Prevention of Surgical-Site Infections

Transcription

1 correspondence Prevention of Surgical-Site Infections To the Editor: The study by Bode et al. highlights the value of mupirocin chlorhexidine prophylaxis in preventing nosocomial S. aureus infection. Although resistance rates are low in the Netherlands, where Bode et al. conducted their study, our experience at the Providence Veterans Affairs Medical Center has shown that increased use of mupirocin can result in increased rates of resistance in methicillin-resistant S. aureus. Our targeted presurgical surveillance and decolonization program for methicillin-resistant S. aureus with mupirocin chlorhexidine started in 2006, followed by facility-wide surveillance and provider-initiated decolonization in Using Pearson correlation coefficients, we have been evaluating mupirocin resistance in S. aureus since June 2004 and assessing the effect of facilitylevel use of mupirocin. 1-4 In 980 isolates of methicillin-resistant S. aureus, we found that an increase in the monthly use of mupirocin had a significant association with subsequent increases in low-level resistance after 1 month (P = 0.05) and in highlevel resistance after 2 months (P = 0.03). Mupirothis week s letters 1540 Prevention of Surgical-Site Infections 1544 Increased Ambulatory Care Copayments and Hospitalizations 1545 Systolic Heart Failure 1546 Health Care System Rankings 1547 Failure to Validate Association between 12p13 Variants and Ischemic Stroke 1550 Autologous Pancreatic Islet Transplantation for Severe Trauma To the Editor: I am concerned about the generalizability of the findings of Bode et al. (Jan. 7 issue) 1 regarding the identification of nasal carriers of Staphylococcus aureus and the subsequent use of mupirocin nasal ointment and chlorhexidine soap. First, it is unclear how the researchers identified patients who were expected to be hospitalized for 4 or more days, since no specific protocol is provided. Second, key surgical data are lacking, despite the preponderance of surgical patients (88%). No data are provided on the appropriateness of antimicrobial prophylaxis, in particular the timing of administration; 84 of 828 surgical patients (10%) received no prophylaxis. Similarly, no data are provided to explain the high rate of infection: among surgical patients, S. aureus infection developed in 3.6% of those receiving prophylaxis and 8.4% of those receiving placebo; 11 to 12% of patients had non S. aureus infection. Would the intervention be as effective in a hospital with lower baseline rates of infection? Finally, it is unclear whether such a screening protocol would work in settings with a high rate of methicillin-resistant S. aureus infection, as is the case in many community hospitals. 2 Deverick J. Anderson, M.D., M.P.H. Duke University Medical Center Durham, NC dja@duke.edu 1. Bode LG, Kluytmans JA, Wertheim HF, et al. Preventing surgical-site infections in nasal carriers of Staphylococcus aureus. N Engl J Med 2010;362: Anderson DJ, Sexton DJ, Kanafani ZA, Auten G, Kaye KS. Severe surgical site infection in community hospitals: epidemiology, key procedures, and the changing prevalence of methicillin-resistant Staphylococcus aureus. Infect Control Hosp Epidemiol 2007;28:

2 correspondence cin resistance in methicillin-susceptible S. aureus was uncommon during the 4.5-year period (9 cases of resistance in 342 isolates), and the use of mupirocin was not correlated with lagged resistance in the subsequent 12 months. These data suggest an ecologic association between mupirocin use and resistance in methicillin-resistant S. aureus. These findings have implications for facilities instituting decolonization programs, since the increased use of mupirocin may reduce the drug s effectiveness. Kerry L. LaPlante, Pharm.D. University of Rhode Island Kingston, RI kerrytedesco@uri.edu Aisling R. Caffrey, M.S., Ph.D. Veterans Affairs Medical Center Providence, RI Kalpana Gupta, M.D., M.P.H. Massachusetts Veterans Epidemiology Research Information Center Boston, MA 1. Cookson BD. The emergence of mupirocin resistance: a challenge to infection control and antibiotic prescribing practice. J Antimicrob Chemother 1998;41: Deshpande LM, Fix AM, Pfaller MA, Jones RN. Emerging elevated mupirocin resistance rates among staphylococcal isolates in the SENTRY Antimicrobial Surveillance Program (2000): correlations of results from disk diffusion, Etest and reference dilution methods. Diagn Microbiol Infect Dis 2002;42: M100-S19: performance standards for antimicrobial susceptibility testing: 19th informational supplement. Wayne, PA: Clinical and Laboratory Standards Institute, Patel JB, Gorwitz RJ, Jernigan JA. Mupirocin resistance. Clin Infect Dis 2009;49: To the Editor: We do not believe that preoperative S. aureus screening and decolonization provides only a marginal benefit or should be reserved for the highest-risk surgical patients, as suggested by Wenzel 1 in the editorial accompanying the article by Bode et al. S. aureus, a virulent pathogen, can cause a substantial number of surgical-site infections and deaths across a wide spectrum of patients. 