Almyroudis, N.G., Osawa, R., Samonis, G., Wetzler, M., Wang, E.S., McCarthy, P.L., & Segal, B.H. (2016). Discontinuation of systematic surveillance and contact precautions for vancomycin-resistant enterococcus (VRE) and its impact on the incidence of VRE faecium bacteremia in patients with hematologic malignancies. Infection Control and Hospital Epidemiology, 37, 398–403. 

DOI Link

Study Purpose

To evaluate if discontinuing systematic VRE surveillance and contact isolation of colonized patients affects the incidence of vancomycin-resistant enterococcus (VRE) faecium bacteremia

Intervention Characteristics/Basic Study Process

Patients with hematologic malignancies admitted to inpatient leukemia, lymphoma, or bone marrow transplantation (BMT) services on three hematology units and one ICU unit received active VRE surveillance and contact isolation precautions for those colonized with VRE from March 2008 to February 2011 per the institutional policy. Admitted patients underwent weekly surveillance on Mondays by perianal swabs, irrespective of admission date. Patients found to be colonized with VRE were placed on contact isolation for the current and subsequent admissions concordant with the Society of Healthcare Epidemiology of America guidelines. 
 
August 2010: levofloxacin prophylaxis during neutropenia or until fever/infection in adults with acute leukemia/hematopoietic stem cell transplanation (HSCT)
 
March 1, 2011: discontinuation of VRE surveillance/contact isolation of colonized and infected patients
 
Contact precautions/mandatory hand hygiene for colonization with methicillin-resistant Staphylococcus aureus (MRSA), multidrug-resistant Gram-negative bacteria, and Clostridium difficile infection (C-diff) continued during both study periods. 
 
August 2013: daily chlorhexidine-impregnated washcloths for all patients
 
Environmental cleaning practices did not change during either study period.

Sample Characteristics

  • N = 2,319   
  • AGE = 5–94 years
  • MALES: 58/59%, FEMALES: 42/41%
  • CURRENT TREATMENT: Other
  • KEY DISEASE CHARACTERISTICS: Hematologic malignancies: leukemia, lymphoma, BMT
  • OTHER KEY SAMPLE CHARACTERISTICS: Underlying malignancies were ALL/AML, CML/MDS/MPD, HL/NHL/CLL, and plasma cell malignancies. No significant differences existed in either group.

Setting

  • SITE: Single site   
  • SETTING TYPE: Inpatient    
  • LOCATION: Buffalo, NY

Phase of Care and Clinical Applications

  • PHASE OF CARE: Multiple phases of care
  • APPLICATIONS: Pediatrics, elder care, palliative care

Study Design

Prospective nonrandomized observational study comparing the incidence of VRE faecium bacteremia in colonized patients with hematologic malignancies during the period of active surveillance/contact precautions versus no active surveillance/contact precautions

Measurement Instruments/Methods

During both study periods, VRE faecium bacteremia was the primary endpoint, and rates of VRE faecium bacteremia were measured as the number of events/1,000 patient days per month. VRE faecium bacteremia and recurrent VRE faecium bacteremia was measured as single or polymicrobial in one blood culture or more. Recurrent bacteremia was considered as a subsequent episode in the same patient occurring after completion and after sterilization of the bloodstream was documented. Microbial species level/susceptibilities were measured by the institutional clinical laboratory.
 
VRE colonization isolates were measured by a perianal swab, stool, or other nonsterile body site. Incidence of bacteremia related to MRSA and C-diff was also collected and was measured as number of events per 1,000 patient days of care per month. Antibiotic exposure for bacteremic patients within one month before development of bacteremia was also measured as days of antibiotic therapy per 1,000 patient days of care per year. Nursing hours per patient day were collected for both study periods.
 
Time series analysis was used to evaluate trends.

Results

Comparing study periods, no significant difference existed in incidence of VRE bacteremia, MRSA bacteremia, and C-diff. Antibiotic utilization was not significantly different between study periods. Levofloxacin prophylaxis had no affect on the incidence of VRE bacteremia. Daily chlorhexidine bathing showed no effect on VRE colonization/bacteremia. No significant difference existed in aggregate antibiotic use and incidence of bacteremia ≤ 30 days prior between study periods. Nursing hours/patient day was not significantly different during study periods. No significant difference existed in patient demographics, patients per service, or underlying hematologic malignancies between study groups/periods.

Conclusions

In a single-site institution (with sporadic molecular epidemiology of VRE faecium in patients with hematologic malignancies), the incidence of VRE faecium bacteremia was not significantly different comparing study periods—active surveillance/contact precaution per institutional policy and after discontinuation of policy. Incidence of MRSA bacteremia and C-Diff remained stable.

Limitations

  • Baseline sample/group differences of import
  • Risk of bias (no control group)
  • Risk of bias (no blinding)
  • Risk of bias (no random assignment)
  • Unintended interventions or applicable interventions not described that would influence results
  • Findings not generalizable
  • Single-site study
  • Only hematologic malignancies, not generalizable to other cancer types
  • Predictive ability limited because of different sequential groups rather than the same group with time matched controls
  • Nursing hours/day does not clearly estimate care burden/nurse/patient ratio
  • Rates of colonization comparing groups not provided
  • Treatment phase or types not collected comparing groups
  • Lack of data on compliance with hand hygiene and lack of molecular epidemiologic data on VRE isolates from the second study period
  • Role of active surveillance and contact precautions was not examined in clonal outbreaks, so the study may not be applicable to other patient populations with cancer or in outbreaks with different molecular data

Nursing Implications

A single-site study revealed that VRE bacteremia incidence in hematologic malignancy inpatients was not affected by VRE surveillance/contact precautions. Nursing practice measured as hours/patient day was not an effective measure for influencing nursing-sensitive infection-related outcomes. Larger multisite trials that include nursing-sensitive measures are needed to identify the most effective practices essential to prevent/control VRE bacteremia in high-risk patients.