Transmission of Resistant Bacteria in Intensive Care

Transcription

1 c o r r e s p o n d e n c e Transmission of Resistant Bacteria in Intensive Care To the Editor: Antibiotic resistance is one of the most important problems in medicine today. As pointed out by Platt (April 14 issue) 1 in an editorial accompanying two studies on the prevention of infection with resistant bacteria, the investigators had apparently discordant results. In the study by Huskins et al., 2 in which researchers intervened with infection-control measures in intensive care units (ICUs), there was no diminution in the rate of either colonization or infection. In the second study, by Jain et al., 3 researchers had more success in reducing the rate of infection caused by methicillin-resistant Staphylococcus aureus (MRSA) in Veterans Affairs (VA) hospitals. In neither study was there an attempt to reduce the use of antibiotics. Since there is scant restraint on prescribing antibiotics, a report published some four decades ago remains pertinent. 4 A neurosurgical ward had been plagued with antibiotic-resistant klebsiella, including several instances of meningitis. Multiple control measures failed. The epidemic was aborted when all antibiotics were stopped. One wonders whether patients in ICUs who were not receiving antibiotics could be kept in different areas from those receiving them. Such a policy should diminish unnecessary use. The physician would have to consider that ordering antibiotics for a particular patient would result in transfer to an area that would render the patient at increased risk for exposure to resistant organisms. Stephen J. Seligman, M.D. New York Medical College Valhalla, NY stephen_seligman@nymc.edu 1. Platt R. Time for a culture change? N Engl J Med 2011;364: Huskins WC, Huckabee CM, O Grady NP, et al. Intervention to reduce transmission of resistant bacteria in intensive care. N Engl J Med 2011;364: Jain R, Kralovic SM, Evans ME, et al. Veterans Affairs initiative to prevent methicillin-resistant Staphylococcus aureus infections. N Engl J Med 2011;364: Price DJ, Sleigh JD. Control of infection due to Klebsiella aerogenes in a neurosurgical unit by withdrawal of all antibiotics. Lancet 1970;2: To the Editor: The two reports on MRSA surveillance by Huskins et al. and Jain et al. are important contributions with seemingly opposite conclusions. The study by Huskins et al. suggests that ICU surveillance does not have an important effect on MRSA. An important concept to consider is MRSA days captured in predicting a successful program. 1 This statistic considers the sensitivity of MRSA testing, the length of stay for a patient with MRSA, and the reporting turnaround time. This prediction rule indicates no benefit when captured isolation days range from 53 to 72%. 2 As expected, the study by Huskins this week s letters 761 Transmission of Resistant Bacteria in Intensive Care 766 Abiraterone and Increased Survival in Metastatic Prostate Cancer 768 FOLFIRINOX versus Gemcitabine for Metastatic Pancreatic Cancer 769 Selenium and the Course of Mild Graves Orbitopathy 771 gp100 Peptide Vaccine in Melanoma 772 Childhood Diarrhea Deaths after Rotavirus Vaccination in Mexico n engl j med 365;8 nejm.org august 25,

2 et al. supports this rule; the problem was using a culture-based test (with 90% sensitivity) 3 combined with a mean reporting time of 5.2 days, which yielded only 41% of patient-days in the ICU being covered after the reported result. For successful MRSA control, captured isolation days have exceeded 80%. 2,4 Conversely, more than 90% of VA hospitals use real-time polymerase-chainreaction testing with a turnaround time of less than 1 day for testing at admission. A fundamental difference between these two studies thus hinges on the rapid availability of results to optimize the treatment of patients. The data presented by Jain et al. could be used to confirm this rule, if desired. The findings could potentially be used to address an important public policy question regarding MRSA control that could have a major effect on legislation and the optimization of patient care, 5 since an ineffective MRSA screening program will be costly without benefit to patients. Does the laboratory-testing approach matter? We believe it does. Lance R. Peterson, M.D. NorthShore University HealthSystem Evanston, IL lance1@uchicago.edu Tobi Karchmer, M.D. BD Diagnostics La Jolla, CA Fred C. Tenover, Ph.D. Cepheid Sunnyvale, CA Dr. Peterson reports receiving grant support for molecular diagnostics from Cepheid, BD Diagnostics, MicroPhage, Nanosphere, Roche, and Syntezza and lecture fees from Cepheid, BD Diagnostics, and Roche. No other potential conflict of interest relevant to this letter was reported. 