To explore the economic impact to a hospital of universal methicillin-resistant Staphylococcus aureus (MRSA) screening.

We used a decision tree model to estimate the direct economic impact to an individual hospital of starting universal MRSA screening and contact precautions. Projected costs and benefits were based on literature-derived data. Our model examined outcomes of several strategies including non-nares MRSA screening and comparison of culture versus polymerase chain reaction–based screening.

Under baseline conditions, the costs of universal MRSA screening and contact precautions outweighed the projected benefits generated by preventing MRSA-related infections, resulting in economic costs of $104,000 per 10,000 admissions (95% CI, $83,000–$126,000). Cost-savings occurred only when the model used estimates at the extremes of our key parameters. Non-nares screening and polymerase chain reaction–based testing, both of which identified more MRSA-colonized persons, resulted in more MRSA infections averted but increased economic costs of the screening program.

We found that universal MRSA screening, although providing potential benefit in preventing MRSA infection, is relatively costly and may be economically burdensome for a hospital. Policy makers should consider the economic burden of MRSA screening and contact precautions in relation to other interventions when choosing programs to improve patient safety and outcomes.

Infect Control Hosp Epidemiol 2015;36(1): 2–13

To create a national policy model to evaluate the projected cost-effectiveness of multiple hospital-based strategies to prevent methicillin-resistant Staphylococcus aureus (MRSA) transmission and infection.

Cost-effectiveness analysis using a Markov microsimulation model that simulates the natural history of MRSA acquisition and infection.

Hypothetical cohort of 10,000 adult patients admitted to a US intensive care unit.

We compared 7 strategies to standard precautions using a hospital perspective: (1) active surveillance cultures; (2) active surveillance cultures plus selective decolonization; (3) universal contact precautions (UCP); (4) universal chlorhexidine gluconate baths; (5) universal decolonization; (6) UCP + chlorhexidine gluconate baths; and (7) UCP+decolonization. For each strategy, both efficacy and compliance were considered. Outcomes of interest were: (1) MRSA colonization averted; (2) MRSA infection averted; (3) incremental cost per colonization averted; (4) incremental cost per infection averted.

A total of 1989 cases of colonization and 544 MRSA invasive infections occurred under standard precautions per 10,000 patients. Universal decolonization was the least expensive strategy and was more effective compared with all strategies except UCP+decolonization and UCP+chlorhexidine gluconate. UCP+decolonization was more effective than universal decolonization but would cost $2469 per colonization averted and $9007 per infection averted. If MRSA colonization prevalence decreases from 12% to 5%, active surveillance cultures plus selective decolonization becomes the least expensive strategy.

Universal decolonization is cost-saving, preventing 44% of cases of MRSA colonization and 45% of cases of MRSA infection. Our model provides useful guidance for decision makers choosing between multiple available hospital-based strategies to prevent MRSA transmission.

Infect Control Hosp Epidemiol 2015;36(1): 17–27

Nasal swab culture is the standard method for identifying methicillin-resistant Staphylococcus aureus (MRSA) carriers. However, this method is known to miss a substantial portion of those carrying MRSA elsewhere. We hypothesized that the additional use of a sponge to collect skin culture samples would significantly improve the sensitivity of MRSA detection.

Hospitalized patients with recent MRSA infection were enrolled and underwent MRSA screening of the forehead, nostrils, pharynx, axilla, and groin with separate swabs and the forehead, axilla, and groin with separate sponges. Staphylococcal cassette chromosome mec (SCCmec) typing was conducted by polymerase chain reaction (PCR).

A total of 105 MRSA patients were included in the study.

At least 1 specimen from 56.2% of the patients grew MRSA. Among patients with at least 1 positive specimen, the detection sensitivities were 79.7% for the swabs and 64.4% for the sponges. Notably, 86.4% were detected by a combination of sponges and nasal swab, and 72.9% were detected by a combination of pharyngeal and nasal swabs, whereas only 50.9% were detected by nasal swab alone (P<0.0001 and P=0.0003, respectively). Most isolates had SCCmec type II (59.9%) and IV (35.7%). No correlation was observed between the SCCmec types and collection sites.

Screening using a sponge significantly improves MRSA detection when used in addition to screening with the standard nasal swab.

Infect Control Hosp Epidemiol 2014;36(1): 28–33

We sought to determine whether the bacterial burden in the nares, as determined by the cycle threshold (CT) value from real-time MRSA PCR, is predictive of environmental contamination with MRSA.

Patients identified as MRSA nasal carriers per hospital protocol were enrolled within 72 hours of room admission. Patients were excluded if (1) nasal mupirocin or chlorhexidine body wash was used within the past month or (2) an active MRSA infection was suspected. Four environmental sites, 6 body sites and a wound, if present, were cultured with premoistened swabs. All nasal swabs were submitted for both a quantitative culture and real-time PCR (Roche Lightcycler, Indianapolis, IN).

