The incidence of and clinical variables associated with vancomycin-resistant enterococcal colonization in mechanically ventilated patients

Study objectives: (1) To determine in our ICU the incidence of vancomycin-resistant enterococcus (VRE) colonization in mechanically ventilated patients without a history of VRE infection or colonization; and (2) to determine the risk factors and outcome variables associated with VRE colonization in these patients. Design: A prospective cohort study conducted between January 1996 and March 1998. Setting Medical and cardiac critical care units in a tertiary care urban university hospital. Patients: Mechanically ventilated patients without evidence of pneumonia at the onset of ventilation. Interventions: None. Measurements and results: Patients underwent rectal cultures by standard methods on day 1, day 3 or 4, day 6 or 7, and day 14 of intubation to detect VRE. Thirteen of 83 patients (16%) had rectal cultures positive for VRE (VRE+) at some point while being mechanically ventilated during their stay in the ICU. In comparison, approximately 15 of 2,100 medical ICU patients (0.7%) had clinical VRE, infections as determined by the hospital's infection control program during a e-year period. VRE+ patients had a higher incidence of immunosuppression than patients who had rectal cultures negative for VRE (VRE-) (9 of 13 [69%] vs 16 of 70 [23%], respectively; p < 0.01) and neutropenia (4 of 13 [31%] vs 5 of 70 [7%], respectively; p < 0.01). Hospital length of stay (LOS) was longer in VRE+ patients than in VRE- patients (27 +/- 17 days vs 17 +/- 14 days, respectively; p = 0.05), whereas pre-ICU hospital LOS and ICU LOS were similar in both patient groups. Five of 67 patients (7%) were VRE+ on day 1 of intubation, suggesting colonization at a prior site of care. Three of 29 patients who had subsequent rectal cultures converted to VRE+ while in the ICU. This group had a higher incidence of immunosuppression and neutropenia, and received more vancomycin compared with the patients who remained VRE- (p < 0.01). However there was no significant difference in the use of other broad-spectrum antibiotics (such as antipseudomonal penicillins, third-generation cephalosporins, quinolones, and clindamycin), enteral tube feedings, or sucralfate between the two groups. In addition, a topical antibiotic paste (a gentamicin, nystatin, polymixin slurry) that,vas placed in the oropharynx to prevent bacterial overgrowth was not found to increase the incidence of VRE colonization in this patient population. Conclusions: The incidence of VRE colonization was surprisingly high: 16% in mechanically ventilated patients in a hospital in which VRE was not previously known to be endemic. Risk factors for the acquisition of VRE colonization included immunosuppression, neutropenia, and vancomycin use. Increased LOSs and hospital costs were seen in VRE+ patients compared to VRE- patients. Whether VRE colonization is a contributor to severe disease that leads to prolonged hospitalization and increased resource allocation or whether it is simply a marker of disease severity cannot be determined from this study. To the extent that specific antibiotic protocols are used to reduce antibiotic-resistant flora in the ICU, monitoring the incidence of VRE in the stool specimens of immunocompromised, mechanically ventilated patients can be a simple and useful tool to assess ne effect of these strategies.