2 A total of 91 patients with S. aureus colonization in three studies 3-5 underwent preoperative screening and decolonization; no S. aureus infections developed in these patients, as compared with 32 of 533 patients (6%) with such colonization who received chlorhexidine antisepsis but not decolonization (P = 0.009). Screening also facilitates better perioperative prophylaxis (e.g., for methicillin-resistant S. aureus). Without screening, appropriate switching from cefazolin to vancomycin often does not happen. In one cited study, 4 surgeons routinely switched antibiotics, and methicillin-susceptible S. aureus caused the only surgicalsite infection in a patient who did not receive mupirocin. In another study, 3 surgeons did not switch antibiotics, and methicillin-resistant S. aureus was rare and prophylaxis usually appropriate. In a third study, 5 which was focused on patients with methicillin-resistant S. aureus, those who underwent decolonization also received appropriate perioperative prophylaxis, whereas the other patients who were colonized with methicillin-resistant S. aureus usually did not receive a glycopeptide, a factor that may have influenced the development of 29 surgical-site infections with the methicillin-resistant strain. One advantage of screening is that colonized patients can be isolated to prevent spread. More than 30 studies have shown that active detection and isolation were effective at controlling methicillin-resistant S. aureus infections among surgical patients, and 12 studies of cost-effectiveness reported savings with such prophylaxis. Barry M. Farr, M.D. University of Virginia Health System Charlottesville, VA William R. Jarvis, M.D. Jarvis and Jason Associates Hilton Head Island, SC Dr. Jarvis reports having received consulting fees from Becton Dickinson, Johnson & Johnson, Bard, Kimberly-Clark, and 3M. No other potential conflict of interest relevant to this letter was reported. 1. Wenzel RP. Minimizing surgical-site infections. N Engl J Med 2010;362: Kaye KS, Anderson DJ, Choi Y, Link K, Thacker P, Sexton DJ. The deadly toll of invasive methicillin-resistant Staphylococcus aureus infection in community hospitals. Clin Infect Dis 2008; 46: Price CS, Williams AE, Phillips G, Dayton M, Smith W, Morgan S. Staphylococcus aureus nasal colonization in preoperative orthopaedic outpatients. Clin Orthop Relat Res 2008;466: Salgado CD, Greco CS, Buddington RS, Farr BM. Decrease in total knee replacement (TKR) infection rate after implementing new infection control measures. Presented at the 13th Annual Scientific Meeting of the Society for Healthcare Epidemiology of America, Arlington, VA, April 5 8, abstract. 5. 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: To the Editor: In their study comparing chlorhexidine alcohol with povidone iodine for surgical-site antisepsis, Darouiche et al. (Jan. 7 issue)

3 report that the use of a chlorhexidine alcohol preparation significantly reduced the rate of overall surgical-site infection, as compared with povidone iodine. The authors note that alcoholbased products pose a small risk of fire in the operating room. We wish to point out that an estimated 50 to 200 operating room fires from all causes occur in the United States each year. Given that such fires are preventable, this figure is unacceptably high. With the common use of electrocautery and oxygen, an alcohol-based antiseptic completes the fire triad. Most operating room fires involving alcohol-based antiseptics that have been reported to the Food and Drug Administration (FDA) occurred with the use of 26-ml applicators. The use of a large volume of liquid prolongs the drying time and increases the likelihood of pooling in drapes or soaking hair, increasing the risk of fire. If an alcohol-based applicator is used for antisepsis and the surgical procedure involves electrocautery, we encourage operating room personnel to use the smallest-volume applicator that is practical, adhere to label directions regarding drying time, and avoid pooling in drapes and hair. 2 Karen D. Weiss, M.D., M.P.H. Steven F. Osborne, M.D. Priscilla Callahan-Lyon, M.D. Food and Drug Administration Silver Spring, MD karen.weiss@fda.hhs.gov 1. Darouiche RO, Wall MJ Jr, Itani KMF, et al. Chlorhexidine alcohol versus povidone iodine for surgical-site antisepsis. N Engl J Med 2010;362: American Society of Anesthesiologists Task Force on Operating Room Fires. Practice advisory for the prevention and management of operating room fires. Anesthesiology 2008;108: To the Editor: Darouiche et al. report a significant reduction in the rate of surgical-site infection in a group receiving antisepsis with chlorhexidine alcohol, as compared with povidone iodine, before clean-contaminated surgery. The number of patients who would need to be treated to avoid one infection was 17. In my opinion, the external validity of this estimate, which relies on disease incidence in both groups, has not been fully discussed. Since the introduction of the routine prophylactic use of antibiotics, the rate of surgical-site infection in hospitals that are participating in the National Nosocomial Infections Surveillance system was reported to be 3.3% in clean-contaminated surgery, 1 as my colleagues and I reported using definitions and post-discharge surveillance recommended by the Centers for Disease Control and Prevention (CDC). 