Trauma intensive care unit (TICU) patients requiring mechanical respiratory support frequently develop ventilator-associated pneumonia (VAP). Oral and oropharyngeal bacteria are believed to be responsible for many cases of VAP, but definitive evidence of this relationship is lacking. Earlier studies used conventional culture-based methods for identification of bacterial pathogens, but these methods are insufficient, as some bacteria may be uncultivable or difficult to grow. The purpose of this study was to use a culture-independent molecular approach to analyze and compare the bacterial species colonizing the oral cavity and the lungs of TICU patients who developed VAP. Bacterial samples were acquired from the dorsal tongue and bronchoalveolar lavage fluid of 16 patients. Bacterial DNA was extracted, and the 16S rRNA genes were PCR amplified, cloned into Escherichia coli, and sequenced. The sequencing data revealed the following: (i) a wide diversity of bacterial species in both the oral and pulmonary sites, some of them novel; (ii) known and putative respiratory pathogens colonizing both the oral cavity and lungs of 14 patients; and (iii) a number of bacterial pathogens (e.g., Dialister pneumosintes, Haemophilus segnis, Gemella morbillorum, and Pseudomonas fluorescens) in lung samples that had not been reported previously at this site when culture-based methods were used. Our data indicate that the dorsal surface of the tongue serves as a potential reservoir for bacterial species involved in VAP. Furthermore, it is clear that the diversity of bacterial pathogens for VAP is far more complex than the current literature suggests.Ventilator-associated pneumonia (VAP) frequently occurs in patients requiring mechanical respiratory support, with incidence from 8% to 28% and mortality rates from 24% to 76%, depending on the population studied and the techniques used for the diagnosis of pneumonia (3). VAP is often associated with prolonged hospitalization of trauma patients, which can result in additional hospital charges of about $40,000 per patient (1).The etiology of VAP is variable, depending on the time of onset, duration of hospitalization, population studied, and hospital setting (3). For example, Staphylococcus aureus, Streptococcus pneumoniae, and Haemophilus influenzae usually predominate in early-onset VAP. Aerobic gram-negative bacteria, including members of the family Enterobacteriaceae, have been isolated in both early-and late-onset pneumonia. Pseudomonas aeruginosa and Acinetobacter and Enterobacter species are often isolated from late-onset pneumonia. Trauma intensive care unit (TICU) patients are at high risk of infection with S. aureus, among a variety of other microorganisms. In the postsurgical population, H. influenzae and S. pneumoniae dominate in trauma patients but not in patients with other diagnoses (3, 5). In one study, oral anaerobic bacteria such as Prevotella, Veillonella, and Fusobacterium spp. were isolated from VAP patients (4).Several different routes of infection by VAP pathogens hav...