At study enrollment, 82 patients had a positive MRSA-PCR. A negative correlation of moderate strength was observed between the CT value and the number of MRSA colonies in the nares (r=−0.61; P<0.01). Current antibiotic use was associated with lower levels of MRSA nasal colonization (CT value, 30.2 vs 27.7; P<0.01). Patients with concomitant environmental contamination had a higher median log MRSA nares count (3.9 vs 2.5, P=0.01) and lower CT values (28.0 vs 30.2; P<0.01). However, a ROC curve was unable to identify a threshold MRSA nares count that reliably excluded environmental contamination.

Patients with a higher burden of MRSA in their nares, based on the CT value, were more likely to contaminate their environment with MRSA. However, contamination of the environment cannot be predicted solely by the degree of MRSA nasal colonization.

Clostridium difficile infection (CDI) is a common hospital-acquired infection. Previous reports on the incidence, risk factors, and impact of CDI on resources in the surgical population are limited. In this context, we study CDI across diverse surgical settings.

We prospectively identified patients with laboratory-confirmed postoperative CDI after 40 different general, vascular, or gynecologic surgeries at 52 academic and community hospitals between July 2012 and September 2013. We used multivariable regression models to identify CDI risk factors and to determine the impact of CDI on resource utilization.

Of 35,363 patients, 179 (0.51%) developed postoperative CDI. The highest rates of CDI occurred after lower-extremity amputation (2.6%), followed by bowel resection or repair (0.9%) and gastric or esophageal operations (0.7%). Gynecologic and endocrine operations had the lowest rates (0.1% and 0%, respectively). By multivariable analyses, older age, chronic immunosuppression, hypoalbuminemia (≤3.5 g/dL), and preoperative sepsis were associated with CDI. Use of prophylactic antibiotics was not independently associated with CDI, neither was sex, body mass index (BMI), surgical priority, weight loss, or comorbid conditions. Three procedure groups had higher odds of postoperative CDI: lower-extremity amputations (adjusted odds ratio [aOR], 3.5; P=.03), gastric or esophageal operations (aOR, 2.1; P=.04), and bowel resection or repair (aOR, 2; P=.04). Postoperative CDI was independently associated with increased length of stay (mean, 13.7 d vs 4.5 d), emergency department presentations (18.9 vs 9.1%) and readmissions (38.9% vs 7.2%, all P<.001).

Incidence of postoperative CDI varies by surgical procedure. Postoperative CDI is also associated with higher rates of extended length of stay, emergency room presentations, and readmissions, which places a potentially preventable burden on hospital resources.

Infect Control Hosp Epidemiol 2015;36(1): 40–46

To determine the frequency, risk factors, and outcomes for vancomycin-resistant Enterococcus (VRE) colonization and infection in patients with newly diagnosed acute leukemia.

Retrospective clinical study with VRE molecular strain typing.

A regional referral center for acute leukemia.

Two hundred fourteen consecutive patients with newly diagnosed acute leukemia between 2006 and 2012.

All patients had a culture of first stool and weekly surveillance for VRE. Clinical data were abstracted from the Intermountain Healthcare electronic data warehouse. VRE molecular typing was performed utilizing the semi-automated DiversiLab System.

The rate of VRE colonization was directly proportional to length of stay and was higher in patients with acute lymphoblastic leukemia. Risk factors associated with colonization include administration of corticosteroids (P=0.004) and carbapenems (P=0.009). Neither a colonized prior room occupant nor an increased unit colonization pressure affected colonization risk. Colonized patients with acute myelogenous leukemia had an increased risk of VRE bloodstream infection (BSI, P=0.002). Other risk factors for VRE BSI include severe neutropenia (P=0.04) and diarrhea (P=0.008). Fifty-eight percent of BSI isolates were identical or related by molecular typing. Eighty-nine percent of bloodstream isolates were identical or related to stool isolates identified by surveillance cultures. VRE BSI was associated with increased costs (P=0.0003) and possibly mortality.

VRE colonization has important consequences for patients with acute myelogenous leukemia undergoing induction therapy. For febrile neutropenic patients with acute myelogenous leukemia, use of empirical antibiotic regimens that avoid carbapenems and include VRE coverage may be helpful in decreasing the risks associated with VRE BSI.

Infect Control Hosp Epidemiol 2015;36(1): 47–53

The findings and conclusions in this report are those of the authors and do not necessarily represent the official position of the Centers for Disease Control and Prevention or the Agency for Toxic Substances and Diseases Registry.

Describe the impact of standardizing state-specific summary measures of antibiotic resistance that inform regional interventions to reduce transmission of resistant pathogens in healthcare settings.

Analysis of public health surveillance data.