2 Consequently, the number that would need to be treated may apply only to centers with the unusual 16.1% rate of surgicalsite infection after clean-contaminated surgery. Jean-Jacques Parienti, M.D., Ph.D. Centre Hospitalier Universitaire Caen, France parienti-jj@chu-caen.fr 1. Culver DH, Horan TC, Gaynes RP, et al. Surgical wound infection rates by wound class, operative procedure, and patient risk index. Am J Med 1991;91(3B):152S-157S. 2. Parienti JJ, Thibon P, Heller R, et al. Hand-rubbing with an aqueous alcoholic solution vs traditional surgical hand-scrubbing and 30-day surgical site infection rates: a randomized equivalence study. JAMA 2002;288: Dr. Bode and colleagues reply: In response to Anderson s question: the identification of patients who were expected to be hospitalized for at least 4 days was based on the opinion of the treating physician on the day of admission. This criterion aimed to identify patients with a higher-thanaverage risk of health care associated infections. This partly explains the high rate of infection that we observed. The inclusion of patients with a high risk of S. aureus infection is another reason. We recruited 109 patients from nonsurgical areas, 391 from cardiothoracic areas, 172 from orthopedic areas, 95 from vascular-surgery areas, and 150 from other surgical wards. The baseline infection rate, which depends on the type of patients selected for this intervention, will definitely influence the number of patients needed to screen and treat to prevent one S. aureus infection. However, since the effect of the intervention is on endogenous infection, the effect size will not be changed. We did not assess the timing of administration of antimicrobial prophylaxis, since prophylaxis was given according to national guidelines and any deviation would be subject to randomization. In agreement with these guidelines, 10% of surgical patients did not receive prophylaxis. The goal of decolonizing S. aureus carriers on 1542

4 correspondence hospital admission is not total, long-term eradication of the bacteria but to temporarily eliminate or reduce the load of S. aureus when the risk of endogenous infection is increased. Mupirocinsusceptible strains, including both methicillinsusceptible and methicillin-resistant S. aureus, will be susceptible to this intervention. Thus, the rate of endogenous health care associated S. aureus infection can be reduced, regardless of the prevalence of methicillin-resistant S. aureus. However, the permanent eradication of this resistant strain may well require additional measures. The report by LaPlante et al. of the emerging resistance to mupirocin emphasizes the need to restrict the use of mupirocin to the treatment of S. aureus carriers. However, from their data, it is unclear how long mupirocin was used by patients undergoing the decolonization program and whether the strains that were found were clonally related. Several studies have suggested that perioperative, short-term use of mupirocin is not associated with the emergence of resistance. 1-3 We agree with Farr and Jarvis that by implementing a screening program, appropriate prophylaxis and isolation methods can be chosen, and that screening and decolonization should not be restricted to patients at highest risk for infection. S. aureus infection can have a devastating outcome in patients, regardless of their a priori level of risk. Lonneke G.M. Bode, M.D. Henri A. Verbrugh, M.D., Ph.D. Margreet C. Vos, M.D., Ph.D. Erasmus University Medical Center Rotterdam, the Netherlands l.bode@erasmusmc.nl 1. Boelaert JR, De Smedt RA, De Baere YA, et al. The influence of calcium mupirocin nasal ointment on the incidence of Staphylococcus aureus infections in haemodialysis patients. Nephrol Dial Transplant 1989;4: Fawley WN, Parnell P, Hall J, Wilcox MH. Surveillance for mupirocin resistance following introduction of routine perioperative prophylaxis with nasal mupirocin. J Hosp Infect 2006; 62: van Rijen MM, Bonten M, Wenzel RP, Kluytmans JA. Intranasal mupirocin for reduction of Staphylococcus aureus infections in surgical patients with nasal carriage: a systematic review. J Antimicrob Chemother 2008;61: Dr. Darouiche and colleagues reply: We appreciate the concern expressed by Weiss and colleagues about operating room fires, which, although rare, can result in serious outcomes and, therefore, must be prevented. The majority of such fires involve the head and neck, a region that has proximity to oxygen, the presence of hair and drapes, and the potential pooling of antiseptic solutions. The FDA-approved labeling for the 26-ml applicator for chlorhexidine alcohol states: Prep carefully: Do not use 26-ml applicator for head and neck surgery, do not use on an area smaller than 8.4 in. 8.4 in., use a smaller applicator instead. Do not drape or use ignition source until solution is completely dry (minimum of 3 minutes on hairless skin, up to 1 hour in hair). Do not allow solution to pool; remove wet materials from prep area. We also encourage the surgical team to review the practices recommended by the Association of Perioperative Registered Nurses. In our study, we showed the operating room staff how to properly use the antiseptic preparations, and no fires occurred. We commend Farr and Jarvis on their quest for better infection control. The editorial by Wenzel specifically assessed the incremental value of an approach that targets a