Abstract 160: Single-center experience with Bilateral Inferior Petrosal Sinus Sampling (BIPSS) in adrenocorticotropin-dependent Cushing's syndrome

Objective: To calculate the measures of diagnostic accuracy of BIPSS and determine the optimal cut-point values for central to peripheral ACTH ratios to differentiate between Cushing's Disease (CD) and Ectopic Cushing's Syndrome (ECS). Methods: The initial diagnosis of Cushing's Syndrome (CS) was made as per the endocrine society guidelines. Patients with s. ACTH > 15 pg/mL were considered to have ACTH-dependent CS. Baseline ACTH (both the inferior petrosal sinuses and peripheral) values were obtained, and since 2014, additionally stimulated ACTH values (5 and 10-minutes post one unit iv vasopressin) were obtained as standard protocol at our institute. Since the past few years, prolactin corrected ACTH ratios are also being calculated. The highest central to peripheral ACTH gradient > 2 in baseline samples (bC:P), >3 in samples after vasopressin stimulation (sC:P), and >0.8 in PRL-corrected ratios (pC:P) was used as reference value indicative of CD. An inter-petrosal gradient >1.4 was considered to be suggestive of an adenoma located on the side of the petrosal sinus with the higher ACTH concentration. To determine the final etiological diagnosis, histopathology (HP) and immunohistochemistry (IHC), as well as biochemical recovery after surgery were considered. Results: 51 patients with ACTH-dependent CS underwent BIPSS from 2005 to 2020. Four patients with incomplete data/follow-up, one patient with unilateral petrosal sinus catheterization, one patient with Cortisol Resistance Syndrome were excluded from the analysis. Mean age of our cohort was 37.5 ± 15.1 years, with 77.8% females. 39 patients had CD (positive HP-IHC in 34, and remission post-surgery in additional 5 with hyperplasia), while 6 had ectopic CS (one each of bronchial carcinoid, lung NET, thymic NET, pancreatic NET; and two with undiagnosed ectopic source). The measures of diagnostic accuracy are shown in [Table 1]. On ROC analysis, cut-point ratio of 2.55 for bC:P value gave a sensitivity of 89.7% and specificity of 100% (AUC = 0.964); sC:P ratio of 2.95 had 96.3% sensitivity and 100% specificity (AUC = 0.991); 0.93 for pC:P ratio provided a sensitivity of 93.8% and specificity of 100% for diagnosis of CD (AUC = 0.984). 71.4% (10/14) patients with left sided adenoma, and 85% (17/20) patients with right sided adenoma were correctly lateralised on BIPSS. In patients whom adenoma was detected on pMRI, 66.7% (6/9) were correctly lateralised to the left side, and 94.1% (16/17) to the right side.