Clinical Microbiology and Infection Prevention - Journal of Clinical ...

JOURNAL OF CLINICAL MICROBIOLOGY, Sept. 2011, p. S57–S60 0095-1137/11/$12.00 doi:10.1128/JCM.00690-11 Copyright © 2011, American Society for Microbiology. All Rights Reserved.

VOL. 49, NO. 9 SUPPL.

Clinical Microbiology and Infection Prevention Daniel J. Diekema1,2* and Michael A. Saubolle3,4 Division of Infectious Diseases, University of Iowa College of Medicine,1 and Infection Prevention Program, University of Iowa Hospitals and Clinics,2 Iowa City, Iowa; Division of Infectious Diseases, Laboratory Sciences of Arizona/Sonora Quest Laboratories, Phoenix, Arizona3; and Department of Medicine, University of Arizona College of Medicine, Tucson, Arizona4

* Corresponding author. Mailing address: Division of Infectious Diseases, University of Iowa College of Medicine, 200 Hawkins Drive, Iowa City, IA 52246. Phone: (319) 356-8615. Fax: (319) 356-4916. E-mail: [email protected]. S57

Downloaded from http://jcm.asm.org/ on December 9, 2017 by guest

relatedness, which requires maintenance of an organism bank (Table 1). The laboratory should also act in a consultative capacity with the IPP to help determine whether an outbreak is “real” or a potential pseudo-outbreak due to contamination of specimens outside or within the laboratory. In addition, the laboratory can help generate hypotheses as to the potential source of an outbreak, its reservoir, and its mode of spread, through molecular typing of the suspected organisms and through testing the environment and/or personnel as necessary. Outbreaks are often detected retrospectively, either after they have resolved or when they are already beginning to wane. Therefore, a major challenge to the CML is in detecting outbreaks early enough to allow effective intervention and impact morbidity and mortality. Effective and regular communication between CML personnel and infection preventionists is essential to this effort, as many outbreaks are first noted at the CML benches when technologists notice unusual clusters of organisms or antimicrobial resistance patterns. The future of early detection will likely be found in real-time data-mining software that can raise flags based upon subtle changes in rates of test ordering or positive test results. Antimicrobial stewardship. Every hospital must now have an antimicrobial stewardship program, guidelines for which have been published by the Infectious Diseases Society of America and the Society for Healthcare Epidemiology of America (4). Antimicrobial stewardship efforts are directly dependent on reports from the CML, so good communication between the laboratory, pharmacy, IPP, and a stewardship team is essential. For guiding empirical antimicrobial therapy, unit-specific and tailored antibiograms should be updated on a regular basis and provided to clinicians at the bedside. Such antibiogram data can also be used for evaluation of trends in important antimicrobial resistance rates and for education of clinicians regarding optimal antimicrobial use. For guiding directed antimicrobial therapy, patient-specific culture and susceptibility data are needed. This allows for a prospective audit of antimicrobial use with feedback to the prescriber. A major challenge to effective stewardship is an inability to obtain antimicrobial susceptibility data from the CML in a timely and efficient manner. Reducing the analytic turnaround time is only the starting point, however. The data then have to be incorporated quickly into antimicrobial management, which often cannot be done if laborious chart reviews are required for each patient on antimicrobial therapy. Computer decision support systems can solve this problem by automatically identifying potential opportunities to optimize therapy, using realtime analysis of data from several sources (e.g., pharmacy, electronic medical record, and laboratory) (11).

