Vibration response imaging in the detection of pleural effusions: a feasibility study.

BACKGROUND

Computerized analysis of the regional distribution of breath sound intensity during respiration has generated interest as a possible diagnostic modality.

OBJECTIVES

We hypothesized that pleural effusions would create a dependent region of absent breath sounds and thus vibration response imaging (VRI) could be used in the detection of such pleural effusions.

METHODS

A prospective, single-blinded and open-labeled trial was carried out, and VRI recordings were compared to upright chest X-rays, bedside ultrasound examinations and volume of fluid drained via thoracentesis. VRI images were interpreted by a physician who was blinded to the patients' clinical history, physical examination and diagnostic tests. Quantitative assessment of pleural effusion size in the VRI images was performed by ImageJ software and an automatic pixel count analysis.

RESULTS

VRI recordings were performed on 57 consecutive patients and correctly predicted the diagnosis in 45 cases (45/56, 80%) as compared to chest X-rays. The calculated sensitivity, specificity, positive predictive value and negative predictive value for diagnosis of pleural effusion were 86% (62/72), 93% (37/40), 95% (62/65) and 79% (37/47), respectively, in a per-hemithorax analysis. In the quantification of effusion size, there were high correlations between VRI images and chest X-ray area as assessed by ImageJ (r = 0.67) and pixel count (r = -0.77). The level of agreement between VRI readings and ultrasonography was 75% (41/55), and correlation with the volume of fluid drained in therapeutic thoracentesis was moderate (r = -0.49). No side effects from the VRI recordings were documented.

CONCLUSIONS

VRI can be used to detect and quantify pleural effusions.