Early diagnosis and estimation of pulmonary congestion and edema in patients with left-sided heart diseases from histogram of pulmonary CT number.

To establish an early diagnosis and have a quantitative estimation of congestion or edema, pulmonary CT scans were performed to detect the increase of regional lung water volume of patients with congestive heart failure. CT scanning of the upper, middle, and lower lung fields at maximum inspiration and hemodynamic analysis to ensure a hemodynamically chronic stable condition after the heart failure therapy for 5 weeks by cardiac catheterization within 2 weeks before CT scanning were performed in 10 normal subjects and 25 patients with left-sided heart failure. Patients were divided into two groups according to the presence or absence of pulmonary congestion or edema on chest radiographs, and the mean pulmonary wedge pressure (mPWP). The pulmonary congestion group (14 patients) had pulmonary congestion or edema, and mPWP was greater than 13 mm Hg despite the heart failure therapy. The nonpulmonary congestion group (11 patients) had no pulmonary congestion or edema, and mPWP was lower than 10 mm Hg. Histograms of pulmonary CT numbers were graphed to a set region of interest (ROI) on transverse CT images in all subjects; then changes of pulmonary CT numbers when pulmonary congestion or edema occurred were estimated from those histograms. We found that when severe pulmonary congestion or edema occurred in 6 patients with severe left heart failure, the pulmonary CT numbers within the ROI rapidly increased from -650 to -750 Hounsfield units (HU). We defined the amount ratio between the pixels (pixel counts between -650 and -750 HU within these R0Is) and the pixels (pixel counts between -300 and -950 HU) of the CT number (which were all pixels composing these R0Is) x100(%) as "%PXL." %PXL showed a linear correlation with mPWP. Compared with the %PXL in the normal group, the %PXL in the pulmonary congestion group was significantly higher than that in the normal group (p<0.05, p<0.01, p<0.001) in all lung fields. Similarly, compared with the %PXL in the normal group, %PXL in the nonpulmonary congestion group was significantly higher than that in the normal group in the middle and lower lung fields (p<0.05, p<0.01) in the posterior field. From these results, we concluded that %PXL was a good index to identify abnormal water content usually not recognizable by routine chest radiograph in the lungs.