1. Robicsek A, Beaumont JL, Paule SM, et al. Universal surveillance for methicillin-resistant Staphylococcus aureus in 3 affiliated hospitals. Ann Intern Med 2008;148: Peterson LR, Diekema DJ. To screen or not to screen for methicillin-resistant Staphylococcus aureus. J Clin Microbiol 2010;48: Paule SM, Mehta M, Hacek DM, Gonzalzles TM, Robicsek A, Peterson LR. Chromogenic media vs real-time PCR for nasal surveillance of methicillin-resistant Staphylococcus aureus: impact on detection of MRSA-positive persons. Am J Clin Pathol 2009; 131: Bowler WA, Bresnahan J, Bradfish A, Fernandez C. An integrated approach to methicillin-resistant Staphylococcus aureus control in a rural, regional-referral healthcare setting. Infect Control Hosp Epidemiol 2010;31: Peterson A, Marquez P, Terashita D, Burwell L, Mascola L. Hospital methicillin-resistant Staphylococcus aureus active surveillance practices in Los Angeles County: implications of legislation-based infection control, Am J Infect Control. 2010; 38: To the Editor: The use of active detection and isolation measures to control hospital infections that are lethal to health care workers (e.g., smallpox, multidrug-resistant tuberculosis, and the severe acute respiratory syndrome) is without controversy. After 3 years of uncontrolled spread of MRSA infections, the University of Virginia Hospital began actively detecting and isolating MRSA-colonized patients in 1980 and eradicated the endemic strain within 1.5 years. Many subsequent studies reported control of antibiotic-resistant pathogens. Long-standing opponents of MRSA control measures demanded a randomized, controlled trial without saying what bias ICU randomization prevented or why randomization was not needed to document isolation control of other hospital pathogens. The study of a 6-month ICU intervention by Huskins et al. showed no control. A 2.7-year study of active detection and isolation in VA hospitals by Jain et al. showed significant control hospital-wide. Which outcome was correct? We believe the latter. We think that the study by Huskins et al. provided false negative results by agreeing with other negative studies, which generally had problems implementing active control measures. The study by Huskins et al. had some important issues (e.g., unacceptable delay detecting colonization and inadequate isolation compliance) that could have compromised its validity. Barry M. Farr, M.D. University of Virginia Health System Charlottesville, VA William R. Jarvis, M.D. Jason and Jarvis Associates Hilton Head Island, SC Dr. Jarvis reports receiving consulting fees from Johnson & Johnson, Becton Dickinson, Teleflex, Bard, Carefusion, and Kimberly-Clark. No other potential conflict of interest relevant to this letter was reported. To the Editor: Huskins et al. report a lack of benefit for active surveillance and expanded barrier precautions in reducing the incidence of MRSA and vancomycin-resistant enterococcus (VRE) in adult ICUs. The use of barrier precautions was less than what the protocol specified, which raises the possibility that stricter compliance might have been effective. An alternative scenario is that patient-to-patient transmission was not the principal mode of acquisition of VRE and MRSA. At our tertiary cancer center, molecular studies showed that the majority of VRE- 762 n engl j med 365;8 nejm.org august 25, 2011

