Diagnosis and grading of intrapulmonary vascular dilatation in cirrhotic patients with contrast transesophageal echocardiography

Background/Aims: The use of transesophageal contrast echocardiography (TOCE ) in the diagnosis of intrapulmonary vascular dilatation (IVD) and hepatopu lmonary syndrome (HPS) needs to be studied. We tested the specificity of TO CE using traditional criteria and the value of a new method based on TOCE, a grading scale and a selected contrast. Methods: 1) Several solutions were tested and two were selected: 20% mannit ol and 0.9% saline. 2) 71 cirrhotic patients and 20 controls were studied. Left atrium opacification with contrast was classified into 6 degrees by TO CE, Mild and significant IVD were considered in relation to results in cont rols, Patients mere studied with saline and mannitol-TOCE, Results were com pared to transthoracic contrast echocardiography (TTCE), to gas exchange ab normalities and to Child class. Results: The reproducibility of TOCE grading was excellent, (kappa >0.9). I VD detection using TTCE, mannitol-TOCE and saline-TOCE was 29.5%, 55% (25% mild and 30% significant), and 45% (38% mild and 7% significant), respectiv ely. The best agreement with TTCE (reference method) was obtained with mann itol-TOCE, using significant IVD as the cut point. By this criterion, 18% r eached the criteria of HPS using TTCE and 22% using mannitol-TOCE. Patients with IVD by TTCE had non-significant changes in gas exchange determination s, Patients with significant IVD by saline TOCE had lower mean PaO2 levels (67.3+/-14 vs. 79.5+/-11 mmHg, p<0.05) than patients without IVD, Patients with significant IVD by mannitol TOCE had higher mean AaPO(2) (29.3+/-14 vs . 19.7+/-9 mmHg; p<0.005) and lower mean PaCO2 levels (30.1+/-4.4 vs. 33.3/-4.8 mmHg; p<0.05) than patients without IVD, Severity of IVD by TOCE corr elated to Child class (r=0.43;p<0.001). Conclusions: The presence of contrast in the left atrium cannot be a criter ion of IVD when TOCE is used. Our semi-quantitative scale has proved to be feasible and reproducible, presenting a good agreement with TTCE, and has s hown better correlation with gas exchange abnormalities and Child class. Sa line TOCE appears to be more specific in the detection of hypoxemic patient s with IVD, but mannitol TOCE adds sensitivity.