Politics

Diagnostic accuracy and potential clinical value of the LightCycler SeptiFast assay in the management of bloodstream infections occurring in neutropenic and critically ill patients

Description
Diagnostic accuracy and potential clinical value of the LightCycler SeptiFast assay in the management of bloodstream infections occurring in neutropenic and critically ill patients
Categories
Published
of 6
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.
Related Documents
Share
Transcript
  Diagnostic accuracy and potential clinical value of the LightCycler SeptiFastassay in the management of bloodstream infections occurring in neutropenicand critically ill patients Dayana Bravo a , Jose´ Blanquer b , Mar Tormo c , Gerardo Aguilar d , R afael Borra´s a,e ,Carlos Solano c , Marı´a A. Clari a , Elisa Costa a , Beatriz Mun˜oz-Cobo a ,Mo´nica Argu¨eso b , Jose´ Roberto Pineda b , David Navarro a,e, * a Microbiology Service, Hospital Clı´ nico Universitario, Av. Blasco Iba´ n˜ ez 17, 46010 Valencia, Spain b Intensive Care Unit, Hospital Clı´ nico Universitario, Valencia, Spain c Hematology and Oncology Service, Hospital Clı´ nico Universitario, Valencia, Spain d  Anesthesiology and Critical Care Department, Hospital Clı´ nico Universitario, Valencia, Spain e Department of Microbiology, School of Medicine, University of Valencia, Spain 1. Introduction Bloodstream infections are associated with significant morbid-ityandmortalityinbothneutropenicandcriticallyillpatients. 1,2 Afavorable clinical outcome is directly related to rapid etiologicaldiagnosis and prompt administration of effective antimicrobialtherapy. 3,4 Early therapeutic management of febrile episodes inthese patients includes the empirical administration of broad-spectrum antimicrobial agents, since the results of blood cultures,currentlythegoldstandardmethodforthedetectionofbacteremiaand fungemia, are not available until 48–72 h after samplecollection. 5 Reassessment of antimicrobial therapy upon receiptof blood culture results is then recommended, as a more targetedantimicrobial regimen minimizes toxicity, reduces costs, andlowers the probability of the emergence of microbial resistance.Conventional blood culture methods have several drawbacks,such as the possibility of providing false-positive results due tocontamination and false-negative results due to their relativeinsensitivity, particularly in patients on antimicrobial therapy atthe time of sampling. 5 In this sense, it has been shown that only30% of blood cultures obtained at the onset of fever are positive infebrile neutropenic patients. 2,6 PCR-based methods detecting microbial DNA directly in bloodspecimens are a promising approach for the rapid diagnosis of bacteremia and fungemia. 7 Among these, the LightCycler 1 SeptiFast Test, a commercially available multiplex real-time PCR assay able to detect up to 25 bacterial and fungal organismscommonly involved in systemic infections, has recently been thefocus of much attention. The performance of the SeptiFast test hasbeen evaluated in different clinical settings. 8–20 Overall, the International Journal of Infectious Diseases 15 (2011) e326–e331 A R T I C L E I N F O  Article history: Received 10 November 2010 Received in revised form 13 December 2010 Accepted 8 January 2011 Corresponding Editor:  J. Peter Donnelly,Nijmegen, the Netherlands Keywords: Real-time PCR Diagnostic accuracy of the SeptiFast assayBloodstream infectionBlood cultureNeutropeniaCritically ill patients S U M M A R Y  Objectives:  TheobjectivesofthisstudyweretocomparetheperformanceoftheLightCyclerSeptiFastTestMGRADE and conventional blood culture in the etiological diagnosis of febrile episodes occurring inneutropenic and critically ill patients (in the intensive care unit; ICU), and to assess the potential clinicalvalue of the SeptiFast test in patient management. Methods:  A total of 86 febrile episodes occurring in 33 neutropenic patients and 53 ICU patients wereanalyzed.BloodsamplesforbloodcultureandSeptiFasttestingwereobtainedattheonsetoffever,beforethe