Central line–associated bloodstream infection (CLABSI) data from intensive care units (ICUs) of facilities reporting to the National Healthcare Safety Network in 2011 were analyzed. For CLABSI due to methicillin-resistant Staphylococcus aureus (MRSA), extended-spectrum cephalosporin (ESC)-nonsusceptible Klebsiella species, and carbapenem-nonsusceptible Klebsiella species, we computed 3 state-level summary measures of nonsusceptibility: crude percent nonsusceptible, model-based adjusted percent nonsusceptible, and crude infection incidence rate.

Overall, 1,791 facilities reported CLABSIs from ICU patients. Of 1,618 S. aureus CLABSIs with methicillin-susceptibility test results, 791 (48.9%) were due to MRSA. Of 756 Klebsiella CLABSIs with ESC-susceptibility test results, 209 (27.7%) were due to ESC-nonsusceptible Klebsiella, and among 661 Klebsiella CLABSI with carbapenem susceptibility test results, 70 (10.6%) were due to carbapenem-nonsusceptible Klebsiella. All 3 state-specific measures demonstrated variability in magnitude by state. Adjusted measures, with few exceptions, were not appreciably different from crude values for any phenotypes. When linking values of crude and adjusted percent nonsusceptible by state, a state’s absolute rank shifted slightly for MRSA in 5 instances and only once each for ESC-nonsusceptible and carbapenem-nonsusceptible Klebsiella species. Infection incidence measures correlated strongly with both percent nonsusceptibility measures.

Crude state-level summary measures, based on existing NHSN CLABSI data, may suffice to assess geographic variability in antibiotic resistance. As additional variables related to antibiotic resistance become available, risk-adjusted summary measures are preferable.

Infect Control Hosp Epidemiol 2015;36(1): 54–64

Manual surveillance of healthcare-associated infections is cumbersome and vulnerable to subjective interpretation. Automated systems are under development to improve efficiency and reliability of surveillance, for example by selecting high-risk patients requiring manual chart review. In this study, we aimed to validate a previously developed multivariable prediction modeling approach for detecting drain-related meningitis (DRM) in neurosurgical patients and to assess its merits compared to conventional methods of automated surveillance.

Prospective cohort study in 3 hospitals assessing the accuracy and efficiency of 2 automated surveillance methods for detecting DRM, the multivariable prediction model and a classification algorithm, using manual chart review as the reference standard. All 3 methods of surveillance were performed independently. Patients receiving cerebrospinal fluid drains were included (2012–2013), except children, and patients deceased within 24 hours or with pre-existing meningitis. Data required by automated surveillance methods were extracted from routine care clinical data warehouses.

In total, DRM occurred in 37 of 366 external cerebrospinal fluid drainage episodes (12.3/1000 drain days at risk). The multivariable prediction model had good discriminatory power (area under the ROC curve 0.91–1.00 by hospital), had adequate overall calibration, and could identify high-risk patients requiring manual confirmation with 97.3% sensitivity and 52.2% positive predictive value, decreasing the workload for manual surveillance by 81%. The multivariable approach was more efficient than classification algorithms in 2 of 3 hospitals.

Automated surveillance of DRM using a multivariable prediction model in multiple hospitals considerably reduced the burden for manual chart review at near-perfect sensitivity.

Infect Control Hosp Epidemiol 2015;36(1): 65–75

Rapidly growing nontuberculous mycobacteria (RGM) are considered rare pathogens, causing central line–associated bloodstream infection. We identified an outbreak of central line–associated bloodstream infection due to RGM at a hematology-oncology ward during a 5-month period.

Outbreak investigation and literature review.

A Japanese tertiary care center.

Adults who were hospitalized at the hematology-oncology ward from October 15, 2011, through February 17, 2012.

A total of 5 patients with a bloodstream infection due to RGM (4 cases of Mycobacterium mucogenicum and 1 case of Mycobacterium canariasense infection) were identified; of these, 3 patients had acute myeloid leukemia, 1 had acute lymphocytic leukemia, and 1 had aplastic anemia. Four of the 5 patients received cord blood transplantation prior to developing the bloodstream infection. All central venous catheters in patients with a bloodstream infection were removed. These patients promptly defervesced after catheter removal and their care was successfully managed without antimicrobial therapy. Surveillance cultures from the environment and water detected M. mucogenicum and M. canariasense in the water supply of the hematology-oncology ward. The isolates from the bloodstream infection and water sources were identical on the basis of 16S-rRNA gene sequencing.

The source of RGM in the outbreak of bloodstream infections likely was the ward tap water supply. Awareness of catheter-related bloodstream infections due to nontuberculous mycobacteria should be emphasized, especially where immunocompromised patients are at risk. Also, using antimicrobials after catheter removal to treat central line–associated bloodstream infection due to RGM may not be necessary.

