrospectively on the day before SAB was diagnosed, which has also been vali-dated as a predictor for mortality in SAB [17].Type of acquisitionwas classi 󿬁 ed as community-associated, healthcare-associated, or Table 2. De 󿬁 nitions of Quality-of-Care Indicators for  Staphylococcus aureus   Bacteremia Selected After Systematic Review of theLiterature Quality-of-Care Indicator Definition FormulaReferences inSupplementaryDataFollow-upblood cultures Performanceof control blood cultures 48 – 96 hafterantimicrobial therapy was startedregardless of clinical evolutionPatients in whom follow-up blood cultureswere collected ×100/patients aliveat 96 h[9 – 11][14] [15] [21]Earlysource control Removal of nonpermanent vascularcatheterwhenever the catheterwas suspectedorconfirmed asthe source of SAB, or drainageof an abscess in <72 hPatients in which the amenablesourcewas removed in <72 h ×100/patientswith source amenableof removal/ drainage[9 – 11] [13]Echocardiographyinpatients with clinicalindicationsPerformanceof echocardiographyin patientswith complicated bacteremia (see definitionin Methods) or predisposing conditions forendocarditisPatients with echocardiography×100/ patients with complicated bacteremiaorpredisposingconditionforendocarditis,alive at least 96 h[9] [10] [12 – 15][71 – 75]Early use of intravenouscloxacillin for MSSA asdefinitive therapyDefinitive therapywith intravenouscloxacillin(at least 2 g every 6 h oradjusted based onrenal function in renal failure) in casesofmethicillin-susceptiblestrains (allergicpatients excluded). Treatment shouldbestarted within the first 24 h after methicillinsensitivity was available. For hemodialysispatients, cefazolin 2 g aftereachhemodialysis sessionwas acceptableDefinitive therapy with intravenouscloxacillin ×100/nonallergic patientswithmethicillin-susceptibleisolates[9] [13] [79][81]Adjustment of vancomycindose accordingto troughlevelsMeasurementof trough levels of vancomycinin patientstreated forat least 3 d with thisantibiotic and adjustment of dose in order toachieve plasmatrough levels between 15and 20 mg/L in survivorsPatients with trough level of vancomycindetermined and doseadjusted ×100/ patientstreated with vancomycin foratleast 3 d[24] [59] [76 – 80]Treatment durationaccording to thecomplexityof infectionDuration of antimicrobialtherapyof at least 14d for uncomplicated bacteremia and 28 d forcomplicated bacteremia. Sequential oraltreatment with fluoroquinolone plusrifampin, trimethoprim-sulfamethoxazole, orlinezolid was considered acceptedinselectedcasesPatients with appropriate duration oftherapy ×100/patients alive at 14 or 28 din casesof uncomplicated orcomplicated bacteremia, respectively[10] [12 – 14] [21][78] Abbreviations: MSSA, methicillin-susceptible  Staphylococcus aureus  ; SAB,  Staphylococcus aureus   bacteremia. Evidence-Based Bundle on SAB  ã  CID 2013:57 (1 November)  ã  1227   b  y g u e  s  t   on O c  t   o b  e r 2  3  ,2  0 1 4 h  t   t   p :  /   /   c i   d  . oxf   or  d  j   o ur n a l   s  . or  g /  D o wnl   o a  d  e  d f  r  om   nosocomial, following Friedman criteria [19]. Primary sources of SAB were de 󿬁 ned according to the Centers for Disease Controland Prevention [20], and evaluated in consensus by 2 investiga-tors in each on the participant centers. Sources of SAB associatedwith high mortality in previous studies were classi 󿬁 ed as high-risk sources;these includedendocarditis,endovascularinfectionsotherthan catheter-related, central nervous system infections, intra-abdominal infections, and respiratory tract infections [21, 22]. We considered empirical antibiotic treatment as appropriate if at least 1 active drug according to in vitro susceptibility resultshad been initiated in the  󿬁 rst 12 hours after the blood culturewas obtained. Persistent SAB was de 󿬁 ned as the isolation of   S.aureus  in blood cultures obtained from peripheral veins for  ≥ 3days despite active antimicrobial therapy according to suscepti-bility testing. For the purpose of clinical decisions, SAB wasconsidered as complicated if any of the following criteria werepresent: persistent bacteremia; development of endocarditis ormetastatic