3 correspondence colonized patients carried unique strains, arguing against a common source of transmission. 1 When infection-control interventions fail, it is important to try to learn why. Molecular analysis of MRSA and VRE isolates would have been useful to distinguish a common source of transmission that would be amenable to more intensive control precautions from sporadic acquisition. If the majority of isolates were unique, then efforts would be most productively devoted to understanding host factors (e.g., coexisting conditions and genetic polymorphisms) and environmental exposures (e.g., antibiotics) that predict the risk of acquisition. Nikolaos G. Almyroudis, M.D. Brahm H. Segal, M.D. Roswell Park Cancer Institute Buffalo, NY brahm.segal@roswellpark.org 1. Almyroudis NG, Lesse AJ, Hahn T, et al. Molecular epidemiology and risk factors for colonization by vancomycin-resistant Enterococcus in patients with hematologic malignancies. Infect Control Hosp Epidemiol 2011;32: To the Editor: The trial by Huskins et al. should be interpreted with care before barrier precautions are abandoned as one of the primary tools to prevent transmission. First of all, screening for MRSA included nares only, even though 25% of carriers may be missed with only nasal screening. 1 Second, barrier precautions were assigned only 35.0 to 50.7% of all ICU patient-days because of late reporting of the screening results. This turnaround time is too slow to assign barrier precautions. 2 Third, hand-hygiene adherence was 15.2 to 60.1%, not accounting for the technique. 3 And fourth, not all VRE-colonized patients require contact isolation. 4 This important study clearly shows that only the use of barrier precautions with a higher degree of compliance than what was reported will result in a clinically significant effect on the transmission of resistant pathogens. Andreas F. Widmer, M.D. University of Basel Basel, Switzerland widmera@uhbs.ch 1. Mertz D, Frei R, Jaussi B, et al. Throat swabs are necessary to reliably detect carriers of Staphylococcus aureus. Clin Infect Dis 2007;45: Harbarth S, Fankhauser C, Schrenzel J, et al. Universal screening for methicillin-resistant Staphylococcus aureus at hospital admission and nosocomial infection in surgical patients. JAMA 2008;299: Widmer AF, Conzelmann M, Tomic M, Frei R, Stranden AM. Introducing alcohol-based hand rub for hand hygiene: the critical need for training. Infect Control Hosp Epidemiol 2007;28: Tschudin Sutter S, Frei R, Dangel M, Gratwohl A, Bonten M, Widmer AF. Not all patients with vancomycin-resistant enterococci need to be isolated. Clin Infect Dis 2010;51: To the Editor: Jain et al. report a laudable reduction in rates of MRSA transmission and infection after the institution of a bundled prevention program. It appears that the calculation of transmissions of MRSA did not exclude from the denominator the inpatient days associated with patients who were not at risk for acquisition of MRSA (i.e., patients who tested positive for MRSA on admission or who had a history of MRSA, a number that increased over time with active surveillance). If this is the case, then the reported relative reduction in transmission rates of 17% in the ICU and 21% in non-icu areas may be inflated by the use of a denominator that includes both at-risk and already-colonized patients. 1 Huskins et al. used patient-days at risk as the denominator and did not find a reduction in MRSA acquisition. If the impressive decrease in the rate of MRSA infection was not associated with a concurrent significant decrease in transmissions, then horizontal prevention measures may be more important than vertical measures, such as nasal screening. 2 Kalpana Gupta, M.D., M.P.H. Judith M. Strymish, M.D. VA Boston Healthcare System kalpana.gupta@va.gov Elizabeth Lawler, D.Sc., M.P.H. Massachusetts Veterans Epidemiology Research Information Center 1. Weinstein RA, Huang SS. Health care-associated infection: assessing the value and validity of our measures. Clin Infect Dis 2009;48: Wenzel RP. Minimizing surgical-site infections. N Engl J Med 2010;362:75-7. To the Editor: Jain et al. report a significant decrease in rates of health care associated (HCA) transmission and infection caused by MRSA in VA hospitals by using a MRSA bundle. The n engl j med 365;8 nejm.org august 25,

