Daisy‐like cells in a non‐neoplastic pericardial effusion

Abstract

Cells showing a peculiar hyperconvoluted nuclei that resemble flower petals were first described in cytological specimens from the female pelvic peritoneum obtained through culdocentesis. 1 In this original report, the cells are described as reminiscent of daises, a term that is still in use. Subsequent reports describing daisy cells are scarce and they all have in common a similar source: peritoneal pelvic washings from female patients. [2][3][4] There is no mention of these cells in fluids other than of abdomino-pelvic origin. The reason for this striking nuclear morphological change is unknown. Hormonal influence and mechanical forces acting on the pelvic peritoneum have been postulated. 1-4 Whatever the cause is, it is important to be familiarized with this phenomenon to avoid misinterpreting the cells as atypical. In this article, we describe characteristic daisy cells in a non-neoplastic pericardial effusion.Papanicolaou stained material, the nuclear membrane was evident and the chromatin showed a vesicular to finely granular aspect with no hyperchomatism. Most cells showed one or more small nucleoli. Neutrophils were also present and showed a smaller or similar size as daisylike cells. No eosinophils or multinucleated giant cells were seen. Daisy cells expressed vimentin and showed focal expression of CD163. They were negative for calretinin, cytokeratin 7, cytokeratin AE1/AE3, WT1, and CD68. The pericardial window biopsy specimen showed relevant pericardial fibrosis with extracellular matrix deposition and plump fibroblasts. The mesothelial layer was absent and instead a thin fibrin layer with occasional neutrophils, lymphocytes and fibroblastic reparative cells was present. No daisy-like cells were observed in the histologic analysis. No neoplastic cells or granulomas were present. Immunohistochemistry showed numerous myofibroblasts (expressing alpha-smooth muscle actin) a subset of which expressed cytokeratins (AE1/AE3) and calretinin as often happens in serosal surfaces under reparative conditions. A few CD68 macrophages were present. There was no immunoexpression of Langerin, S100 protein and CD1a. The cells reported here show cytomorphological features identical to classical daisy cells described in peritoneal fluids. 1-4 These reports include no immunocytochemical studies confirming the mesothelial nature of these cells. However, it seems the most reasonable explanation. Images most often illustrate monolayered clusters of cells in which cohesivity is appreciated. Daisy cells are large with central nuclei and a moderate amount of cytoplasm. In some images, they are seen within a monolayer in continuity with normal looking mesothelial cells. Reactive pericardial mesothelial changes can be of greater intensity leading to diagnostic difficulties and a higher number of nonconclusive diagnoses. 5-7 The most worrisome changes are mesothelial hypercellularity and hypertrophy, cluster formation and presence of psamomma bodies. It seems that such changes can be related to mechanical forces driven by heart be...