Among 137 thymectomized patients with nonthymomatous myasthenia gravis (MG), six were reoperations. Thymectomy had initially been performed by the transcervical approach, but was ineffective. Extended thymectomy was then performed by median sternotomy. The reoperations revealed residual thymus (average weight, 19 gm) in all cases. Postoperative courses were favorable in three patients and unfavorable in three others. The latter had received high-dose steroid therapy before the second operation. We conclude that total thymectomy by the transcervical approach is almost impossible, and that reoperation may accelerate improvement in some cases.

No abstract

In order to quantify the lymphocyte to epithelial cell ratio (L/E ratio) of the thymoma, actual counting was performed. The L/E ratio was investigated in 25 specimens of 22 cases (22 initial tumors and 3 recurrent tumors). L/E ratios ranged from 0.27 to 4.10, and the mean value was 1.93 f 0.21. The mean L/E ratio of histologically malignant group was lower than that of the benign group. In the three recurrent cases, L/E ratios of the recurred tumor decreased in all cases. It was concluded that the epithelial element increases with the advance.of the tumor. Cancer 40:1222-1228, 1977. €IYMOMA HAS BEEN DEFINED AS A TUMOR T originating from thymic parenchyme composed of epithelial cells and lymphocytes (thymocytes). Therefore, the thymoma fundamentally consists of epithelial cells and lym-phocytes in various proportions and has been classified according to the ratio of those thymic elements by many authors. Because such classifications are made in accordance with the impressions of the observer, we suggest that the criteria of classification are different between authors. In this study, we actually calculated the ratio of lymphocytes to epithelial cells in thymomas (L/E ratio) and demonstrated the correlations between the clinicopathologic findings and the L/E ratio. MATERIALS AND METHODS Thymomas surgically removed at The 1st Department of Surgery, Osaka University Medical School were used. These comprised a total of 22 cases; in three cases recurrent tumors were also investigated. From the 1st 1)epartrnent The patients ranged from 16 to 57 years of age. Fifteen patients were men and seven women. Ten patients demonstrated myasthenia gravis; one, pure red cell aplasia; one, an eosino-philic leukemoid reaction. Method for the Calculation of the L/E Ratio Several sections stained with hemotoxy-lin-eosin were prepared from one specimen. Then microscopic photographs were taken at 10 fields randomly sampled from each section. These photos were divided into 70 demarcated squares (Fig. 1). In order to examine how many squares should be calculated for an available L/E ratio, two methods were used. In six of the 22 initial tumors, L/E ratios were calculated from all 70 squares of a photo. O n the same photo, the accumulated L/E ratios were calculated from various number of squares which were randomly selected. When L/E ratios from various number of squares and those from all 70 squares, were compared, it was clear that the calculation of 10 squares is sufficient for the available L/E ratio of one photo (Fig. 2). In the second method, the statistical test was performed according to the following equation. k 2 X (ir a 1222

In 42 cases of myasthenia gravis with thymoma, the tumor was removed surgically. In four of these cases, myasthenic symptoms had remitted before operation. However, in two of these four cases, myasthenic symptoms returned after operation, despite apparently total thymectomy. In two of the four cases that remitted before operation, the thymus was found to be highly involuted, without germinal centers; in another case, the thymus was found to be moderately involuted, and germinal centers were found. In the nonrelapsing cases, only the tumor itself was removed, with residual thymic tissue being left behind. The thymoma in all cases consisted of polygonal epithelial cells and lymphocytes.