implementation of empirical antimicrobial therapy. Results:  The overall microorganism-to-isolate agreement between the SeptiFast test and blood culturewas 69% ( k = 0.37) in neutropenic patients and 75% ( k = 0.56) in ICU patients. The sensitivity of theSeptiFast assay for clinically relevant episodes of bacteremia and fungemia was 62% in neutropenicpatientsand70%inICUpatients.BasedonSeptiFastresults,empiricaltreatmentsweredeemedadequatein all but one of the febrile episodes. Nevertheless, early antibiotic treatment readjustment was judgedfeasible in most of clinically significant episodes overall. Conclusions:  The SeptiFast assay is a valuable ancillary method for the diagnosis of bloodstreaminfections in neutropenic and ICU patients. In these clinical settings, results of the SeptiFast assay maylead to a more targeted antibiotic therapy early after the onset of fever.   2011 International Society for Infectious Diseases. Published by Elsevier Ltd. All rights reserved. * Corresponding author. Tel.: +34 96 3864657; fax: +34 96 3864173. E-mail address:  david.navarro@uv.es (D. Navarro). Contents lists available at ScienceDirect International Journal of Infectious Diseases journal homepage: www.elsevier.com/locate/ijid 1201-9712/$36.00 – see front matter    2011 International Society for Infectious Diseases. Published by Elsevier Ltd. All rights reserved.doi:10.1016/j.ijid.2011.01.003  sensitivityoftheSeptiFasttesthasbeenreportedtobehigherthanthat of blood culture. Nevertheless, further studies are needed inorder to determine whether the information provided by theSeptiFast test leads to the improved therapeutic management of patients with suspected bloodstream infections.InthepresentstudyweevaluatedthediagnosticaccuracyoftheSeptiFast test in comparison with that of conventional bloodculture in the etiological diagnosis of febrile episodes occurring inneutropenic and critically ill patients. We also retrospectivelyassessed the potential utility of the information provided by theSeptiFast assay in the therapeutic management of these patientsearly after the onset of fever. 2. Materials and methods  2.1. Patients This was an observational study conducted in the HematologyUnit and the Medical and Surgical ICU of the Hospital Clı´nicoUniversitario, Valencia (Spain) over a 8-month period. Theinclusion criterion used was the development of a febrile episodethat required hospital admission or occurred during hospital stay.The exclusion criterion was the receipt of empirical antibiotictreatment prior to blood sampling for analysis. Patients who metthe inclusion criterion and who were under the care of thephysicians listed among the authors of this manuscript wererecruited into the study. The study was conducted following localethics rules.A total of 86 non-consecutive patients were enrolled betweenFebruary and September 2009. The Hematology Unit cohortconsisted of 33 patients (median age 58 years, range 24–80 years;18malesand15females).Theunderlyingclinicalconditionsofthepatients were as follows: hematological malignancy underchemotherapy ( n  = 18), allogeneic stem cell transplantation( n  = 10), and autologous stem cell transplantation ( n  = 5). Patientsin the Hematology Unit presented with febrile (axillary tempera-ture  > 38.0  8 C) neutropenia ( < 0.5  10 9 neutrophils/l) either of unknown srcinorrelated to peripheral orcentral venouscatheterimplantation. Blood samples were drawn at the onset of fever forblood culture (two sets, following local guidelines) and SeptiFasttesting. Patients were on antibacterial (fluoroquinolones; exceptfor autologous transplant patients) and antifungal (posaconazole)prophylaxis at the time of sampling, and were treated empiricallyafter blood collection with either ceftazidime plus amikacin ormeropenem at conventional doses upon the appearance of fever.Vancomycinwas later addedif Gram-positive cocciwere observedon direct smears from two positive blood cultures, or from onepositivebloodcultureifeithermucositisgrade  2waspresentoracatheter-related bacteremia was suspected (presence of phlebitisor signs of inflammation at the insertion site).A total of 53 patients were included in the ICU cohort (medianage 65.5 years, range 23–86 years; 33 males and 20 females).Patients were admitted to the ICU after heart surgery ( n  = 8),neurosurgery or cerebrovascular accident ( n  = 11), or due to acuterespiratory failure ( n  = 11), severe sepsis or septic shock of pulmonary, urinary or soft tissue srcin ( n  = 15), drug intoxication( n  = 3), or other causes ( n  = 5). Patients in the ICU either presentedwith fever or developed fever during their ICU stay. Blood samplesfor blood cultures (three sets) and SeptiFast testing were obtainedat the onset of fever, before the implementation of empiricalantimicrobial therapy. Febrile episodes in these patients weretherapeutically managed at the physician’s discretion on the basisofthesuspectedetiology,localmicroorganismresistancepatterns,theseverityofillness,andthenatureofpriorantibiotictreatments.Antimicrobial therapy administered to patients is described whendeemed appropriate.  2.2. Microbiological studies Blood was drawn by sterile venous puncture and processedusing the automated continuous monitoring blood culture systemBactec 9420 BC (Becton Dickinson, Heidelberg, Germany). Each setof blood cultures consisted of a pair of bottles for aerobic andanaerobicbacteriaandanadditionalbottleforfungalrecovery.Theblood cultures were incubated for a maximum of 7 days. Eachbottle was inoculated with 10 ml of blood. The different sets of blood cultures were obtained at intervals of 30 min. Direct smearexamination (Gram staining) was performed from positive bloodcultures as soon as detected, the clinicians being immediatelynotified of the results by telephone. This usually occurred  > 16 hafter blood culture incubation. Direct antimicrobial susceptibilitytesting was then performed by the disk diffusion method. Isolatedmicroorganisms were identified by standard methods and criteria.Ethylenediaminetetraacetic acid (EDTA)-whole blood wassimultaneously obtained from patients for the detection of microbial DNA using the LightCycler SeptiFast Test MGRADE(Roche Diagnostics GmbH, Manheim, Germany), as recommendedbythemanufacturer.TotalDNAwasextractedfrom1.5 mlofbloodby means of the SeptiFast Lys Kit on the MagNA Lyser instrument(Roche Diagnostics). Real-time detection of PCR products wasperformed using the LightCycler 2.0 instrument (Roche Diagnos-tics). A full description of the procedure, as well as the targetmicrobial species of the SeptiFast test, have previously beenreported. 8,14 PCR testing was performed retrospectively on speci-mens that had been stored at   20  8 C so that results had noinfluence on antimicrobial therapy. A single blood sample perfebrile episode was obtained for SeptiFast testing.Microbiological cultures of samples from other sites (lowerrespiratory tract specimens, urine, or catheter tips) were per-formed when clinically indicated. Ventilator-associated pneumo-nia was diagnosed following previously published criteria. 21 Personnel performing the SeptiFast assay were blinded to theresults of the blood culture.  2.3. Definitions and interpretation of the data Significant bacteremia or fungemia was defined by microbialgrowth in one or more blood culture sets. For typical microbialcontaminants, such as coagulase-negative  Staphylococcus  species(CoNS),  Streptococcus spp ,  Corynebacterium spp , or  Bacillus spp , theisolation of the same bacterial species with an identicalantimicrobial susceptibility profile in at least two blood culturesfromdifferentbloodsamplingswasrequired.Inpatientsattendingthe Hematology Unit, bacterial isolation in one blood culturecoupled with the isolation of the same species with an identicalantimicrobialsusceptibilitypatternonasemiquantitativecathetertip culture was also considered to represent a significantbacteremia. In this clinical setting, the significance of either theisolation of a potentially contaminating microorganism in a singleset of blood cultures or the detection of CoNS DNA in blood by theSeptiFast assay was judged on the basis of clinical grounds.Discordant results between the assays were resolved on the basisof available clinical (i.e., response to therapy) and microbiological(cultures from other sites) data.Thesensitivity(truepositives/(truepositives + falsenegatives))of the SeptiFast assay was calculated taking only relevant isolates(non-contaminant) into consideration. We used simple Kappastatistics to measure the degree of agreement between the assaysusing SPSS v. 15.0 (SPSS Inc., Chicago, IL, USA).A retrospective analysis of the test results and clinical data wasperformedinordertoassesstheadequacyofempiricaltherapeuticregimens administered to patients and to determine whether theinformation provided by the SeptiFast assay would have had any D. Bravo et al./International Journal of Infectious Diseases 15 (2011) e326–e331  e327  therapeutic relevance. An empirical therapeutic regimen wasdeemed adequate when it provided cover for the causalmicroorganism and the pathogen was found to be susceptible toat least one of the antimicrobials used. The appropriateness of adjustment or de-escalation of empirical antimicrobial therapy onthe basis of the SeptiFast results was assessed taking intoconsideration the local antimicrobial susceptibility profiles of the detected microorganisms. Appropriate streamlining of empir-ical antimicrobial therapy included the following potentialadjustments: (1) suppression of one out of two broad-spectrumantibiotics providing cover for the causal microorganisms (i.e.,meropenem or levofloxacin for  Enterobacteriaceae ), (2) suppres-sion of an antibiotic with activity against beta-lactam-resistantGram-positive cocci (vancomycin, linezolid, or daptomycin) whenGram-negativebacterial DNA wasdetected,and (3)suppression of broad-spectrumantibiotics(i.e.,meropenem)whenGram-positivecocci were detected, provided that an antibiotic with activityagainst beta-lactam-resistant Gram-positive cocci (vancomycin,linezolid,ordaptomycin)wasincludedinthetherapeuticregimen.The appropriateness and feasibility of antibiotic adjustments were judged independently by two of the authors. 3. Results  3.1. Performance of the SeptiFast assay in febrile neutropenic patients Blood samples from 33 febrile neutropenic patients wereexamined. All but three febrile episodes resolved within the studyperiod.DataregardingthemicroorganismsdetectedbytheSeptiFastassay and recovered by blood culture are summarized in Table 1.Twenty-five episodes were associated with a positive bloodculture. A single microorganism was isolated in all cases. Out of these 25 episodes, nine had a positive result by the SeptiFast test,which was concordant with that of the blood culture in six cases(CoNS  n  = 3,  Streptococcus spp, Klebsiella pneumoniae/oxytoca , and Serratia marcescens ) and discordant in the remaining three cases.Microbiological findings in the latter cases were as follows: in onecase  Stenotrophomonas maltophilia  was detected by the SeptiFastassay, while CoNS (contaminant) was isolated from the bloodculture; in another episode,  Acinetobacter baumannii  and  Escher-ichia coli  were detected by the SeptiFast assay, but only the latterwas recovered from the blood culture. The presence of   S.maltophilia  DNA in blood was deemed clinically meaningless asthefebrileepisoderesolvedwithanantibioticregimenthatdidnotprovide cover for the microorganism (meropenem). The clinicalrelevance of   A. baumannii  DNA detection by the SeptiFast assaycould not be ascertained. In the remaining case,  Enterococcus faecium  was isolated from the blood culture and  Staphylococcusaureus  was detected by the SeptiFast assay.Sixteen blood culture-positive episodes had a negative Septi-Fast result. Of these, 12 yielded CoNS, which were interpreted ascontaminants in eight cases. In two cases microorganisms notincluded in the SeptiFast panel list were isolated ( Corynebacteriumspp  and  Bacillus spp ). In the remaining two cases  E. coli  was missedby the SeptiFast assay.No blood culture-negative samples tested positive by theSeptiFastassay,andbothmethodsyieldednegativeresultsineightepisodes. The overall microorganism-to-isolate agreement (nega-tive results in both assays and positive results in both assaysyielding the same microbial species) between the assays was 48%,but increased to 69% ( k = 0.37) if specimens yielding CoNS judgedascontaminantsand microorganismsnotincludedintheSeptiFastpanel list were excluded from the analysis. The sensitivity forclinically relevant microorganisms of the SeptiFast assay was 62%and of blood culture was 74%.  3.2. Potential therapeutic relevance of the information provided bythe SeptiFast assay in febrile neutropenic patients Retrospective analysis of the clinical and microbiological datafor episodes in which both the SeptiFast assay and the bloodculture tested positive indicated the following (Table 2): (1) theempirical treatment was deemed adequate in six out of the sevenclinicallyrelevantfebrileepisodes;intheremainingcasetheresultprovided by the SeptiFast assay would have recommended achange in the empirical antibiotic regimen because of the lack of cover forthe detected microorganism( S. maltophilia ), even thoughthe clinical significance of this finding was later dismissed; (2)adjustment of empirical therapy on the basis of the SeptiFastresults would have been considered appropriate in the above-mentioned case ( S. maltophilia ) and in an additional three febrileepisodes due to  S. aureus ,  Streptococcus mitis  and CoNS, respec-tively, in which either meropenem or the combination of ceftazidime/amikacin could have been suspended; (3) implemen-tationofvancomycinonthebasisofGramstainresultswouldhavebeen avoided in six patients in whom clinically irrelevant CoNSwere isolated had the result of the SeptiFast assay (negative) beentaken into consideration; (4) the SeptiFast assay missed fourclinically relevant CoNS; (v) results of the SeptiFast assay wouldhave led to the inappropriate addition of vancomycin to theempirical antibiotic regimen in two febrile episodes (CoNSdetected were interpreted as clinically irrelevant).  3.3. Performance of the SeptiFast assay in febrile episodes in the ICU  patients Blood samples from 53 febrile episodes were examined. All butseven episodes resolved within the study period. Data regarding  Table 1 Microorganisms detected by the SeptiFast assay and recovered by blood culture in febrile episodes occurring in neutropenic patientsSpecies Number of samples with a positive result by:Blood culture only SeptiFast assay only Both assaysCoagulase-negative staphylococci a 13 0 3 Staphylococcus aureus  0 1 0 Enterococcus faecium  1 0 0 Streptococcus spp / Streptococcus mitis  0 0 1 Corynebacterium spp b 1 - - Bacillus spp b 1 - - Escherichia coli  2 0 1 Serratia marcescens  0 0 1 Klebsiella pneumoniae/oxytoca  0 0 1  Acinetobacter baumannii  0 1 0 Stenotrophomonas maltophilia  0 1 0 a Eight coagulase-negative staphylococci detected only by blood culture and two detected by both methods were judged to be contaminants. b Not included in the SeptiFast master list. D. Bravo et al./International Journal of Infectious Diseases 15 (2011) e326–e331 e328  the microorganismsdetected by the SeptiFast assay and recoveredby blood culture are summarized in Table 3.Twenty-five episodes were associated with a positive bloodculture. A single microorganism was isolated in all cases. Of these,12 had a positive result by the SeptiFast test and 13 had a negativeSeptiFast result. In the former episodes, concordant results wereobtained in 10 cases and discordant results in two cases (CoNS inblood culture and  E. coli  by the SeptiFast assay, and CoNS in bloodcultureand  Aspergillus spp bytheSeptiFastassay). Thedetection of   Aspergillus spp  DNA by the SeptiFast assay was considered as afalse-positive result on the basis of analytical galactomannan-negative results on serum and clinical grounds. In the latterepisodes, the following microorganisms were recovered fromblood culture: CoNS ( n  = 9), six of which were considered ascontaminants, methicillin-susceptible  S. aureus  (MSSA), methicil-lin-resistant S.aureus (MRSA), E.coli ,and Enterobactercloacae .FourepisodeswithanegativebloodcultureresulttestedpositivebytheSeptiFast assay. The microorganisms detected in these episodeswere  E. cloacae/aerogenes ,  Pseudomonas aeruginosa , CoNS and Streptococcus spp . The latter two were considered contaminantsbased on clinical grounds.  P. aeruginosa  was later recovered from asecond set of blood cultures (not shown). A negative result in bothtests was obtained in 24 febrile episodes. The overall microorgan-ism-to-isolate agreement between the two assays was 64%, butincreased to 75% ( k = 0.56) when the clinically irrelevant micro-organisms recovered from blood culture or detected by theSeptiFastassaywereexcludedfromtheanalysis.Inthissetting,thesensitivity of the SeptiFast assay was 70% and of blood culture was73%.  3.4. Potential therapeutic relevance of the information provided bythe SeptiFast assay in febrile patients in the ICU  As shown in Table 4, empirical treatments administered tofebrile ICU patients with clinically relevant positive results by theSeptiFast assay alone or by the SeptiFast assay and blood culturewere deemed adequate in all cases. Nevertheless, antibioticadjustments consisting of: (1) suppression of either broad-spectrum antibiotics (meropenem, levofloxacin, or the combina-tion ceftazidime/amikacin) or antibiotics targeting beta-lactam-resistant Gram-positive bacteria (linezolid or daptomycin), (2)additionof thelattertypeof antimicrobials,and (3)substitutionof broad-spectrum antibiotics (levofloxacin or cefepime) by carba-penems (on the basis of the antibiotic resistance profile of thecirculating microorganisms, quinolone-resistant and extended-spectrum beta-lactamase (ESBL)-producing  Enterobacteriaceae )were considered potentially feasible in all but three episodes; intwo out of the latter episodes,  P. aeruginosa  was detected and the  Table 3 Microorganisms detected by the SeptiFast assay and recovered from blood culture in febrile episodes occurring in ICU patientsSpecies Number of samples with positive result by:Blood culture only SeptiFast assay only Both assaysCoagulase-negative staphylococci a 9 1 1 Staphylococcus aureus  2 0 2 Enterococcus faecalis  0 0 1 Streptococcus spp b 0 1 0 Escherichia coli  1 1 1 Serratia marcescens  0 0 1 Klebsiella pneumoniae/oxytoca  0 0 1 Enterobacter cloacae/aerogenes  1 1 1 Pseudomonas aeruginosa  0 1 1 Candida albicans  0 0 1  Aspergillus spp c 0 1 0ICU, intensive care unit. a Seven out of the 11 coagulase-negative staphylococci were considered contaminants. b Interpreted as a contaminant. c The detection of   Aspergillus spp  DNA was judged clinically irrelevant based on microbiological and clinical data.  Table 2 PotentialtherapeuticrelevanceoftheinformationprovidedbytheSeptiFastassayinfebrileneutropenicpatientswithclinicallysignificantmicroorganismsisolatedbybloodcultureMicroorganism detectedby SF / BC resultProbable srcinof infectionEmpirical treatment and its adequacy. a Feasibility of antibiotic adjustment Serratia marcescens  / identical Unknown Meropenem. Adequate. Adjustment not feasible Klebsiella pneumoniae / oxytoca  /  Klebsiella pneumoniae Unknown Meropenem. Adequate. Adjustment not feasible Stenotrophomonas maltophilia  / CoNS Unknown Meropenem. Inadequate. b Adjustment feasible:addition of trimethoprim–sulfamethoxazole Escherichia coli+Acinetobacter baumannii  /  Escherichia coli  Unknown Meropenem. Adequate. Adjustment not feasible Staphylococcus aureus  /  Enterococcus faecium  Unknown Meropenem. Vancomycin was added on the basisof Gram stain results. Adequate. Adjustment feasible: suppression of meropenem Streptococcus spp  /  Streptococcus mitis  Unknown Ceftazidime/amikacin. Vancomycin was added on the basisof Gram stain results. Adequate. Adjustment feasible: suppressionof broad-spectrum antibioticsCoNS / identical c Catheter d Ceftazidime/amikacin. Vancomycin was added on the basis of Gram stain results. Adequate. Adjustment feasible: suppressionof broad-spectrum antibioticsSF, SeptiFast assay; BC, blood culture; CoNS, coagulase-negative staphylococci. a An empirical therapeutic regimen was deemed adequate when providing coverage for the causal microorganism. b Eventhough Stenotrophomonasmaltophilia waseventuallyjudgedasclinicallyirrelevant,additionofantimicrobialsspecificallytargetingthismicroorganismwouldhavebeen considered adequate within that time frame. c Two out of the three CoNS detected by the SeptiFast assay and recovered from blood culture were interpreted as contaminants. d CoNS ( Staphylococcus epidermidis ) with an identical antimicrobial susceptibility profile were isolated from blood and from the semiquantitative catheter tip culture. D. Bravo et al./International Journal of Infectious Diseases 15 (2011) e326–e331  e329  combinationofmeropenemandlevofloxacinwasdeemednottobeadjustable(duetocirculatingstrainsresistanttoeitherquinolonesor carbapenems, or both).  E. cloacae/aerogenes  was detected by theSeptiFast assay in the remaining episode. In this case, themeropenem therapy was judged not to be adjustable, asderepressed AmpC beta-lactamase-producing  Enterobacter spp strains have been shown to circulate in this setting. 4. Discussion The primary objective of the current study was to compare theperformanceoftheSeptiFastassaywiththatofconventionalbloodculture in the etiological diagnosis of suspected bloodstreaminfections occurring in neutropenic (under antibiotic prophylaxis)and critically ill (antibiotic-naı¨ve) patients presenting with fever,with or without focal signs of infection. We were particularlyinterestedinassessingtheanalyticalperformanceofbothmethodsat the precise moment of the onset of fever, when decisionsregarding the antibiotic regimen to be implemented have thegreatestimpactontheclinicaloutcome.Thus,samplesdrawnonlyonce from any given patient were considered for analysis. Overall,we found that the blood culture was more sensitive than theSeptiFast assay. This was mostly due to the great number of CoNSdeemed to be clinically irrelevant that were recovered from bloodculture but not detected by the SeptiFast assay. This was notunexpected, as the SeptiFast assay has been designed to avoid thedetection of CoNS DNA present at low concentrations (contami-nating bacteria). 8 In fact, when clinically irrelevant microorgan-isms isolated from blood culture were excluded from the analysis,both assays were found to display comparable sensitivities.The fact that neutropenia minimizes the likelihood of detectingmicrobial DNA in blood may have been responsible for therelatively low diagnostic yield of the SeptiFast assay in ourhematology patients. 16 Previously published data on the sensitivi-ty of the SeptiFast assay in comparison with that of blood cultureare rather heterogeneous. In some studies, the SeptiFast assay hasbeen found to be clearly superior, 9–11,17–19 whereas in others bothmethodshaveeitherdisplayedsimilarsensitivitiesortheSeptiFastassay has compared unfavorably with respect to blood cul-ture. 15,16,20 Differences in the characteristics and clinical condi-tionsofpatients(ratherheterogeneousinourstudy),thespectrumof microorganisms detected, the number of blood culture setsdrawn, the volume of blood drawn for culture, and the presence orabsence of ongoing antibiotic therapy at the time of samplingamong the studies may account for these discrepancies.For clinically relevant microorganisms, the overall microorgan-ism-to-isolate agreement between the SeptiFast and the bloodculture results was relatively modest (69% for patients in theHematology Unit and 75% for those in the ICU) and in line withpreviouslyreporteddata. 16,18,20 TheSeptiFastassayfailedtodetectseveral clinically significant species, such as  E. coli  ( n  = 3),  E.cloacae , MSSA, and MRSA. The failure of the SeptiFast assay indetecting the latter two species may be related to the fact thatbacterial lysis in this system is more difficult for Gram-positivebacteria. 19 With respect to the former species, analogous data tothose reported herein have been published. 15–18,20 TheinterpretationofthedetectionofbacterialorfungalDNAintheabsenceofapositivebloodcultureisnotstraightforward,sincecirculating DNA may come from non-viable microorganisms orfrom microorganisms inhibited by antimicrobials. Furthermore,microbial DNA detected by the SeptiFast assay may even comefromcontaminantenvironmentalmicroorganisms.Ourexperienceon this matter is rather conflicting. On the one hand, the presenceof bacterial DNA in blood in two out of four febrile episodesoccurring in critically ill patients ( P. aeruginosa  and  E. cloacae/ aerogenes ) was deemed relevant on the basis of clinical findingsand microbiological data. On the other hand, the clinicalsignificance of the detection of DNA from  S. maltophilia ,  A.  Table 4 Potential therapeutic relevance of the information provided by the SeptiFast assay in febrile patients in the ICU with clinically significant positive results obtained either bythe SeptiFast assay alone or by the SeptiFast assay and blood cultureMicroorganism detectedby SF assay / BC resultsProbable srcinof infection a Empirical treatment and its adequacy. b Feasibility of antibiotic adjustment Enterococcus faecalis  / identical Urinary tract Ceftazidime/gentamicin/vancomycin. Adequate. Adjustment feasible:suppression of broad-spectrumantibiotic therapy Staphylococcus aureus  / MSSA Unknown Meropenem/levofloxacin. Adequate. Adjustment feasible: additionof vancomycin/linezolid/daptomycin and suppression of broad-spectrum antibiotic therapy Staphylococcus aureus  / MRSA Catheter Meropenem. Linezolid added on the basis of Gram stain results. Adequate. Adjustment feasible:suppression of meropenem Enterobacter cloacae  / negative Respiratory c Meropenem/linezolid. Adequate. Adjustment feasible: suppression of linezolid Pseudomonas aeruginosa  / negative Respiratory c Meropenem/levofloxacin. Adequate. Adjustment not feasible Serratia marcescens  / identical Respiratory c Levofloxacin/linezolid. Adequate. Adjustment feasible: addition of a carbapenemagent and suppression of linezolid Escherichia coli  / CoNS d Respiratory c Cefepime. Linezolid added on the basis of Gram stain results. Adequate. Adjustmentsuitable: carbapenem therapy instead of cefepime and suppression of linezolid Escherichia coli  / identical Unknown Levofloxacin/daptomycin. Adequate. Adjustment feasible: carbapenem therapyinstead of levofloxacin and suppression of linezolid Enterobacter cloacae/aerogenes  /  Enterobacter cloacae Respiratory c Meropenem. Adequate. No adjustment feasible Pseudomonas aeruginosa  / identical Respiratory c Meropenem/levofloxacin. Adequate. No adjustment feasible Klebsiella pneumoniae/oxytoca  /  Klebsiella pneumoniae Urinary Meropenem/levofloxacin. Adequate. Adjustmentfeasible: suppression of levofloxacin Candida albicans  / identical Catheter Levofloxacin. Caspofungin added on the basis of Gram stain. Adequate. Adjustment feasible:suppression of levofloxacinCoNS / CoNS Catheter Meropenem. Linezolid added on the basis of Gram stain. Adequate.Adjustment feasible: suppression of meropenemICU, intensive care unit; SF, SeptiFast assay; BC, blood culture; CoNS, coagulase-negative staphylococci; MSSA, methicillin-susceptible  Staphylococcus aureus ; MRSA,methicillin-resistant  Staphylococcus aureus . a The same bacterial species was recovered from the indicated source. b An empirical therapeutic regimen was deemed adequate when providing cover for the causal microorganism. c Ventilator-associated pneumonia. d CoNS considered as a contaminant. D. Bravo et al./International Journal of Infectious Diseases 15 (2011) e326–e331 e330
Search
Similar documents
View more...
Related Search
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
SAVE OUR EARTH

We need your sign to support Project to invent "SMART AND CONTROLLABLE REFLECTIVE BALLOONS" to cover the Sun and Save Our Earth.

More details...

Sign Now!

We are very appreciated for your Prompt Action!

x