VRE Colonization and Infection at the NIH Clinical Center (NIHCC)
Methods: Starting in July 2009, perirectal swabs were collected on admission and weekly from non-colonized patients and plated on selective media. In July 2010, swabs were tested by VanA/VanB PCR (Cepheid), replaced in Dec 2010 with VanA PCR. Patients with PCR+ results were cultured.
Results: Surveillance swabs identified VRE colonization in 65%; clinical cultures identified 35%. Of 215 identified by active surveillance, 24% later grew VRE from clinical cultures.
Among the 215, 65% grew VRE in culture, and 35% had positive PCR, but negative cultures (PCR+/Cx-). Only 30% of PCR+/Cx- patients with subsequent swabs grew VRE from cultures. PCR+/Cx- swabs grew 41 organisms: vancomycin-susceptible E. faecium/faecalis (27%), vancomycin-resistant E. faecalis (5%), E. gallinarum/casseliflavus (44%), and others (24%). PCR had a positive predictive value of 43%, and 95% of identified organisms were not VRE.
Of 140 with initial positive surveillance cultures, 33% grew VRE from clinical cultures. 32% had ≥3 subsequent negative swabs, but 21% later grew VRE a median of 46 days after the last negative surveillance culture.
Overall, 19% had VRE bacteremia. 44% of all patients died, including 49% initially identified in clinical culture, 72% of those who had VRE bacteremia, 44% identified in surveillance culture and 39% who were PCR+/Cx-.
Conclusion: VRE colonization is tenacious; only 17% of our patients appeared to become decolonized. The low positive predictive value of PCR testing is likely influenced by vancomycin resistance genes in non-VRE bacteria. Mortality in patients with VRE colonization/infection is high, and colonization may be a surrogate marker for underlying disease severity. Understanding the dynamics of VRE colonization is essential for reducing its prevalence.
H. Y. Hughes,
A. V. Michelin, None
E. S. Snitkin, None
D. K. Henderson, None
T. N. Palmore, None