ERS/AIPO Intensive Care corner

View all Respiratory Digests

Lung ultrasound combined with procalcitonin for a diagnosis of ventilator-associated pneumonia

Author(s): Juandi Zhou, Jia Song, Shijin Gong, Weihang Hu, Minjia Wang, Aibing Xiao, Cong Zhang and Zhouzhou Dong

Respiratory Care May 2019, 64 (5) 519-527

Respiratory critical care

Digest Author(s): Silvia Mongodi / 10 May, 2019

Ventilator-Associated Pneumonia (VAP) is the most frequent nosocomial infection in Intensive Care Units (ICU) and is associated with longer ICU length of stay, prolonged mechanical ventilation, increased use of antimicrobials, higher health-care costs and mortality.

The early diagnosis of VAP remains a challenge for the intensivist, with uncertain consensus about timeline for administering antibiotics. A combination of radiological/clinical criteria (Clinical Pulmonary Infection Score - CPIS) is currently used for VAP diagnosis; however, it demonstrated low specificity and sensitivity. Lung Ultrasound (LUS) is a useful bedside tool to improve early VAP detection: subpleural consolidations and dynamic linear/arborescent air-bronchogram within lobar/hemilobar consolidations demonstrated to be sensible and specific signs for VAP, respectively. Moreover, the LUS score may be used to monitor lung recovery and antibiotic efficacy.

The primary aim of this prospective multicentre study by Zhou et al. was to evaluate the diagnostic performance of LUS alone and combined with Procalcitonin (PCT) in patients clinically suspected for VAP. The combined use of LUS and PCT showed sensitivity 81,3% and specificity 85,5%, with significantly higher diagnostic accuracy (AUC 0,865) than white blood cell count, PCT alone, C-reactive protein or CPIS. In particular, the CPIS was confirmed to be poorly accurate (AUC 0,665). LUS alone showed high sensitivity (91,7%), but low specificity (63,2%). This study confirms the potential interest of LUS combined to clinical/biochemical parameters to improve bedside diagnosis of VAP.

The main limitations of the study are:

  • diagnostic criteria - semiquantitative culture of tracheal aspirate combined to CT-scan is not recommended (1); 22,4% and 13,2% of non-VAP patients had strongly or very strongly positive semiquantitative culture respectively, leading to difficult to interpret results;
  • ultrasound criteria - the authors considered as “positive signs” for VAP a wide range of pathological findings: subpleural consolidations (defined by the shred sign) and lobar/hemilobar consolidations (visualized as tissue-like pattern) with any air-bronchogram (dynamic, static, absent). Subpleural consolidations demonstrated to be a sensible but not specific sign for VAP, being an ultrasound features of lung contusions, ARDS, pulmonary embolism. Within lobar/hemilobar consolidations, static/absent air-bronchogram suggests non-patent airways and is visualized in obstructive atelectasis. Dynamic punctiform air-bronchogram rules out atelectasis but is not specific for pneumonia. Only linear/arborescent dynamic air-bronchogram has been demonstrated to be highly specific for VAP. This may justify the high sensitivity but low specificity of LUS reported by Zhou and colleagues.

    Comment:

  • Classical clinical/radiological criteria have poor diagnostic performance for VAP and a reliable bedside tool is needed.
  • Lung ultrasound is a useful tool for bedside early diagnosis and monitoring of VAP, mainly if associated with other clinical/laboratory findings (i.e. PCT, purulent secretions).
  • Further studies are needed to well define diagnostic ultrasound criteria for VAP.