ANTHROPOMETRIC INDEX FOR PECTUS EXCAVATUM

Eduardo Baldassari Rebeis, José Ribas Milanez de Campos, Ângelo Fernandez, Luíz Felipe Pinho Moreira, Fabio B. Jatene

Pectus excavatum (PEX) is a depression of the sternum in relation to the costal cartilages. Clinical and objective measures for classifying the defect are rare and difficult to apply. The present study aimed to create an anthropometric index (AI) for PEX as a method for diagnosis and for preoperative and postoperative assessment by comparing it to the Haller index (HI) and to the lower vertebral index (LVI). From December 2001 to February 2004, 2 groups of patients were studied at our institution: a) 30 patients with normal configuration of the thoracic cage, upon physical examination; b) 20 patients with PEX. The latter underwent surgery according to the Ravitch technique modified by Robicsek, and they were evaluated in the postoperative period. All patients were assessed by means of the AI (clinical), HI (tomographic), and LVI (radiographic) measures at the level of deepest deformity in the case of the PEX patients, and in the distal third of the sternum in the normal patients. The patients who had undergone surgery were once again measured between the 60th and the 80th postoperative days. There was a high correlation between the AI and the HI (80% P < .001) and between the AI and the LVI (79% P < .001). The accuracy of the 3 indices was similar, in that the following cut points were established: AI = 0.12, HI = 3.10, and LVI = 0.25. Upon analyzing the preoperative results, we verified that for the 3 indices, over 75% of the patients with pectus excavatum were above the cut points and were confirmed as having the defect. In the postoperative results, the value of the indices found below the cut point was considered within the normal standard, and this occurred in 100% for the AI, in over 50% for the HI, and in 50% for the LVI. The AI allowed adequate measurement of the defect, maintaining a) a high correlation with the HI and the LVI and a high accuracy, similar to the already acknowledged and published indices and b) an efficient comparison between the preoperative measurement and the postoperative results.