Endocrine Abstracts

Background: Prolactinomas (PRL) are pituitary adenomas mainly characterized by hyperprolactinemia. In addition to the endocrine effects of prolactin, metabolic alterations have been described in PRL patients. Changes in inflammatory parameters have recently been identified in pituitary adenoma patients. Since both, metabolic (MM) and inflammatory markers (IM) showed promising results in characterization/prognosis of tumor patients, it is tempting to speculate whether those might be useful also in PRL patients.

Patients, methods, results: In this retrospective analysis of medically treated PRL, 23 patients with microprolactinoma (56% women, mean age 34.5) and 30 patients with macroprolactinoma (37% women, mean age 40.7) were included. No difference between micro- and macroprolactinoma patients were evident regarding age, sex, and rate of obesity, hypertension and prediabetes/diabetes mellitus. At baseline, macroprolactinoma patients presented with higher heart rate (HR) (P=0.005) and higher likelihood of thyrotropic- (P=0.007) and gonadotropic-insufficiency (P=0.006). We found a significant correlation between prolactin and BMI (rs=0.364; P=0.007) as well as PRL-size with HbA1c (rs=0.413; P=0.032), BMI (rs=0.316; P=0.021) and HR (rs=0.284; P=0.050). Considering prolactin or PRL-size separately in a multivariate analysis with BMI, HbA1c and HR a significant positive association persisted between prolactin and HbA1c (P=0.009), whilst HR and BMI were positively associated with each other and independent from prolactin or PRL-size. HbA1c had also a negative correlation with testosterone (rs=-0.478; P=0.038), which was not persistent after including prolactin in the multivariate analysis. No correlation could be identified at baseline between prolactin/PRL-size with the studied IM (Glasgow Prognostic Score, Neutrophil-Platelet-Score, Neutrophile-to-Lymphocyte-Ratio [NLR], Platelet-to-Lymphozyte-Ratio [PLR], Prognostic Nutrition Index, Systemic Immune Inflammation Index) and other studied MM (LDL- and HDL-cholesterol, triglyceride, blood pressure [BP]). An association between NLR and fT4 (rs=0.329; P=0.038) as well as LDL (rs=-0.617; P=0.014) was identified, which did not persist in a multivariate analysis considering both variables together and PRL-size. A correlation between fT4 and PLR (rs=0.351; P=0.026) could not be confirmed in the multivariate analyses with prolactin level and PRL-size, separately. In 47 patients complete follow-up data (median follow-up time 17 months, interval 2–141 months) were available. Cabergoline dosage required to achieve normoprolactinemia correlated with baseline LDL (rs=0.493, P=0.052), systolic (rs=0.341; P=0.024) and diastolic (rs=0.324; P=0.032) BP as well as baseline testosterone (rs=-0.447; P=0.019). Tumor shrinkage correlated with LDL at baseline (rs=0.570; P=0.033). Only systolic and diastolic BP remained predictive for Cabergoline dosage required to achieve normoprolactinemia in the regression analysis.

Conclusion: Metabolic but not inflammatory markers might be related with initial presentation and outcome in PRL.