Background: The treatment of choice in prolactinomas are dopamine agonists, with surgery reserved for cases refractory to or poorly tolerant of pharmacotherapy. Little is known on the associations between clinical, neuroradiological and pathological features influencing the efficacy of transphenoidal surgery.
The Aim: To evaluate clinical, neuroradiological and neuropathological aspects of surgically treated prolactinomas in women, with a particular emphasis on factors affecting treatment efficacy.
Methods: This cohort study included 44 female patients aged 29.11±8.7 yr, qualified for surgery due to unsuccessful medical treatment. Clinical, pathological and imaging parameters were evaluated in light of surgical outcomes.
Results: The group included 25 microadenomas (56.8%) and 19 macroadenomas (43.2%). The mean maximum tumor diameter was 11.41±5.2 mm (range: 331 mm). The median pre-operative PRL level was 220.5 mcg/dL (IQR: 151.8365.5). There was a significant correlation between PRL levels and the maximum tumor diameter (rs=0.544, P<0.001) and between PRL levels and patients age (borderline significance: rs=0.273, P=0.073). The median PRL level was 3.4 mcg/dL (IQR: 1.212.9) on postoperative day 1 and 9.4 mcg/dL (IQR: 4.117.4) at month 3. PRL levels were within reference range in 81.8% on 1st day and at month 3, respectively. Mean follow-up was 81.7±34.4 months, with a 72.7% remission rate at the end of follow-up. The rates of secondary hypothyroidism was low at 2.3%, with no cases of hypoadrenalism or permanent diabetes insipidus. Remission rates were higher in microadenomas than macroadenomas (84.0% vs. 57.9%; P=0.054). Low tumor invasiveness (Knosp 01) was associated with a higher remission rate compared to grade 24 tumors (85.7% vs. 22.2%; P<0.001). None of the patients with Knosp grade 34 tumors (n=5) were cured. Plurihormonal adenomas (mostly alpha-subunit immunopositivity) were detected in 7 patients (15.9%). Plurihormonality was associated with lower remission rates at month 3 compared to pure lactotroph tumors (57.1% vs. 90.6%, respectively; P=0.059). Ki-67 expression was ≥3% in 25% of cases. There was no association between Ki-67 expression and tumor diameter or remission rate (P=0.135 and P=0.884, respectively). A logistic regression model showed that the remission rate at month 3 depended mainly on tumor invasiveness (Knosp 2-4) rather than tumor plurihormonality (P=0.002 and P=0.292, respectively).
Conclusion: Surgical treatment of prolactinomas in women is an effective and safe procedure, with success rate depending more on tumor invasiveness (Knosp grades 2-4) than on other parameters, such as tumor size, plurihormonality, or Ki-67 expression.
18 - 21 May 2019
European Society of Endocrinology