This study analyzed 91 prolactinoma patients treated with bromocriptine over 19 years at Songklanagarind Hospital. Among them, 86 had non-cystic prolactinomas, and 5 had cystic prolactinomas. MRI findings, prolactin levels, and predictors of treatment response were thoroughly evaluated, with the following key findings:

• Patient Characteristics: The majority of patients were female (78.0%), with a mean age of 36.7 years (SD 12.4). Dyslipidemia was the most common comorbidity, affecting 10.8% of patients. The study primarily focused on non-cystic prolactinomas.
• Pre-treatment MRI Findings: MRI revealed significant differences between homogeneous and heterogeneous prolactinomas. Heterogeneous prolactinomas were associated with:
• Larger tumor size and volume: Median maximum diameter was 26.4 mm for T1W heterogeneous prolactinomas vs. 6.7 mm for homogeneous ones (p < 0.001).
• Higher prolactin levels: Median prolactin levels were 470 ng/mL for heterogeneous prolactinomas vs. 128.8 ng/mL for homogeneous ones (p < 0.001).
• A greater proportion of macroadenomas (90.2% in T1W heterogeneous vs. 80% microadenomas in homogeneous, p < 0.001).
• Post-treatment MRI Findings:
• Both groups showed similar rates of tumor volume reduction (>50%) after treatment (59.8% in homogeneous vs. 61.3% in heterogeneous groups, p > 0.05).
• Prolactin normalization rates were comparable, with no significant difference between the two groups (51.6% in homogeneous vs. 41.8% in heterogeneous groups, p = 0.515).
• Cystic Changes After Bromocriptine:
• New or increased cystic components were observed in 53.5% of patients, while 39.5% had no or stable cystic changes, and 7% showed a decrease.
• Patients with pre-existing cystic components were more likely to experience changes, with 71.7% showing an increase and 100% of those with decreased cystic components starting with pre-treatment cystic features (p < 0.001).
• Cystic reduction correlated with lower prolactin levels post-treatment (p = 0.018). Figure 2
• Hemorrhagic Changes After Bromocriptine:
• Hemorrhagic changes were significant, with 32.6% of patients developing new or increased hemorrhage, primarily among those with pre-existing hemorrhagic components (82.1%).
• Stable hemorrhagic components were observed in 60.5% of cases, and 7% showed a decrease in hemorrhagic features. Pre-treatment hemorrhagic status strongly predicted post-treatment changes (p < 0.001).
• Predictors of Treatment Response:
• Patients were classified into good response (55.8%) and poor response (44.2%) groups based on a ≥50% tumor size reduction and prolactin normalization.
• Multivariate logistic regression identified key predictors: Figure 3
• Cavernous sinus invasion: Reduced likelihood of favorable response (adjusted OR = 0.05, p = 0.002).
• Optic chiasm invasion: Increased likelihood of favorable response (adjusted OR = 7.89, p = 0.014).
• Sphenoid sinus invasion: Strongest predictor of good response (adjusted OR = 12.67, p = 0.009). 
• The figure 1 showing MRI before and after treatment: Decreased size and volume after treatment in patient having sphenoid and optic chiasm invasion.
• Prolactin Levels and Hemorrhagic Changes:
• No significant correlation between hemorrhagic changes and prolactin levels (p = 0.128), suggesting that hemorrhagic components are less relevant to prolactin normalization.