Preoperative Prediction Of Granulation Pattern Subtypes In Gh-Secreting Pituitary Adenomas

Objective The aim of this study was to establish a preoperative prediction method for sparsely granulated (SG) growth hormone (GH)-secreting pituitary adenoma, an aggressive tumour subtype with high recurrence risk, in acromegaly patients.Methods Eighty-three patients with GH-secreting pituitary adenomas were included in this study. GH measurements, cytokeratin immunostaining and electron microscopy were performed to detect granulation patterns. Preoperative factors, including general, radiological and endocrinological features and acute octreotide suppression test outcomes, were compared between SG and densely granulated (DG) groups. The predictive capabilities of these features were analysed using a receiver operating characteristic (ROC) curve, and the most predictive features were combined to establish a grading scale.Results Thirty-nine of the 83 patients had SG GH-secreting pituitary adenomas; 44 had DG tumours. SG tumours tended to occur in younger patients and have larger diameters and volumes, higher Knosp grades, lower GH indexes and normalized insulin-like growth factor-1 (IGF-1) level, and a lower increment GH% after octreotide treatment. The tumour size, Knosp grade, GH index and increment GH% after octreotide treatment had good predictive performance, with area under the curve (AUC) values ranging from 0.70 to 0.80. Combining four parameters, including diameter, Knosp grade, GH index and increment GH% after octreotide treatment, we established a grading scale for predicting SG GH-secreting pituitary adenomas with an AUC of 0.84 and relatively high sensitivity and specificity.Conclusions We propose a predictive method for distinguishing SG and DG GH-secreting pituitary adenomas preoperatively. This method will help physicians identify candidates for presurgical medical treatment and neurosurgeons determine radical surgical strategies for high-risk tumours.