Your browser doesn't support javascript.
loading
The diagnostic value of automated quantitative DNA cytometry for pancreatic malignancy / 中华消化内镜杂志
Chinese Journal of Digestive Endoscopy ; (12): 157-162, 2018.
Artigo em Chinês | WPRIM | ID: wpr-711498
ABSTRACT
Objective To estimate the diagnostic value of cytology, DNA-ICM(DNA-image cytometry),cytology combined with DNA-ICM for pancreatic malignancy,and to explore the cut-off value for DNA-ICM. Methods Patients with suspicious pancreatic malignancy were retrospectively identified. In total,145 EUS-FNA specimens acquired from 140 separate patients were examined by cytology and DNA-ICM. Diagnostic values among cytology, DNA-ICM and the combination of the techniques in detecting pancreatic malignancy were compared. Results Compared with cytology, DNA-ICM had a lower sensitivity (63.0% VS 82.4%)and accuracy(69.7% VS 85.5%). After combining the techniques, the diagnostic value for pancreatic malignancy significantly improved compared with that by cytology(0.941 VS 0.912, P=0.007 0)or DNA-ICM only(0.941 VS 0.815, P<0.000 1). By using the Youden index, the cut-off value for DNA-ICM to detect pancreatic malignancy was one cell with DI(DNA index)≥2.5. Notably,with this standard, the sensitivity and accuracy of DNA-ICM significantly increased to 72.3% and 77.2%, and those of the combined techniques increased to 91.6% and 93.1%, respectively. Conclusion Automated DNA-ICM is an objective and effective method for pancreatic malignancy. Although DNA-ICM has a lower diagnostic value than that of conventional cytology, an improved value was obtained after combining the techniques.

Texto completo: DisponíveL Índice: WPRIM (Pacífico Ocidental) Tipo de estudo: Estudo diagnóstico Idioma: Chinês Revista: Chinese Journal of Digestive Endoscopy Ano de publicação: 2018 Tipo de documento: Artigo

Similares

MEDLINE

...
LILACS

LIS

Texto completo: DisponíveL Índice: WPRIM (Pacífico Ocidental) Tipo de estudo: Estudo diagnóstico Idioma: Chinês Revista: Chinese Journal of Digestive Endoscopy Ano de publicação: 2018 Tipo de documento: Artigo