VAP is the most common acquired infection in mechanically ventilated patients. It occurs more than 48 hours after intubation with a 2-10 times higher mortality risk. There is no gold standard for diagnosis and therapy is usually empiric. A number of studies have demonstrated the appropriateness of the initial antibiotic regimen as a vital factor in determining outcome. Therefore, the correct choice of the empiric regimen is crucial. A VAP protocol was introduced at our institution in October 2001 which included cefepime and gentamicin as the empiric antibiotics of choice (alternative: cefepime and levofloxacin in patients with acute renal failure).
Institutional Review Board approval was obtained. A retrospective chart review was conducted in medical and coronary intensive care unit patients before and after implementation of the VAP protocol (pre-protocol group, n=10 patients vs. post-protocol group, n=15). The pre-protocol study period was January/March 2001 vs. post-protocol study period was January/ March 2002. Inclusion criteria: Available medical records on patients categorized as having VAP based on infection control criteria.
Median age was 73 years (range, 46-77 years) in the pre-protocol vs.58 years (range, 29-86 years) post-protocol group, respectively. Median ventilator days were 30 (range, 5-98) vs. 32 (range, 9-137), P= 0.63). Median ICU length of stay (28 days, range, 5-105 ) vs. 29 (range, 10-137) days, P= 0.52). Median hospital length of stay (30 days, range, 5-105) vs. 39 (range, 10-137) days, P= 0.51). No difference in overall mortality ( P= 0.337, OR 1.94 ( 0.29-13.8), RR 1.33 (0.65-2.73) or attributable mortality ( P=0.346, OR 3.3 (0.22-109.9), RR 1.9 ( 0.37-9.76) were noted.
The implementation of a VAP protocol did not impact healthcare utilization in our institution. Further evaluation is needed to determine severity of illness and other confounding variables that may attribute to health care utilization.
Further studies are needed to assess underlying factors which account for health care utilization.
S. Yakoob, None.