Colheita de aspirado traqueal para o diagnóstico microbiológico de pneumonia associada à ventilação mecânica : comparação de duas técnicas

Abstract

The accurate diagnosis of ventilation-associated pneumonia (VAP) remains one of the most difficult challenges faced by professionals who care for patients submitted to mechanical ventilation (MV), as proved by the lack of a gold standard diagnosis and the shortage of validation protocols that cover several distinct facets of multidisciplinary nature involving such diagnosis. The main purpose of this study was to assess the qualitative and quantitative agreement of the results of crops of tracheal aspirate (TA) collected through traditional and protected techniques for the microbiological diagnosis of VAP. To that end, we have performed a clinical, prospective, comparative and single-blinded study. The TA samples were collected by the same professionals: a nurse and a physiotherapist, performer and assistant, respectively. At the time of clinical suspicion of VAP, 40 patients aged ≥ 18 years, who were submitted to MV for a period over than 48 hours, were included in the study. It was obtained a pair of samples of tracheal aspirates from each patient, which was collected by traditional and protected techniques, one immediately after the other one. The enforcement order of the techniques was random, according to randomization held by available online software. Once samples have been collected, they are coded and packed in a thermal box. Next, they are immediately transported to a microbiology laboratory and processed on conventional cultivation techniques. The results were expressed in colony forming units (CFU) per milliliter (ml) and the agents were identified by automation (AES Vitek ® 2 Device, BioMerieux, France). The study was approved by the Ethics Committee on Researches involving Human Beings, under the Protocol nº 114311/2012. The data were statistically analyzed by means of the Minitab for Windows (version 14). Chi-square Test (χ²) and Fisher Exact Test were conducted, both with a level of 95% reliability (significant p value of <0,05) and applied at the intersection of categorical variables. Most of patients were male, elderly, with many days of hospitalization and MV, affected by several comorbidities, with increased number of leukocytes and rods and all of them were being submitted to antibiotic therapy. Through the traditional technique, 23 (57,5%) patients showed, in their samples, microbial growth ≥ 105 CFU/ml. There was qualitative and quantitative agreement between the techniques (p=0,004), however, the greater amount of extras microorganisms and doubtful results were associated with the protected technique. The most common microorganisms identified by both techniques were Acinetobacter baumannii (n=7), Candida albicans (n=6) and Pseudomonas aeruginosa (n=5). These preliminary results did not show significant differences in the results of cultivation of samples through protected and traditional techniques, but raised the necessity for further studies, including analysis of genetic concordance of agents isolated in the endotracheal tube wall and in tracheobronchial secretions.