In their recent study, the primary objective of Bello and colleagues was to assess the frequency of microbiologic concordance between bilateral lung samples, while secondary objectives were the identification of factors affecting concordance, the assessment of the impact of unilateral versus bilateral lung samples on the choice of appropriate treatment, and the diagnostic accuracy of chest X-ray in VAP management.
This prospective single-centre study included 79 adult patients (age > 18 years) with suspected VAP and no contraindication for bronchoscopy. After review of chest X-rays, bilateral BAL samples were collected by two experienced endoscopists with the use of two separate bronchoscopes and were immediately sent to the lab for separate quantitative cultures. Appropriateness of treatment for cases with discordant BAL cultures was assessed by a second team of physicians in simulated prescribing sessions.
In 36 (46%) patients cultures were bilaterally positive, in 22 (28%) were unilaterally positive, whereas in 21 patients cultures were bilaterally negative. Overall, 92 out of 158 BAL cultures (58.2%) were positive and the main isolates in decreasing order were Pseudomonas aeruginosa, Staphylococcus aureus and Klebsiella pneumoniae.
The presence of an infiltrate on chest X-ray in the sampling area predicted BAL culture positivity with a sensitivity of 76% and a positive predictive value of 61%. Concordance between right and left BAL cultures was found in 47 (59.5%) patients. In multivariate analysis the only factors associated with BAL culture concordance were the presence of bilateral infiltrates on chest X-ray and purulent respiratory secretions. In a simulated prescribing setting, treatment based on unilateral sampling was judged as appropriate in >90% of the cases with discordant results.
Despite the low rate of concordance between right and left lung bronchoscopic sampling, the rate of appropriate antimicrobial coverage was high in the simulated prescribing experiment, whereas, chest X-ray proved inadequate in guiding BAL sampling.
The fact that the study was conducted in a single ICU, with a high incidence of multi-drug resistant pathogens and a trend towards administration of broad-spectrum and/or multiple antibiotic regimens, may limit the extrapolation of the results, as this practice may off-set a potential negative effect of ‘blind’ sampling. Also, the use of a simulation experiment to judge the appropriateness of therapy is not an optimal method to assess the therapeutic value of bilateral versus unilateral BAL cultures. A clinical trial with randomisation of the patients to receive treatment according to unilateral or bilateral cultures results is warranted.
Take home message
In patients with clinically suspected VAP the culture results’ concordance between right and left lung samples was low, especially in the absence of purulent secretions or bilateral opacities on chest X-ray. However, as the choice of pulmonary sampling area had limited effects on the appropriateness of treatment, a single sampling of respiratory secretions, regardless of radiographic opacities, seems a reliable method for VAP management, particularly in ICUs with a high prevalence of multi-drug resistant pathogens.
This article review was submitted by EJRC members Dr Evi Tsigou & Dr Despoina Koulenti on behalf of the WG on Pneumonia.
Bello G, Pennisi MA, Di Muzio F, De Pascale G, Montini L, Maviglia R, Mercurio G,
Spanu T, Antonelli M. Clinical impact of pulmonary sampling site in the diagnosis of ventilator-associated pneumonia: A prospective study using bronchoscopic bronchoalveolar lavage. J Crit Care. 2016;33: 151-7. doi: 10.1016/j.jcrc.2016.02.016