Abstract 2384: Correlation of Clinical and Radiological Signs with the Thoracic Impedance during the Evolution of Acute Heart Failure
Currently, there are no reliable non-invasive techniques to predict acute heart failure (AHF) before appearance of clinical signs. Treatment is initiated only when overt edema occurs. To validate the ability of lung impedance (LI) to predict evolving AHF in patients hospitalized for acute myocardial infarction (AMI) by comparison to radiological findings. We used a new non-invasive technique for LI measurement that is 15-fold more sensitive than the traditional technique. A radiological score (RS) was devised for quantification of lung fluid with the following signs: blood redistribution, cardiac silhouette, new pleural effusion, peribronchial cuffing, Kerley B-lines and batwing edema. These determined RS used for correlation with LI. A RS of 0 −2 indicated no AHF, 3– 4 signified interstitial edema, and 5–10 - different degrees of alveolar edema. 584 patients admitted for AMI were screened. 509 of enrolled patients had normal physical examination and chest x-ray on admission and were monitored for 94.6 ± 35.7 hours. RS score was 0.8 ± 0.5 at study onset. 392 of 516 patents did not develop AHF. Their maximal LI decrease was 5.6% from baseline (95% CI, 0 −12%, p = 0.3) when RS was 1.3 ± 0.5 (p = 0.06). 124 patients with AMI developed clinical and radiological signs of AHF. At the preclinical stage, LI decreased by 14.9 ± 1.6% from baseline (p < 0.0001) and RS was 2.9 ± 0.5 (p < 0.05). When rales appeared, LI decreased by 20.9 ± 2.7% (p < 0.0001) and RS increased to 4.5 ± 0.8, (p < 0.001). When full AHF developed, LI decreased by 36.1 ± 5.5% and corresponding RS increased to 8.0 ± 1.1. Five patients died at this stage and the rest successfully treated, with LI and RS returning to baseline. LI highly correlated with RS (r = −0.9, p < 0.001). Importantly, the time elapsed from LI decrease > 12% (12% is maximal LI decrease in group without AHF) to the appearance of lung rales and radiological signs of alveolar lung edema was 242 ± 105 min (CI: 30 to 480 min). The chest x-ray used in this study verified the ability of surface thoracic impedance monitoring to predict AHF. It confirmed that the new method of LI measurement is of sufficient sensitivity to detect evolving AHF in patients with AMI when there is ample time to initiate therapy in an attempt to prevent AHF.