ABSTRACT
AIMS: To evaluate diagnostic accuracy of contrast echocardiography (CE) as compared with CT, for the screening of pulmonary arteriovenous malformations (PAVMs) in hereditary haemorrhagic telangiectasia (HHT); to evaluate the clinical significance of semi-quantitative analysis of a shunt on CE. METHODS AND RESULTS: A blinded prospective study was conducted in 190 consecutive subjects at risk of HHT who underwent screening for PAVMs, including clinical evaluation, pulse oximetry, standard and CE, and chest multirow CT without contrast medium. A semi-quantitative analysis of the shunt size was performed according to the contrast echo opacification of the left-sided chambers: Grade 0, no bubbles; 1, occasional filling with <20 bubbles; 2, moderate filling; 3, complete opacification. The first 100 patients were compared with 100 controls. A total of 119 (63%) patients had positive CE (32.2% Grade 1, 13.1% Grade 2, 11% Grade 3, 6.3% with patent foramen ovale). The overall diagnostic performance of CE was sensitivity 1.00, specificity 0.49, positive predictive value (PPV) 0.32, negative predictive value (NPV) 1.00. The PPV for the different grades was 0.00 for Grade 1, 0.56 for Grade 2, 1.00 for Grade 3; the NPV of Grade 0 was 1.00. A significant correlation was found between the CE grading and the number of PAVM, and complications (P < 0.0001). CONCLUSION: CE is an extremely sensitive procedure for the detection of PAVMs with substantial clinical impact.
Subject(s)
Arteriovenous Malformations/diagnostic imaging , Echocardiography/methods , Pulmonary Artery/abnormalities , Pulmonary Veins/abnormalities , Telangiectasia, Hereditary Hemorrhagic/complications , Tomography, X-Ray Computed/methods , Adolescent , Adult , Aged , Arteriovenous Malformations/diagnosis , Arteriovenous Malformations/etiology , Case-Control Studies , Chi-Square Distribution , Child , Child, Preschool , Embolization, Therapeutic , Female , Genetic Predisposition to Disease , Humans , Male , Microbubbles , Middle Aged , Oximetry , Prospective Studies , Pulmonary Artery/diagnostic imaging , Pulmonary Veins/diagnostic imaging , Sensitivity and Specificity , Single-Blind Method , Young AdultABSTRACT
In the last two decades the use of Computed Tomography (CT) has played an important role in the diagnosis of pneumoconioses because of its spatial resolution and lack of technical limitations if compared to traditional chest radiography, first of all in the early stages. Nevertheless the staging of these diseases is still based on the International Labour Office (ILO) classification issued in 1980, whose application is obtained comparing radiographical findings with a standard set of radiographs provided by ILO itself. Starting from the early nineties many studies attempted to propose a classification system similar to the ILO one, combining different symbols to describe a CT thoracic finding or introducing a "score system" to classify different CT patterns (fibrosis, pleural plaques, etc.) and their severity. Despite these efforts no classification system has been accepted yet. After a short description of technical and practical issues of using CT in the diagnosis of pneumoconioses, this article aims at summarizing different classification attempts.