single organism (S. aureus) for screening and decolonization in patients in whom an approach that targets all organisms (optimal surgical-site antisepsis) had already been implemented. As shown in Figure 1 of the editorial, the incremental benefit of targeting S. aureus was 7.5 percentage points in patients whose risk of a surgical-site infection was already reduced by 40%. Although of marginal benefit in this particular scenario, S. aureus screening and decolonization could be more beneficial in other scenarios. Although Parienti correctly states that the National Nosocomial Infections Surveillance project reported a 3.3% incidence of surgical-site infection after clean-contaminated surgery, he did not acknowledge that only 30% of the participating hospitals conducted postdischarge surveillance, and no formal protocol for postdischarge surveillance was followed. 1 Since 14 to 84% of surgicalsite infections manifest after hospital discharge, 2 a lack of postdischarge surveillance underestimates the true rate of such infections. As we discussed in our article, the rate of infection in our study was similar to or lower than the rates reported in trials that used the CDC s definition of surgical-site infection and that implemented adequate follow-up. Even if we were to apply the 41% reduction in infection to a 3.3% baseline infec- 1543

5 tion rate, the estimated number needed to treat to prevent one case of surgical-site infection would be 74, which still is quite appealing from both a clinical and an economic standpoint. Rabih O. Darouiche, M.D. Michael E. DeBakey Veterans Affairs Medical Center Houston, TX rdarouiche@aol.com Michael C. Mosier, Ph.D. Washburn University Topeka, KS David H. Berger, M.D. Michael E. DeBakey Veterans Affairs Medical Center Houston, TX 1. Culver DH, Horan TC, Gaynes RP, et al. Surgical wound infection rates by wound class, operative procedure, and patient risk index. Am J Med 1991;91(3B):152S-157S. 2. Parienti JJ, Thibon P, Heller R, et al. Hand-rubbing with an aqueous alcoholic solution vs traditional surgical hand-scrubbing and 30-day surgical site infection rates: a randomized equivalence study. JAMA 2002;288: Increased Ambulatory Care Copayments and Hospitalizations To the Editor: Trivedi et al. (Jan. 28 issue) 1 report that increasing copayments for outpatient care raised the health care costs for elderly Medicare plan enrollees. I believe this finding should be interpreted with caution. First, in the fee-for-service portion of Medicare, beneficiaries who pay copayments out of pocket have much lower total spending for acute care than those who do not. 2 Second, the study may overstate statistical significance. The authors selected 18 plans and analyzed data on persons in those plans, but they did not control for the clustering of hospitalizations within each plan s entire enrolled population. I performed a Monte Carlo simulation using public-use data from the Healthcare Effectiveness Data and Information Set, repeatedly selecting 18 randomly chosen plans matched to control plans. In 9% of 1000 trials, the absolute difference in rates of hospitalization was as large as that reported in the study by Trivedi et al. That finding suggests far more uncertainty than the published confidence intervals imply. Free health care may improve quality, particularly for the poor, but it is unlikely to lower costs, even for the elderly. Christopher Hogan, Ph.D. Direct Research Vienna, VA 1. Trivedi AN, Moloo H, Mor V. Increased ambulatory care copayments and hospitalizations among the elderly. N Engl J Med 2010;362: Report to the Congress: improving incentives in the Medicare program. Washington, DC: Medicare Payment Advisory Commission, June 2009: (Accessed April 1, 2010, at The Authors Reply: The study cited by Hogan compared spending among Medicare beneficiaries with and without supplemental coverage. 1 With the use of such cross-sectional designs, there is difficulty in accounting for adverse selection (i.e., persons with a greater anticipated use of health services opt for more comprehensive insurance benefits). Further, Hogan s description of the findings was incorrect: there was no difference in spending for acute hospital care between the two groups. Hogan simulated utilization using a public data set that included one observation per Medicare plan. In contrast, we analyzed data from an average of 25,000 enrollees per plan and accounted for each enrollee s use of health care services in the year before the copayment increase. However, the use of one observation per plan would still yield highly significant results. Fifteen of the 18 case plans in our study had increases in inpatient admissions, as compared with the concurrent trends in the matched control plans (P<0.004). Finally, the selection of a different set of controls probably would not have altered our results. The stable rates of utilization of inpatient care observed among the control plans were very similar to the rates observed in the entire national sample. Amal N. Trivedi, M.D., M.P.H. Vincent Mor, Ph.D. Alpert Medical School of Brown University Providence, RI amal_trivedi@brown.edu 1. Report to the Congress: improving incentives in the Medicare program. Washington, DC: Medicare Payment Advisory Commission, June 2009: (Accessed April 1, 2010, at