Health care-associated infections (HAIs) represent one of the most common complications of care, affecting 5 to 10% of patients admitted to acute-care hospitals worldwide. These HAIs are associated with enormous morbidity and mortality, resulting in more than 90,000 deaths each year in the United States (8). Every health care facility must therefore have a program charged with monitoring and preventing infections in the health care environment. Preventing infections requires the ability to detect them when they occur, which is why the clinical microbiology laboratory (CML) plays a key role in HAI prevention. Some of the major roles played by the CML in infection prevention include contributions to (i) surveillance, (ii) outbreak detection and management, (iii) antimicrobial stewardship, (iv) deliberations of the infection control committee, and (v) education. In each of these areas, the CML faces new challenges as it seeks to contribute fully to the urgent task of preventing HAIs. Surveillance. Review of CML data remains the most common method for case finding in HAI surveillance; therefore, the most important role of the CML is to promptly and accurately detect nosocomial pathogens and their antimicrobial resistance patterns (1, 5). The CML must also work with the infection prevention program (IPP) and the information technology (IT) department to determine how microbiology data are delivered and linked to other surveillance data to streamline this process. Major surveillance challenges facing the CML include the continued emergence of novel infectious agents (e.g., the 2009 H1N1 influenza A virus), novel antimicrobial-resistant pathogens (e.g., vancomycin-intermediate/resistant Staphylococcus aureus [VISA/VRSA], carbapenem-resistant Enterobacteriaceae [CRE]), and new governmental and public health mandates (e.g., mandated active surveillance culturing in some states and in the VA health care system and public reporting of HAI rates). Finally, the CML is under increasing pressure to provide rapid test results. As hospital lengths of stay decrease (12), the window of clinical relevance becomes smaller and smaller. Outbreak detection and management. The CML has important roles to play in any potential outbreak situation, including early recognition of possible clusters and outbreaks, rapid notification of and collaboration with the IPP, additional case finding, and provision of molecular typing for determination of

S58

CAMP CLIN MICRO

J. CLIN. MICROBIOL.

TABLE 1. Steps in nosocomial outbreak investigation and the role of the laboratory at each stepa Investigative step

Role of the clinical microbiology laboratory

a

Adapted from reference 5.

Infection control committee participation. It is paramount that the clinical microbiologist participates on the infection prevention/control committee and acts as a consultant to infection preventionists. He or she is the best person to provide expertise in the interpretation of culture results, advice about the utility of microbiological approaches to an infection control problem, and input regarding the CML resources needed to accomplish the goals of the committee. He or she should describe how changes in the methods used for detection, identification, and susceptibility testing of nosocomial pathogens will impact the IPP. The benefits of close collaboration and interaction between the CML and the IPP are difficult to measure but are real. One large survey of CML directors found that those hospitals with CML directors on the infection prevention/control committee were more likely to have CML involvement in formulary decisions, to produce an annual antibiogram, and to provide molecular typing support (6). Several ongoing trends challenge these valuable personal interactions between CML and IPP personnel. Consolidation of CML services, off-site moves of CML laboratories, and total reliance on the electronic medical record to the exclusion of first-hand observation (e.g., review of plates or Gram stains) too often keep clinicians and infection preventionists out of the CML and keep CML personnel confined to the laboratory. Education. CMLs also play an essential role in the education of future hospital epidemiologists and infection preventionists. Most hospital epidemiologists are trained in infectious diseases (though some are also laboratorians and CML directors). The ACGME requirements for training in infectious diseases require structured clinical microbiology training, and infection preventionists also require microbiology training. The trends referred to above (CML consolidation and off-site moves) complicate this educational role, as does the difficult financial

situation of many hospitals. Who pays for the time and effort required to deliver this educational content? DISCUSSION As a starting point, the discussants agreed that CMLs should no longer be viewed as merely playing supportive roles for infection prevention but rather should be seen as an essential element of the IPP at every health care facility. Viewing the CML as part and parcel of the IPP may help health care administrators understand the importance of adequate funding for the CML, particularly as CML activities increase to meet infection prevention priorities. Effective prevention saves not only lives but money, and these savings are rarely credited to the CML. Additional discussion focused upon specific aspects of the CML activities outlined above, including (i) increasing calls for active surveillance cultures (ASCs) to detect multidrug-resistant organism (MDRO) carriers, (ii) a more standardized approach to immediate or urgent notification of IPP personnel, (iii) a uniform definition for multiply drug-resistant Gramnegative rods (MDR-GNRs) of infection control significance, and (iv) limitations inherent in current approaches to antibiogram generation. Active surveillance. The role of broadly applied active surveillance cultures (ASCs) in detection of MDRO carriers remains unclear. Nonetheless, many hospitals have adopted this approach for control of methicillin-resistant Staphylococcus aureus (MRSA), often in response to legislative mandates. The issue of MRSA active surveillance is addressed in another paper in this series. Therefore, our discussion was focused on surveillance for MDR-GNRs, which continue to emerge as a major threat in the health care environment (14). Can our experience thus far with MRSA inform our future


Recognize problem.............................................................Surveillance and early warning system, ideally part of the laboratory information system; notify infection control personnel of infection clusters, unusual resistance patterns, possible patient-to-patient transmission Establish case definition ....................................................Assist and advise regarding inclusion of laboratory diagnosis in case definition Confirm cases......................................................................Perform laboratory confirmation of diagnosis Complete case finding........................................................Characterize isolates with accuracy, store all sterile-site isolates and epidemiologically important isolates, search laboratory database for new cases Establish background rate of disease and compare to attack rate during suspected outbreak ............................................Provide data for use in ongoing surveillance, which provide baseline rates for selected units and infection sites; search laboratory database for all prior cases of the entity if baseline rate is not prospectively monitored Characterize outbreak (descriptive epidemiology)..................................................................Perform typing of involved strains, compare to previously isolated endemic strains to determine if the outbreak involves a single strain (this can be done only if selected pathogens are routinely stored 关see above兴) Generate hypotheses about causation: reservoir, mode of spread, vector.................................Perform supplementary studies or cultures as needed, but only if justified by epidemiologic link to transmission: personnel, patients. environment Case-control study or cohort study Institute control measures.................................................Adjust laboratory procedures as necessary Ongoing surveillance to document efficacy of control measures ..........................................Maintain surveillance and early-warning function of the laboratory

VOL. 49, NO. 9 SUPPL., 2011

CONVENTIONAL VERSUS MOLECULAR METHODS

TABLE 2. Can the MRSA experience inform screening for MDR Gram-negative rods? Parameter

Transmissibility of key resistance determinant(s) Complexity of resistance mechanism(s) Lab methods for detection of carrier state Established/effective decolonization protocols Literature base assessing screening approaches

MRSA

MDR-GNR

Low

High

Low

High

Yes

In development

Yes

No

Abundant

Scant

Bacillus anthracis, Yersinia pestis, and orthopox viruses). However, the discussants pointed out that the list of organisms for immediate IP notification differs from one institution to another and that it would be appropriate to establish more uniformity. We recognize that some institutions may wish to include organisms or results that others do not (for example, an MDRO may have become endemic in one hospital, to the point that the IP program no longer requires immediate notification, while another hospital has not yet had introduction of the MDRO and needs such notification). Nonetheless, compiling a standard list of “critical” infection prevention results should be a fairly straightforward matter and would start with a simple survey of laboratories to determine what current practice is at a wide range of facilities. Standardizing the MDR-GNR definitions. In addition to optimizing communication with the IPP, the discussants agreed that a widely accepted definition of what constitutes “MDR” for Gram-negative rods would be helpful. Given the diversity of inherent and acquired resistance among the Gram-negative bacteria, institutions vary substantially in their approach to this. Recent efforts have provided a good start (13), but the utility of these definitions must be tested in diverse health care settings. A further complicating feature is the ability of laboratories to detect MDR-GNRs or to differentiate those that require additional infection prevention interventions (e.g., contact precautions or transmission investigations) from those that do not. For example, the CLSI recently changed breakpoints for Enterobacteriaceae and cephalosporins and carbapenems, obviating the need for ESBL confirmatory testing or modified Hodge testing for clinical use. Unfortunately, many IPPs had become accustomed to using the ESBL/Hodge test results to guide institution of contact (barrier) precautions. In one institution, this change resulted in a 35% increase in the number of MDR-GNR isolates identified and a concomitant increase in the hospital’s use of contact precautions (3). The effectiveness of contact precautions in reducing MDRGNR transmission in health care settings is not known, so of course neither is the relative effectiveness of this intervention against different types of MDR-GNR (e.g., those with transmissible or plasmid-borne resistance determinants versus those with chromosomally mediated determinants). Future studies to address these questions will require a standard approach to defining those MDR-GNRs that ought to trigger specific infection prevention measures. This assumes that the CML has the tools to differentiate among these MDR-GNRs, which remains a challenge. Both ESBL and Hodge confirmatory testing are poor substitutes for molecular methods for detection of ESBLs and carbapenemases, but in their absence most laboratories must rely solely on an organism’s resistance phenotype to determine the likelihood that it carries a resistance determinant of epidemiologic significance. As more carbapenemases emerge (9), there will be an increasing need for rapid, cost-effective methods for their identification. Limitations of the current approach to antibiogram preparation. There was also a short discussion regarding the extent to which the current CLSI guideline for antibiogram preparation (2) may underestimate the problem of emerging resistance by including only the first isolate of a single species from an individual patient. Given that antimicrobial resistance often emerges during hospitalization, later isolates (second, third,


approach to MDR-GNR active surveillance? As outlined in Table 2, there are several important differences between MRSA and the MDR-GNRs that suggest the answer may be “no.” Whereas a single gene (mecA) provides a straightforward gold standard for MRSA detection, resistance mechanisms for MDR-GNR are multiple and highly complex. In contrast to MRSA, many of the most important MDR-GNR resistances are carried on mobile genetic elements (e.g., ESBLs and carbapenemases), laboratory methods for routine detection of the carrier state are still being evaluated, decolonization protocols are not currently available or feasible, and the literature base to support screening is scant. Harris et al. recently outlined the additional data required to inform decisions about MDR-GNR screening, including (i) the performance of available screening methods for organisms of interest (e.g., sensitivity, specificity, and frequency of screening), (ii) the proportion of MDR-GNR infections that are due to in-hospital transmission, (iii) the “undetected ratio” of colonized patients (the proportion of colonized patients not detected by clinical cultures), and finally (iv) the effectiveness of contact precautions in preventing MDR-GNR transmission (7). Although some studies have begun to address these questions (15), most discussants agreed with the two experienced health care epidemiologists who recently stated that “more data are needed… before we launch headlong into implementing large-scale active surveillance programs that threaten to divert resources from other important infection prevention initiatives” (10). In the meantime, participants felt that the use of MDRGNR active surveillance cultures should be limited to new introductions of problematic MDR-GNRs (e.g., carbapenemresistant Enterobacteriaceae) or outbreak settings. At all other times, the emphasis should remain fixed on so-called “horizontal” approaches to infection prevention, those tried-and-true practices that are applied to all at-risk patient populations and are designed to prevent infections due to all pathogens (e.g., hand hygiene and bundled practices for prevention of deviceassociated infections). Standardized approaches to urgent notification of IPP personnel. The detection of certain epidemiologically important pathogens requires immediate notification of the IPP so that transmission prevention measures can be initiated immediately (including, if applicable, notification of public health authorities). Examples of organisms for which urgent notification should occur include Mycobacterium tuberculosis, Neisseria meningitidis, certain MDROs, and agents of bioterrorism (e.g.,

S59

S60

CAMP CLIN MICRO

or tenth!) may provide a very different susceptibility profile. Again, a multilaboratory collaborative study to produce antibiograms using alternative approaches may help determine whether this problem is clinically important.

J. CLIN. MICROBIOL.

4.

5.

SUMMARY The CML is an essential element of the IPP in every health care facility, playing critical roles in surveillance, outbreak detection and management, antimicrobial stewardship, risk assessment and planning (through participation on the infection control committee), and education. Close collaboration between CML and IPP personnel is required to ensure optimal HAI prevention, which saves money and lives.

6.

7.

8.

REFERENCES 1. Barenfanger, J., J. Bente, and G. Havener. 2009. Optimal performance for clinical microbiologists and their interaction with infection control staff. Clin. Microbiol. Newsl. 31:9–15. 2. Clinical and Laboratory Standards Institute. 2009. Analysis and presentation of cumulative antimicrobial susceptibility test data. Approved guideline, 3rd ed. CLSI document M39-A3. Clinical and Laboratory Standards Institute, Wayne, PA. 3. Covington, L. E., et al. 2011. New CLSI recommendations: practical implications of eliminating ESBL testing and implementing new MIC breakpoints for Enterobacteriaceae at a major teaching hospital, abstr. 632. Abstr. Soc.

10.

11.

12.

13. 14. 15.


9.

Session discussants: Eileen Burd, David W. Craft, Gary V. Doern, W. Michael Dunne, Jr., Gina L. Ewald, Betty Forbes, George Goedesky, Larry Hambleton, Sue Kehl, Marjus Kostrzewa, Nathan A. Ledeboer, Susan Novak-Weekley, Carol H. Smith, Melvin P. Weinstein, and Alice S. Weissfeld.

Healthcare Epidemiol. Am. 2011 Annu. Sci. Meet. 1 to 4 April 2011, Dallas, TX. http://shea.confex.com/shea/2011/webprogram/Paper4335.html. Dellit, T. H., et al. 2007. IDSA and SHEA guidelines for developing an institutional program to enhance antimicrobial stewardship. Clin. Infect. Dis. 44:159–177. Diekema, D. J., and M. A. Pfaller. 2007. Infection control epidemiology and clinical microbiology, p. 118–128. In P. R. Murray, E. J. Baron, J. H. Jorgensen, and M. A. Pfaller (ed.), Manual of clinical microbiology, 9th ed. ASM Press, Washington, DC. Diekema, D. J., et al. 2001. Clinical microbiology laboratory support for infection control and antimicrobial resistance control, abstr. K-1213. Abstr. 41st Intersci. Conf. Antimicrob. Agents Chemother., Chicago, IL, 22 to 25 September 2001. Harris, D. P., J. C. McGregor, and J. P. Furuno. 2006. What infection control interventions should be undertaken to control multidrug-resistant Gram-negative bacteria. Clin. Infect. Dis. 43(Suppl. 2):S57–S61. Klevens, R. M., et al. 2007. Estimating healthcare associated infections and deaths in U.S. hospitals, 2002. Public Health Rep. 122:160–166. Leverstein-Van Hall, M. A., et al. 2010. Global spread of New Delhi metallobeta-lactamase 1. Lancet Infect. Dis. 10:830–831. Maragakis, L. L., and T. M. Perl. 2010. How can we stem the rising tide of multidrug-resistant Gram-negative bacilli? Infect. Control Hosp. Epidemiol. 31:338–340. McGregor, J. C., et al. 2006. Impact of a computerized clinical decision support system on reducing inappropriate antimicrobial use: a randomized controlled trial. J. Am. Med. Inform. Assoc. 13:378–384. Meltzer, D. O., and J. W. Chung. 2010. U.S. trends in hospitalization and general physician workforce and the emergence of hospitalists. J. Gen. Intern. Med. 25:453–459. Paterson, D. L., and Y. Doi. 2007. A step closer to extreme drug resistant in gram negative bacilli. Clin. Infect. Dis. 45:1179–1181. Peleg, A. Y., and D. C. Hooper. 2010. Current concepts: hospital-acquired infections due to Gram-negative bacteria. N. Engl. J. Med. 369:1804–1813. Weintrob, A. C., et al. 2010. Natural history of colonization with Gramnegative multidrug-resistant organisms among hositalized patients. Infect. Control Hosp. Epidemiol. 31:330–337.