Assessment of different diagnostic modalities for the detection of ventilator-associated pneumonia

Background
Ventilator-associated pneumonia (VAP) is one of the common causes of nosocomial infection. The diagnosis of VAP is subject to marked interobserver variability. The clinical pulmonary infection score (CPIS) is simple, easily applicable and affordable, but may lead to overestimation and unnecessary use of antibiotics and hence microbiological resistance. However, at least 48 h are needed for definitive diagnosis using the microbiological methods, which may lead to delay in starting treatment. The best diagnostic method for the diagnosis of VAP remains unclear. Therefore, there is a great need for a definitive and simple method for the early detection of VAP.


Aim
This study aimed to assess the accuracy of different diagnostic tests for the early detection of VAP.


Results
In terms of the microbiological diagnosis of VAP, we isolated the same organism by protected endotracheal aspirate (EA) and bronchoalveolar lavage in about 76% of patients. The diagnostic performance of CPIS combined with EA gram stain was better than use of CPIS alone, with an area under the curve (AUC) of 0.81. The VAP lung ultrasound score (VPLUS) had a sensitivity and specificity of 70% at cutoff point more than 2, with an AUC of 0.71. Combining CPIS with VPLUS yielded the best diagnostic performance, with the sensitivity and specificity being nearly 85% at CPIS more than or equal to 8+VPLUS more than or equal to 2, AUC 0.92.


Conclusions
LUS is a very promising tool that aids in the diagnosis of VAP based on identification of specific sonographic features. However, it cannot replace microbiological sampling. The combination of clinical or ultrasound findings with gram stain yielded the best diagnostic performance.