4 median length of stay for patients was 3 days (interquartile range, 2 to 7). HCA infections were defined as a positive clinical culture obtained 48 hours after admission. 1 The rate of HCA infection was calculated by dividing the number of HCA infections by total patient time (i.e., 1000 patient-days) within the specified period. For such rate ratios to be accurate, every observation that contributes to the denominator should be at risk of being in the numerator. However, in this study, patients who were hospitalized for less than 48 hours (about 25% of the cohort) contributed patient time to the denominator but could not have measurable HCA infection. One simple method to avoid such immortal time bias would have been to exclude the patients who were hospitalized for less than 48 hours. This bias could have affected the overall results if there was a differential distribution of immortal time between the study periods. 2 Wassim H. Fares, M.D. David J. Weber, M.D., M.P.H. University of North Carolina Chapel Hill, NC wassim_fares@hotmail.com 1. Horan TC, Andrus M, Dudeck MA. CDC/NHSN surveillance definition of health care-associated infection and criteria for specific types of infections in the acute care setting. Am J Infect Control 2008;36: [Erratum, Am J Infect Control 2008;36: 655.] 2. Suissa S. Immortal time bias in pharmaco-epidemiology. Am J Epidemiol 2008;167: Dr. Huskins and colleagues reply: The interventions that we used in our study were generally similar to those used by Jain et al. (with key differences that were described by Platt in his editorial), but the outcomes were distinct. Although we examined the incidence of colonization or infection with MRSA or VRE, the vast majority of new colonization or infection events represented asymptomatic colonization only. In contrast, Jain et al. reported data primarily on the incidence of MRSA infection, principally MRSA central line associated bloodstream infection (CLABSI) and ventilator-associated pneumonia (VAP). Since initiatives to reduce CLABSI and VAP that were caused by all pathogens were conducted concurrently with the MRSA-specific interventions, it is difficult (and potentially misleading) to directly compare the results of the two studies. We discussed the limitation posed by the turnaround time for the results of surveillance cultures and the implications of this finding on the use of barrier precautions in our study. In their comments, several of the correspondents attribute the results of our study to the percentage of patient-days that patients with MRSA or VRE colonization or infection were cared for with the use of contact precautions. However, their statements ignore the fact that newly admitted patients whose colonization status was unknown were cared for with the use of universal gloving, not just standard precautions, while awaiting the surveillance-culture results. Peterson and Diekema s proposed prediction rule, 1 which assumes that patients are cared for with the use of standard precautions alone while awaiting surveillance-test results, is not applicable to our study. Farr and Jarvis question the need for a randomized, controlled trial of an intervention to reduce transmission of MRSA and VRE. Figure 3 of our article provides the answer. If we examine only the data from intervention ICUs during the intervention period, the incidence of MRSA or VRE colonization or infection appears lower in the period from June through August than in the period from March through May. However, when all the data are viewed, it is clear there is no difference between intervention and control ICUs, which have similar group-level characteristics because of randomization. The failure to include a concurrent, similar control group could have led to an erroneous conclusion regarding the effectiveness of the intervention. We agree with Widmer that culturing multiple body sites may identify more MRSA-colonized patients and that proper, reliable use of alcoholbased hand rubs and barrier precautions is essential. We agree with the comments of Almyroudis and Segal but do not have the results of molecular typing to report. We agree with Seligman that the use of antimicrobial agents may affect the spread of antimicrobial-resistant bacteria. W. Charles Huskins, M.D. Mayo Clinic Rochester, MN huskins.charles@mayo.edu Naomi P. O Grady, M.D. National Institutes of Health Clinical Center Bethesda, MD Donald A. Goldmann, M.D. Harvard Medical School 764 n engl j med 365;8 nejm.org august 25, 2011