foci; presence of Janeway lesions, Osler nodes, orother cutaneous or mucosal lesions suggestive of acute systemicinfection (including petechiae, vasculitis, infarcts, ecchymoses,pustules, Roth spots, or conjunctival hemorrhage) in theabsence of a  󿬁 rm alternate explanation [2];presence of any per-manent prosthetic device; any device-related infection wherethe device could not be removed in the  󿬁 rst 3 days; and SAB inpatients under chronic hemodialysis [2, 23 – 28]. We includedthe variable  “ unfavorable clinical course ”  to re 󿬂 ect the clinicalsituation in the same day the intervention was started (typically,48 hours after the blood cultures were taken); it was de 󿬁 nedas worsening or lack of evident improvement in the signs of sepsis [16] with regard to the situation the day the blood cul-tures were taken. Cure was de 󿬁 ned as the absence of all signsand symptoms of infection and a negative blood culture at theend of antibiotic therapy [29]. Recurrence was de 󿬁 ned as theisolation of   S. aureus  with the same susceptibility pattern fromblood cultures or from a deep-seated focus in the following 3months after clinical cure had been reached. Microbiological Studies The recommendations of the Spanish Society of Infectious Dis-eases and Clinical Microbiology were followed for performing,processing, and interpreting the blood cultures [30, 31]. Sus- ceptibility testing was performed using accepted methods ateach hospital. Statistical Analysis Crude comparisons were performed using the  χ 2 or Fisherexact tests for percentages, as appropriate, and the Mann-Whitney   U   test for continuous variables. Relative risks with95% con 󿬁 dence intervals (Cis) were calculated for the crudeanalysis of adherence to the indicators in the preinterventionand intervention periods. Multivariate analyses were performedusing logistic regression. Variables were selected using thebackward stepwise procedure;  P   values <.2 and <.1 were used ascutoffs for including and deleting variables in the models, re-spectively. The predictive ability of the models was studied by the area under the receiver operating characteristic curves.Effect modi 󿬁 cations between the exposure of interest and other variables were investigated. The software used for the analysiswas the SPSS v17.0 package. RESULTS Systematic Review and De 󿬁 nition of Quality-of-Care Indicators The search strategy retrieved 2828 articles. After reviewing theabstracts, 184 articles were fully reviewed and 81 were selectedaccording to the preestablished criteria (see references in Sup-plementary Data). Six aspects of clinical management from 16articles showing an impact on outcome were selected as QCIs(Table 2): performance of follow-up blood cultures; early source control; performance of echocardiography in patientswith speci 󿬁 c criteria; early use of intravenous cloxacillin incases of methicillin-susceptible  S. aureus  (MSSA) (or cefazolinin patients under hemodialysis) as de 󿬁 nitive therapy in nonal-lergic patients; adjustment of vancomycin dose according totrough levels; and provision of an appropriate duration of therapy according to the complexity of infection. The de 󿬁 ni-tions of QCIs and the formulas used to measure them areshown in Table 2. Analysis of the Impact of Intervention During the study period, 536 episodes of SAB were diagnosedin adult patients admitted to the participating hospitals andwere considered eligible for the study; 28 cases were excluded(Figure 1), so that 287 episodes were  󿬁 nally included in the pre-intervention period and 221 in the intervention period. Thebaseline demographics and clinical characteristics of the pa-tients in both periods are shown in Table 3. The proportion of  vascular catheter – related episodes was higher in the intervention Figure 1.  Flow chart of patients included in the multicenter quasi-experimental study. 1228  ã  CID 2013:57 (1 November)  ã  López-Cortés et al   b  y g u e  s  t   on O c  t   o b  e r 2  3  ,2  0 1 4 h  t   t   p :  /   /   c i   d  . oxf   or  d  j   o ur n a l   s  . or  g /  D o wnl   o a  d  e  d f  r  om 
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks