Blood cultures remain the gold standard for diagnosing bacterial and Candida
bloodstream infections (BSIs), but they are limited by slow turnaround and suboptimal sensitivity.1,2
Data from some but not all retrospective studies suggest that rapid initiation of active antimicrobial therapy correlates with reduced mortality among patients with bacterial or Candida
Although such findings have not been validated in prospective studies,5
the limitations of blood cultures and quality standards for sepsis and septic shock provide a rationale for aggressive empiric broad-spectrum antimicrobial therapy.5,9
These practices may promote unnecessary antimicrobial usage, drug toxicity, and emergence of resistance.5,10,11
Culture-independent diagnostic tests (CIDTs) that detect pathogens and antimicrobial resistance genes within clinical samples are transforming clinical practice.2,12
Syndromic CIDT panels for respiratory specimens, stool, and cerebrospinal fluid facilitate rapid identification or exclusion of infections due to particular organisms and antimicrobial initiation or de-escalation strategies.12
Likewise, commercial molecular tests of positive blood cultures for pathogen or resistance determinant identification are valuable patient care and stewardship tools.12-14
CIDTs of whole blood samples from patients with suspected BSIs have not gained widespread acceptance, in part because costs are high and clinicians are unsure how to incorporate testing into rational management paradigms.12,15
This paper reviews the performance of blood cultures and direct-from-whole-blood CIDTs for bacterial and Candida
BSIs and provides a conceptual framework for using CIDTs in the clinic.
PERFORMANCE OF BLOOD CULTURES AND CIDTS
Blood culture sensitivity for detecting bacteremia is ~73%, 90%, and 98% if 1, 2 and 3 sets of aerobic and anaerobic bottles, respectively, are collected in the absence of antimicrobial treatment.16
Such performance is contingent upon collecting ≥20 mL of blood in each culture set. Smaller volumes, as often collected in hospitalized patients, increase false negativity.2
Blood cultures may require several days for bacteria to achieve detectable concentrations and additional time for species identification and resistance testing.2
Blood cultures are ~50% sensitive for diagnosing invasive candidiasis.1
Moreover, blood cultures become positive late in the course of invasive candidiasis, and incubation times prior to positivity are typically longer than those for detecting bacteria.1
Antimicrobial treatment reduces sensitivity for bacteria and Candida
which is notable because 28% to 63% of blood cultures are collected from patients who are receiving antimicrobial agents.2
Several direct-from-whole-blood CIDTs couple nucleic acid amplification with novel technologies for target detection, including Iridica, SeptiFast, SepsiTest, Magicplex, and T2Direct (T2Bacteria, T2Candida) assays.2
T2Direct is the only FDA-cleared system.17,21,22
Assay characteristics and performance are summarized in the Table2,17,21,26,36,37
. Interpretation of performance is complicated by study heterogeneity and limitations of blood cultures as gold standard. CIDTs require ≤5 mL of blood and provide results within 4 to 10 hours. In general, DNA amplification-based technologies are more sensitive than blood cultures in patients receiving antimicrobials.17,23-25
We will focus on SeptiFast and T2Direct as examples of broad- and narrow-spectrum platforms, respectively.17,21,22,26
Concepts presented here can be applied to other tests.
CIDTS AS BAYESIAN DIAGNOSTICS
CIDTs assign a probability of BSI due to a given pathogen.27-29
Positive and negative predictive values (PPVs, NPVs) are determined by sensitivity and specificity of the test and the patient’s likelihood of BSI. Anticipated PPVs and NPVs in different types of patients can be calculated for pathogens targeted by SeptiFast (sensitivity/ specificity: 75%/90%), T2Bacteria (85%/95%), and T2Candida (90%/98%; Table 2
). Tests are unlikely to be useful if performed every time a blood culture is collected,30
because false positivity exceeds true positivity, and excellent NPVs provide only marginal value over already low pretest likelihoods.
In patients presenting to the hospital with fever or those with sepsis in the absence of septic shock, anticipated SeptiFast and T2Bacteria PPVs for targeted bacteria are each ~50% to 75%. SeptiFast offers an advantage of detecting ~85% to 90% of bacteria and fungi that cause BSIs, whereas T2Bacteria detects 5 bacteria that account for ~50% of BSIs.26,31-33
A potential disadvantage of broad-spectrum panels is that some targets are uncommon causes of BSI and therefore more likely to generate false-positive results. Although T2Bacteria NPVs are excellent for targeted bacteria, they are inferior to SeptiFast NPVs in excluding BSIs due to any bacteria. Candida
are typically rare causes of BSI in these populations,15,27-29
and T2Candida PPVs and NPVs are unlikely to be useful in most cases. Anticipated T2Candida PPVs may approach 67%, and NPVs are 99.7% among septic patients with risk factors for candidemia in whom a causative bacterium is not identified. T2Candida detects species that account for >95% of candidemia at most centers.17,21
The likelihood of bacterial or Candida
BSI increases in septic shock, and anticipated PPVs of each test are ≥70%. However, SeptiFast and T2Bacteria NPVs are just 78% and 62%, respectively, for excluding BSIs in the setting of septic shock; T2Candida NPV remains excellent (99.5%).
INCORPORATING CIDTS INTO CLINICAL AND STEWARDSHIP PRACTICES
Surviving Sepsis guidelines endorse empiric broad-spectrum antibiotic therapy within 1 hour of triage for suspected sepsis and septic shock and rapid de-escalation based on susceptibility of causative organisms or if infection is excluded.9
Other experts propose more nuanced, case-by-case approaches, in which empiric antibiotics are administered immediately for suspected septic shock, but clinical observation and diagnostic testing may be undertaken prior to treatment decisions in at least some patients with suspected sepsis in the absence of shock.5,10,11
In patients with possible sepsis in whom treatment decisions have been deferred pending workup, positive SeptiFast or T2Bacteria results may shorten the time to antibiotic treatment compared with waiting for positive blood cultures. If blood or other cultures are negative and an alternative, noninfectious diagnosis is not established, there is a good chance based on predictive values that a positive CIDT has identified a BSI that would have been missed otherwise. In such cases, continuing antibiotic treatment against the CIDT-identified pathogen is reasonable. Detection of resistance genes by SeptiFast may guide early selection of active antibiotics. T2Bacteria does not include resistance genes, but the panel is directed against ESKAPE pathogens (Enterococcus faecium, Staphylococcus aureus, Klebsiella pneumoniae, Pseudomonas aeruginosa,
and Enterobacter species
) that are often resistant to first-line antibiotics.22,34
In high-risk patients, such as those colonized by a resistant pathogen included in the panel, a positive T2Bacteria result may justify use of an alternative agent.
If SeptiFast or T2Bacteria results are negative and a patient with suspected sepsis is stable, clinicians may decide to withhold antimicrobials pending blood and other culture results. This strategy is reasonable because empiric treatment has not proved superior to culture-directed treatment of BSIs or suspected sepsis, and ~50% of suspected sepsis is ultimately ascribed to noninfectious etiologies.5,10,11
If antimicrobials have been administered, combined negative CIDT and culture results may offer an argument for de-escalation. Approaches to using CIDTs similar to those described for possible sepsis also may be useful for managing febrile patients in the emergency department.
In patients with presumed septic shock, clinicians will not wait for CIDT results before initiating broad-spectrum antibiotics (often in combination). In these patients, SeptiFast and T2Bacteria may have value if antibiotics are administered before blood cultures or whole blood samples are collected. Positive CIDT results may help in streamlining empiric antibiotic regimens. T2Candida is likely to be useful in septic shock (and perhaps in sepsis with risk factors for candidemia),35
since empiric antifungals are not recommended routinely.9
T2Candida PPVs and NPVs would justify initiating and withholding (or discontinuing) antifungal therapy, respectively. Sensitivity and specificity of SeptiFast are lower than T2Candida for Candida species,17,21,26
but the former test still may have value in patients with septic shock who are at risk of candidemia.
We propose a conceptual framework for thinking about how to incorporate CIDTs for BSIs into rational patient management and antimicrobial stewardship strategies. Similar exercises can be undertaken for other populations at risk of BSIs, such as patients with neutropenic fever or transplant recipients with sepsis. CIDT-based management paradigms will require validation in clinical trials. An important question is whether shorter turnaround times and identification of more potential pathogens and resistance determinants using direct-from-whole-blood CIDTs can lead to improved patient outcomes compared with those obtained with molecular testing of positive blood cultures.
Clancy is chief of infectious diseases at the VA Pittsburgh Healthcare System and director of the XDR Pathogen Laboratory at the University of Pittsburgh. His research lab is funded by the Department of Veterans Affairs and National Institutes of Health.
Nguyen is director of the Antimicrobial Stewardship and Transplant Infectious Diseases programs at the University of Pittsburgh Medical Center. Her translational and basic science research lab is funded by the National Institutes of Health and other sources.
1. Clancy CJ, Nguyen MH. Finding the "missing 50%" of invasive candidiasis: how nonculture diagnostics will improve understanding of disease spectrum and transform patient care. Clin Infect Dis.
2013 May;56(9):1284-92. doi: 10.1093/cid/cit006.
2. Sinha M, Jupe J, Mack H, Coleman TP, Lawrence SM, Fraley SI. Emerging Technologies for Molecular Diagnosis of Sepsis. Clin Microbiol Rev.
2018 Feb 28;31(2). pii: e00089-17. doi: 10.1128/CMR.00089-17.
3. Morrell M, Fraser VJ, Kollef MH. Delaying the empiric treatment of candida bloodstream infection until positive blood culture results are obtained: a potential risk factor for hospital mortality. Antimicrob Agents Chemother.
2005 Sep;49(9):3640-5. doi: 10.1128/AAC.49.9.3640-3645.2005.
4. Kollef M, Micek S, Hampton N, Doherty JA, Kumar A. Septic shock attributed to Candida infection: importance of empiric therapy and source control. Clin Infect Dis.
2012 Jun;54(12):1739-46. doi: 10.1093/cid/cis305.
5. Singer M. Antibiotics for Sepsis: Does Each Hour Really Count, or Is It Incestuous Amplification? Am J Respir Crit Care Med.
2017 Oct 1;196(7):800-802. doi: 10.1164/rccm.201703-0621ED.
6. Kumar A, Roberts D, Wood KE, et al. Duration of hypotension before initiation of effective antimicrobial therapy is the critical determinant of survival in human septic shock. Crit Care Med.
2006 Jun;34(6):1589-96. doi: 10.1097/01.CCM.0000217961.75225.E9.
7. Taur Y, Cohen N, Dubnow S, Paskovaty A, Seo SK. Effect of antifungal therapy timing on mortality in cancer patients with candidemia. Antimicrob Agents Chemother.
2010 Jan;54(1):184-90. doi: 10.1128/AAC.00945-09.
8. McGregor JC, Rich SE, Harris AD, et al. A systematic review of the methods used to assess the association between appropriate antibiotic therapy and mortality in bacteremic patients. Clin Infect Dis.
2007 Aug 1;45(3):329-37. doi: 10.1086/519283.
9. Rhodes A, Evans LE, Alhazzani W, et al. Surviving Sepsis Campaign: International Guidelines for Management of Sepsis and Septic Shock: 2016. Crit Care Med.
2017 Mar;45(3):486-552. doi: 10.1097/CCM.0000000000002255.
10. Klompas M, Calandra T, Singer M. Antibiotics for Sepsis-Finding the Equilibrium. JAMA.
2018 Oct 9;320(14):1433-1434. doi: 10.1001/jama.2018.12179.
11. IDSA Sepsis Task Force. Infectious Diseases Society of America (IDSA) POSITION STATEMENT: Why IDSA Did Not Endorse the Surviving Sepsis Campaign Guidelines. Clin Infect Dis.
2018 May 2;66(10):1631-1635. doi: 10.1093/cid/cix997.
12. Ramanan P, Bryson AL, Binnicker MJ, Pritt BS, Patel R. Syndromic Panel-Based Testing in Clinical Microbiology. Clin Microbiol Rev.
2017 Nov 15;31(1). pii: e00024-17. doi: 10.1128/CMR.00024-17.
13. Timbrook TT, Morton JB, McConeghy KW, Caffrey AR, Mylonakis E, LaPlante KL. The Effect of Molecular Rapid Diagnostic Testing on Clinical Outcomes in Bloodstream Infections: A Systematic Review and Meta-analysis. Clin Infect Dis.
2017 Jan 1;64(1):15-23. Epub 2016 Sep 26.
14. Banerjee R, Teng CB, Cunningham SA, et al. Randomized Trial of Rapid Multiplex Polymerase Chain Reaction-Based Blood Culture Identification and Susceptibility Testing. Clin Infect Dis.
2 2015 Oct 1;61(7):1071-80. doi: 10.1093/cid/civ447.
15. Clancy CJ, Nguyen MH. Diagnosing Invasive Candidiasis. J Clin Microbiol.
2018 Apr 25;56(5). pii: e01909-17. doi: 10.1128/JCM.01909-17.
16. Lee A, Mirrett S, Reller LB, Weinstein MP. Detection of bloodstream infections in adults: how many blood cultures are needed? J Clin Microbiol.
2007 Nov;45(11):3546-8. doi: 10.1128/JCM.01555-07.
17. Clancy CJ, Pappas PG, Vazquez J, et al. Detecting Infections Rapidly and Easily for Candidemia Trial, Part 2 (DIRECT2): A Prospective, Multicenter Study of the T2Candida Panel. Clin Infect Dis.
2018 May 17;66(11):1678-1686. doi: 10.1093/cid/cix1095.
18. Rhodes J, Hyder JA, Peruski LF, et al. Antibiotic use in Thailand: quantifying impact on blood culture yield and estimates of pneumococcal bacteremia incidence. Am J Trop Med Hyg.
2010 Aug;83(2):301-6. doi: 10.4269/ajtmh.2010.09-0584.
19. Resti M, Micheli A, Moriondo M, et al. Comparison of the effect of antibiotic treatment on the possibility of diagnosing invasive pneumococcal disease by culture or molecular methods: a prospective, observational study of children and adolescents with proven pneumococcal infection. Clin Ther.
2009 Jun;31(6):1266-73. doi: 10.1016/j.clinthera.2009.06.010.
20. Driscoll AJ, Deloria Knoll M, Hammitt LL, et al. The Effect of Antibiotic Exposure and Specimen Volume on the Detection of Bacterial Pathogens in Children With Pneumonia. Clin Infect Dis.
2017 Jun 15;64(suppl_3):S368-S377. doi: 10.1093/cid/cix101.
21. Mylonakis E, Clancy CJ, Ostrosky-Zeichner L, et al. T2 magnetic resonance assay for the rapid diagnosis of candidemia in whole blood: a clinical trial. Clin Infect Dis.
2015 Mar 15;60(6):892-9. doi: 10.1093/cid/ciu959.
22. De Angelis G, Posteraro B, De Carolis E, et al. T2Bacteria magnetic resonance assay for the rapid detection of ESKAPEc pathogens directly in whole blood. J Antimicrob Chemother.
2018 Mar 1;73(suppl_4):iv20-iv26. doi: 10.1093/jac/dky049.
23. Bloos F, Hinder F, Becker K, et al. A multicenter trial to compare blood culture with polymerase chain reaction in severe human sepsis. Intensive Care Med.
2010 Feb;36(2):241-7. doi: 10.1007/s00134-009-1705-z.
24. Bloos F, Sachse S, Kortgen A, et al. Evaluation of a polymerase chain reaction assay for pathogen detection in septic patients under routine condition: an observational study. PLoS One.
2012;7(9):e46003. doi: 10.1371/journal.pone.0046003.
25. Makristathis A, Harrison N, Ratzinger F, et al. Substantial diagnostic impact of blood culture independent molecular methods in bloodstream infections: Superior performance of PCR/ESI-MS. Sci Rep.
2018 Oct 30;8(1):16024. doi: 10.1038/s41598-018-34298-7.
26. Chang SS, Hsieh WH, Liu TS, et al. Multiplex PCR system for rapid detection of pathogens in patients with presumed sepsis - a systemic review and meta-analysis. PLoS One.
2013 May 29;8(5):e62323. doi: 10.1371/journal.pone.0062323.
27. Clancy CJ, Shields RK, Nguyen MH. Invasive Candidiasis in Various Patient Populations: Incorporating Non-Culture Diagnostic Tests into Rational Management Strategies. J Fungi (Basel).
2016 Feb 6;2(1). pii: E10. doi: 10.3390/jof2010010.
28. Clancy CJ, Nguyen MH. Diagnostic Methods for Detection of Blood-Borne Candidiasis. Methods Mol Biol.
2016;1356:215-38. doi: 10.1007/978-1-4939-3052-4_16.
29. Clancy CJ, Nguyen MH. Non-Culture Diagnostics for Invasive Candidiasis: Promise and Unintended Consequences. J Fungi (Basel).
2018 Feb 19;4(1). pii: E27. doi: 10.3390/jof4010027.
30. Coburn B, Morris AM, Tomlinson G, Detsky AS. Does this adult patient with suspected bacteremia require blood cultures? JAMA.
2012 Aug 1;308(5):502-11. doi: 10.1001/jama.2012.8262.
31. Magill SS, Edwards JR, Bamberg W, et al. Multistate point-prevalence survey of health care-associated infections. N Engl J Med.
2014 Mar 27;370(13):1198-208. doi: 10.1056/NEJMoa1306801.
32. Magill SS, O'Leary E, Janelle SJ, et al. Changes in Prevalence of Health Care-Associated Infections in U.S. Hospitals. N Engl J Med.
2018 Nov 1;379(18):1732-1744. doi: 10.1056/NEJMoa1801550.
33. Laupland KB, Church DL. Population-based epidemiology and microbiology of community-onset bloodstream infections. Clin Microbiol Rev.
2014 Oct;27(4):647-64. doi: 10.1128/CMR.00002-14.
34. Boucher HW, Talbot GH, Bradley JS, et al. Bad bugs, no drugs: no ESKAPE! An update from the Infectious Diseases Society of America. Clin Infect Dis.
2009 Jan 1;48(1):1-12. doi: 10.1086/595011.
35. Clancy CJ, Nguyen MH. Diagnosing candidemia with the T2Candida panel: an instructive case of septic shock in which blood cultures were negative. Diagn Microbiol Infect Dis.
2019 Jan;93(1):54-57. doi: 10.1016/j.diagmicrobio.2018.03.017.
36. Zboromyrska Y, Cilloniz C, Cobos-Trigueros N, et al. Evaluation of the Magicplex Sepsis Real-Time Test for the Rapid Diagnosis of Bloodstream Infections in Adults. Front Cell Infect Microbiol.
2019 Mar 12;9:56. doi: 10.3389/fcimb.2019.00056.
37. Ozenci V, Patel R, Ullberg M, Stralin K. Demise of Polymerase Chain Reaction/Electrospray Ionization-Mass Spectrometry as an Infectious Diseases Diagnostic Tool. Clin Infect Dis.
2018 Jan 18;66(3):452-455. doi: 10.1093/cid/cix743.
38. Singer M, Deutschman CS, Seymour CW, et al. The Third International Consensus Definitions for Sepsis and Septic Shock (Sepsis-3). JAMA.
2016 Feb 23;315(8):801-10. doi: 10.1001/jama.2016.0287.