IMPACT OF INVASIVE AND NONINVASIVE QUANTITATIVE CULTURE SAMPLING ON OUTCOME OF VENTILATOR-ASSOCIATED PNEUMONIA - A PILOT-STUDY

We performed an open, prospective, randomized clinical trial in 51 pat ients receiving mechanical ventilation for more than 72 h, in order to evaluate the impact of using either invasive (protected specimen brus h [PSB] and bronchoalevolar lavage [BAL] via fiberoptic bronchoscopy) or noninvasive (quantitative endotracheal aspirates [QEA]) diagnostic methods on the morbidity and mortality of ventilator-associated pneumo nia (VAP). Patients were randomly assigned to two groups: Group A pati ents (n = 24) underwent QEA, PSB, and BAL; Group B patients (n = 27) u nderwent only QEA cultures. Empiric antibiotic treatment was given acc ording to the attending physician and was modified according to the re sults of cultures and sensitivity in Group A using PSB and BAL results and in Group B based upon QEA cultures. Bacteriologic cultures were d one quantitatively for EA, PSB, and BAL. Thresholds of greater than or equal to 10(5), greater than or equal to 10(3), and greater than or e qual to 10(4) CFU/ml were used for QEA, PSB, and BAL, respectively. Mi crobial cultures from Group A patients were positive in 16 (67%) BAL s amples, 14 (58%) PSB samples, and 16 (67%) QEA samples. In Group B pat ients, QEA microbial cultures yielded positive results in 20 of 27 (74 %) samples. In Group A, there was total agreement between culture resu lts of the three techniques on 17 (71%) occasions. in five (21%) cases , QEA coincided with either BAL or PBS. In only two (8%) cases, QEA cu ltures did not coincide with either PSB or BAL. No cases of positive B AL or PSB cultures had negative QEA cultures. Initial antibiotic treat ment was modified in 10 (42%) patients from Group A and in four (16%) patients from Group B (p < 0.05). The observed crude mortality rate wa s 11 of 24 (46%) in Group A, and 7 of 27 (26%) in Group B, whereas the adjusted mortality rates (observed crude minus predicted at admission ) for Groups A and B were 29 and 10%, respectively. There were no stat istically significant differences when comparing crude and adjusted mo rtality rates of Groups A and B. There were no differences in mortalit y between both groups when comparing pneumonia, considering together P seudomonas aeruginosa and Acinetobacter spp. (Group A, 33% versus Grou p B, 27%). There were no differences between Groups A and B with regar d to ICU stay duration and total duration of mechanical ventilation. I n this pilot study, the impact of bronchoscopy was to lead to more fre quent antibiotic changes with no change in mortality. Further studies with larger population samples are warranted to confirm these findings .