5 correspondence Since publication of their article, the authors report no further potential conflict of interest. 1. Peterson LR, Diekema DJ. To screen or not to screen for methicillin-resistant Staphylococcus aureus. J Clin Microbiol 2010;48: Dr. Jain and colleagues reply: We agree with Peterson et al. that the turnaround time for active surveillance testing may be important in decreasing the rates of MRSA transmission and infection. Laboratory data were available from each VA facility nationwide to estimate the time that elapsed between active surveillance testing on admission and the reporting of the results. The turnaround time decreased from a median of 35.0 hours (interquartile range, 8.6 to 53.5) in October 2007 to a median of 12.5 hours (interquartile range, 3.7 to 26.5) in April 2008; the median time remained the same throughout the remainder of the 33-month analysis period (Samore M, Nielson C: personal communication). The VA s turnaround time is consistent with that of other programs that captured more than 80% of MRSA isolation days and successfully reduced the rates of MRSA transmission and infection. 1,2 Gupta et al. and Fares and Weber raise an important issue about at-risk days. Since these VA quality-improvement data were reported from each facility in aggregate form, rather than according to individual patients, we were unable to identify patients who were not at risk and eliminate their patient-days from the denominator. To address the potential effect that eliminating these patient-days would have on rates of transmission and infection, we reanalyzed the data after systematically reducing the total patient-days in the denominator. With denominator reductions ranging from 25% to 75%, the decrease in rates of MRSA transmission and infection remained unchanged and statistically significant (P<0.001 for trend by the Poisson method). Thus, the reported reduction in rates of transmission and infection did not appear to have been inflated by a denominator that included patient-days for those not at risk. Our data suggest that when population-based, or horizontal, infection-control activities, such as hand hygiene and contact precautions, do not lead to a diminution in infection rates, the addition of interventions that are focused on a single organism, or a vertical approach, as noted by Farr and Jarvis, may drive institutional-culture change and perhaps increased compliance with basic infection-control measures. 3 As we have shown, the overall effect may be a diminution in the rate of infection because of the target organism as well as other problem pathogens. The VA s results, and those of other investigators, 2 are consistent with the concept that universal active surveillance testing is associated with decreased rates of MRSA transmission and infection if the turnaround time is short. 1 The challenge now is to determine the most costeffective strategy for active surveillance testing while retaining efficacy. Martin E. Evans, M.D. Veterans Health Administration Lexington, KY martin.evans@va.gov Stephen M. Kralovic, M.D., M.P.H. Veterans Health Administration Cincinnati, OH Rajiv Jain, M.D. Veterans Health Administration Washington, DC Since publication of their article, the authors report no further potential conflict of interest. 1. Peterson LR, Diekema DJ. To screen or not to screen for methicillin-resistant Staphylococcus aureus. J Clin Microbiol 2010;48: Robicsek A, Beaumont JL, Paule SM, et al. Universal surveillance for methicillin-resistant Staphylococcus aureus in 3 affiliated hospitals. Ann Intern Med 2008;148: Wenzel RP. Minimizing surgical-site infections. N Engl J Med 2010;362:75-7. The editorialist replies: The judicious use of antimicrobials is important for the control of resistant pathogens and for other reasons. Remarkably, the evidence is quite meager regarding the effect of specific practices in ICUs. 1 Much of the available evidence has limited applicability, since it describes before-and-after comparisons in single ICUs, often in response to outbreaks. Lack of knowledge about the best use of antimicrobials underscores the need to develop better evidence to guide care. Robust evaluations will require multicenter studies that engage physicians, pharmacists, nurses, microbiologists, outcomes researchers, and medical-system leaders. The creation of solid evidence to guide the use of antimicrobials in ICUs can be a leading example of the learning health care system in action. 2 Richard Platt, M.D. Harvard Pilgrim Health Care Institute n engl j med 365;8 nejm.org august 25,

