such as the use of corticosteroids, head trauma, and lung structural disease among others
(24,48,49,51,57,262–266).
Patients with none of these risk factors can be started on therapy with reduced-spectrum
drugs such as ceftriaxone; a fluorquinolone (levofloxacin, moxifloxacin); ampicillin/
sulbactam; or ertapenem. If the patient has any of the risk factors for an MDR pathogen,
then a two- to three-drug regimen should be started, including an anti-Pseudomonasbeta-
lactam agent (cefepime or ceftazidime, or piperacillin/tazobactam or imipenem or
meropenem); a second anti-Pseudomonas agent (aminoglycoside or an anti-Pseudomonas
fluoroquinolone such as ciprofloxacin or levofloxacin); and a broad-spectrum agent against
gram-positive microorganisms (linezolid or vancomycin) (see Table 6). Treatment should be
started immediately after obtaining adequate samples for microbiological diagnosis.
Treatment Based on Knowledge of the Etiologic Microorganism
A key issue in the antimicrobial treatment of VAP is the de-escalation of treatment once
microbiological information becomes available. We have already mentioned that antimicrobial
agents should be discontinued when appropriate culture results are negative. Targeted
therapy strives to reduce antibiotic use while detecting no evidence of harm in the
management of patients with VAP (267).
Once 24 to 48 hours have passed, information on the number and type of micro-
organisms growing in culture should be available. According to whether gram-negative
microorganisms or gram-positive microorganisms are lacking, the specific drug against the
corresponding microorganisms can be withdrawn even before the identity and susceptibility
of the etiologic agent is known.
Microorganisms that deserve most attention are MRSA,P. aeruginosa,andA. baumannii.
Vancomycin is presently the standard agent against MRSA, although both industry-
sponsored clinical trials and data from individual centers have consistently reported clinical
failure rates of 40% or greater, at least using a standard dose. New evidence suggests that
vancomycin failure could be related to inadequate dosing (268,269), and some authors argue
that trough levels of around 15 to 20 mg/L are needed (270), although the success of this
strategy requires confirmation in clinical trials. On the other hand, higher vancomycin MICs
themselves may be associated with worse outcomes in patients with VAP due to MRSA. This
was suggested in a prospective cohort study of 95 patients with nosocomial MRSA infection
Table 5 Performance of the Different Culture Methods for Diagnosing VAP
Diagnostic technique Cutoff Sensitivity Specificity References
Conventional
Tracheal aspirates 105 cfu/mL 80% (60–97) 62% (41–74) (227,246–248)
Tracheal aspirates 106 cfu/mL 66% (38–82) 78% (72–85) (229,249,250)
BAL 104 cfu/mL 73% (42–93) 82% (45–100) (203,230,251–255)
Protected specimen
brush
103 cfu/mL 66% (33–100) 90% (50–100) (203,254,255)Plugged telescoping
catheter
103 cfu/mL 72% (54–100) 82% (58–93) (246,256,257)Blind
Tracheal aspirates 105 cfu/mL 94% 50% (205)
Bronchial suction 103 –10^4 cfu/mL 74%–97% 74%–100% (258)
Mini BAL 103 –10^4 cfu/mL 63%–100% 66%–100% (205, 258)
Protected specimen
brush
103 cfu/mL 66% (54–98) 91% (57–100) (248,258,259)Plugged telescoping
catheter
103 cfu/mL 65% 83% (205)Range given in parenthesis.
Abbreviations: CFU, Colony Forming Units; BAL, Bronchoalveolar Lavage.
Nosocomial Pneumonia in Critical Care 191