Florid Vascular Proliferation in Grade 0 Glial Implants from Ovarian Immature Teratoma

Abstract

To the Editor: I read with interest the article published by Baker et al. (1) reporting the presence of florid vascular proliferation in ovarian immature teratomas characterized by an extensive proliferation of endothelial and perithelial cells with glomeruloid or cord-like arrangements. This phenomenon is well known in high-grade intracranial neural tumors. This unusual lesion was, however, initially described in ovarian teratomas in 1983 in a comprehensive monograph on human teratomas (2) and subsequently illustrated in a well known reference text of gynecological pathology (3). Later, Gaudin and Rosai (4) described this phenomenon in a variety of extracranial neural and neuroendocrine tumors, which included ovarian teratomas. The presence of this finding was believed to be a characteristic feature of a neural phenotype and thus could help establish a neurogenic origin for poorly differentiated neoplasms. Although this proliferation in extraovarian sites is associated with high grade or very immature neural areas, in ovarian teratomas it also coexists with mature, grade 0 neural tissues and, indeed, we have observed this situation associated with mature glial peritoneal implants in four instances (Fig. 1) (Table 1). All our cases except case 1 had peritoneal implants at the initial operation but the vascular proliferation did not occur simultaneously in both the primary and in the peritoneal mature metastases except in cases 2 and 3 in which it was present in both the primary tumor and the implants. This discrepancy may have been due to incomplete sampling of the ovarian tumor. Macroscopically, the ovarian tumor in case 3 was extensively hemorrhagic and in cases 1 to 3 the implants were intensely red, which contrasted with the pale yellow color of usual glial implants, which resemble wax drops.TABLE 1: Clinicopathological findings in four cases of ovarian teratomas with florid vascular proliferationFIG. 1.: Grade 0 glial implant from case 2 with proliferating vessels arranged in cords (A); a glomeruloid arrangement of the vascular proliferation was also present (B).As indicated by Baker et al. (1), awareness of this finding may help avoid misdiagnosis of a vascular neoplasm, but it also may be clinically important because it may be responsible for hemoperitoneum. This complication occurred in two of our cases, with fatal consequences in one, because the patient bled uncontrollably during a second-look operation (Fig. 2). The remaining cases had a benign clinical course although in case 2, a minor hemoperitoneum was associated with marked adhesions.FIG. 2.: Superficial denudation, hemorrhage, and necrosis of glial implant in case 1.It seems likely that in ovarian teratoma the mature tissues can produce the same angiogenic factor that is secreted in high-grade neural tumors. Francisco F. Nogales, M.D. David Aguilar, M.D.