6 Since publication of his article, the author reports no further potential conflict of interest. 1. Kaki R, Elligsen M, Walker S, Simor A, Palmay L, Daneman N. Impact of antimicrobial stewardship in critical care: a systematic review. J Antimicrob Chemother 2011;66: Institute of Medicine. Redesigning the clinical effectiveness research paradigm: innovation and practice-based approaches: workshop summary. Washington, DC: National Academies Press, Abiraterone and Increased Survival in Metastatic Prostate Cancer To the Editor: De Bono et al. (May 26 issue) 1 report that the CYP17 inhibitor abiraterone increased survival in patients with metastatic castration-resistant prostate cancer. However, CYP17 inhibition leads to a corticotropin-induced mineralocorticoid excess that needs to be efficiently counteracted. 2 In this study, such hormone derangement was mitigated but not abolished by the concomitant administration of prednisone. A greater incidence of hypokalemia and fluid retention was seen in patients receiving abiraterone plus prednisone than in those receiving placebo plus prednisone, and this may have contributed to the greater cardiac toxicity observed. It would be interesting to know how the investigators managed these side effects and the results obtained. In particular, was the prednisone dose increased? Were mineralocorticoid-receptor blockers introduced? Because of its intrinsic mineralocorticoid activity, 3 prednisone may be not the best drug to be used in this context, and the potential introduction of dexamethasone as an alternative should be discussed. 2 Finally, the authors should discuss the utility of monitoring serum corticotropin levels during abiraterone treatment in order to adjust the supplementation therapy on an individual basis. Alfredo Berruti, M.D. University of Turin Orbassano, Italy alfredo.berruti@gmail.com Anna Pia, M.D. San Luigi Hospital Orbassano, Italy Massimo Terzolo, M.D. University of Turin Orbassano, Italy 1. de Bono JS, Logothetis CJ, Molina A. Abiraterone and increased survival in metastatic prostate cancer. N Engl J Med 2011;364: Miura K, Yasuda K, Yanase T, et al. Mutation of cytochrome P alpha gene (CYP17) in a Japanese patient previously reported as having glucocorticoid-responsive hyperaldosteronism: with a review of Japanese patients with mutations of CYP17. J Clin Endocrinol Metab 1996;81: Hindmarsh PC. Management of the child with congenital adrenal hyperplasia. Best Pract Res Clin Endocrinol Metab 2009; 23: To the Editor: With the approval of cabazitaxel as a second-line therapy after the failure of docetaxel in patients with metastatic castrationresistant prostate cancer, it would be interesting to know the views of de Bono et al. on possible clinical measurements that might help in guiding the choice of cabazitaxel or abiraterone acetate. 1 In their study, treatment was continued until disease progression on the basis of the prostate-specific antigen (PSA) level, radiographic imaging, and clinical findings. However, a clarification of whether abiraterone was discontinued for isolated PSA progression is necessary. Finally, the median progression-free survival (5.6 months) was substantially shorter than the median time to PSA progression (10.2 months) in the abiraterone group, which suggests a decoupling of PSA changes and antitumor activity. Did patients have objective progression in the context of PSA responses? Hence, should radiographic assessment be performed more frequently, or is a minimum of 12 weeks of therapy warranted before objective radiographic assessments in accordance with the recommendations of the Prostate Cancer Working Group 2? 2 Guru Sonpavde, M.D. Texas Oncology Houston, TX guru.sonpavde@usoncology.com Dr. Sonpavde reports receiving lecture fees from Centocor Biotech, Sanofi-Aventis, and Dendreon and serving on an advisory board for Dendreon. No other potential conflict of interest relevant to this letter was reported. 1. de Bono JS, Oudard S, Ozguroglu M, et al. Prednisone plus cabazitaxel or mitoxantrone for metastatic castration-resistant prostate cancer progressing after docetaxel treatment: a randomised open-label trial. Lancet 2010;376: Scher HI, Halabi S, Tannock I, et al. Design and end points of clinical trials for patients with progressive prostate cancer and castrate levels of testosterone: recommendations of the 766 n engl j med 365;8 nejm.org august 25, 2011