Chaiwat Aphivatanasiri1 M.D., Sakda Waraasawapati1 M.D., Chawalit Pairojkul1 M.D., Pichayen Daungthongpon2 M.D. 
1Department of Pathology, Faculty of Medicine, KhonKaen University,
2Department of Surgery, Faculty of Medicine, KhonKaen University
Introduction: Pituitary adenoma has been classified as benign according to slow-growing tumor. However, some atypical pituitary adenoma may have clinical aggressiveness such as parasellar invasion and high recurrence rate.Many studies have failed to demonstrate a correlation between Ki-67 labeling index and behavior of this tumor. Eyeballing is a common method for evaluation of Ki-67, however, reliability and reproducibility are in question. Objectives: The primary objective of this study was to compare two different methods of measuring Ki-67 labeling index in pituitary adenoma correlated with clinical outcome.
Materials and Methods: Histopathology of 95 pituitary adenomas which has been operated between 2009 and 2015 were reviewed. Ki-67 immunohistochemistry has been performed and sent for digital slide scanning. Ki-67 labeling index was evaluated at least 2000 tumor cells in hotspot areas with two methods;(1) Eyeballing estimation consensus by two pathologists, (2) Software-Automated counting with Fiji software (ImageJ-based) and plugin (Ret FM-J) validated with manual counting.
Results: There are 74 fi rst-time operated pituitary adenomas specimens with average age of 46. 93 years. Six cases (8.11% ) has been recurred with re-operation.Means Ki-67 labeling index are 2.65 and 1.17for eyeballing and automated counting, respectively. Automated Ki-67 was signifi cant lower than eyeballing ( p<0. 001) . Neither eyeballing nor automated Ki-67 counting demonstrates signifi cant correlation with aggressiveness and recurrence.
Conclusion: Eyeballing estimation for Ki-67 tends to overrate comparing with digital automated counting. According to no signifi cant correlation between Ki-67 and clinical outcome, using of Ki-67 labeling index in pituitary adenomas should be made with caution.