Infect Control Hosp Epidemiol 2015;36(1): 76–80

To identify risk factors related to Mycobacterium abscessus subsp. bolletii infection during an outbreak, associated with laparoscopic surgery and to propose recommendations for preventing new cases.

A retrospective cohort study.

A private hospital in Manaus, Brazil.

A cohort of 222 patients who underwent laparoscopic surgery between July 2009 and August 2010 by a single surgical team.

We collected information about the patients and the surgical procedure using a standard form. We included sex, age, and variables with P≤0.2 in the bivariate analysis in a logistic regression model. Additionally, we reviewed the procedures for reprocessing the laparoscopic surgery equipment, and the strains obtained with culture were identified by molecular methods.

We recorded 60 (27%) cases of infection. After multivariate analysis, the duration of surgery beyond 1 hour (odds ratio [OR] 2.4; 95% confidence interval [CI] 1.2–4.5), not to have been the first operated patient on a given day (OR, 2.7; 95% CI, 1.4–5.2), and the use of permanent trocar (OR, 2.2; 95% CI, 1.1–4.2) were associated with infection. We observed that the surgical team attempted to sterilize the equipment in glutaraldehyde solution when sanitary authorities had already prohibited it. Eleven strains presented 100% DNA identity with a single strain, known as BRA100 clone.

Because contaminated material can act as vehicle for infection, ensuring adequate sterilization processing of video-assisted surgery equipment was crucial to stopping this single clonal outbreak of nonturbeculous mycobacteria in Brazil.

Infect Control Hosp Epidemiol 2015;36(1): 81–86

To assess the effectiveness of infection control preparedness for human infection with influenza A H7N9 in Hong Kong.

A descriptive study of responses to the emergence of influenza A H7N9.

A university-affiliated teaching hospital.

Healthcare workers (HCWs) with unprotected exposure (not wearing N95 respirator during aerosol-generating procedure) to a patient with influenza A H7N9.

A bundle approach including active and enhanced surveillance, early airborne infection isolation, rapid molecular diagnostic testing, and extensive contact tracing for HCWs with unprotected exposure was implemented. Seventy HCWs with unprotected exposure to an index case were interviewed especially regarding their patient care activities.

From April 1, 2013, through May 31, 2014, a total of 126 (0.08%) of 163,456 admitted patients were tested for the H7 gene by reverse transcription-polymerase chain reaction per protocol. Two confirmed cases were identified. Seventy (53.8%) of 130 HCWs had unprotected exposure to an index case, whereas 41 (58.6%) and 58 (82.9%) of 70 HCWs wore surgical masks and practiced hand hygiene after patient care, respectively. Sixteen (22.9%) of 70 HCWs were involved in high-risk patient contacts. More HCWs with high-risk patient contacts received oseltamivir prophylaxis (P=0.088) and significantly more had paired sera collected for H7 antibody testing (P<0.001). Ten (14.3%) of 70 HCWs developed influenza-like illness during medical surveillance, but none had positive results by reverse transcription-polymerase chain reaction. Paired sera was available from 33 of 70 HCWs with unprotected exposure, and none showed seroconversion against H7N9.

Despite the delay in airborne precautions implementation, no patient-to-HCW transmission of influenza A H7N9 was demonstrated.

Infect Control Hosp Epidemiol 2015;36(1): 87–92

A systematic literature review was performed to assess the impact of surgical-staff behaviors on the risk of surgical site infections. Published data are limited, heterogeneous, and weakened by several methodological flaws, underlying the need for more studies with accurate tools.

To assess the current literature regarding the impact of surgical-staff behaviors on the risk of surgical-site infection (SSI).

Systematic literature review.

We searched the Medline, EMBASE, Ovid, Web of Science, and Cochrane databases for original articles about the impact of intraoperative behaviors on the risk of SSI published in English before September 2013.

We retrieved 27 original articles reporting data on number of people in the operating room (n=14), door openings (n=14; number [n=6], frequency [n=7], reasons [n=4], or duration [n=3]), surgical-team discipline (evidence of distraction; n=4), compliance with traffic measures (n=6), or simulated behaviors (n=3). Most (59%) articles were published in 2009–2013. End points were the 30-day SSI rate (n=8), air-particle count (n=2), or microbiological air counts (n=6); 11 studies were only descriptive. Number of people in the operating room and SSI rate or airborne contaminants (particle/bacteria) were correlated in 2 studies. Door openings and airborne bacteria counts were correlated in 2 observational studies and 1 experimental study. Two cohort studies showed a significant association between surgeon interruptions/distraction or noise and SSI rate. The level of evidence was low in all studies.

Published data about the impact of operating-room behaviors on the risk of infection are limited and heterogeneous. All studies exhibit major methodological flaws. More studies with accurate tools should be performed to address the influence of operating room behaviors on the infectious risk.

Infect Control Hosp Epidemiol 2015;36(1): 93–106