(CSEMP071) INVESTIGATION AND MANAGEMENT OF PRIMARY ALDOSTERONISM WITH ADRENAL VEIN SAMPLING AT THE CHU DE QUÉBEC-UNIVERSITÉ LAVAL

Abstract: Investigation and Management of Primary Aldosteronism with Adrenal Vein Sampling at our centre.

Background: Adrenal vein sampling (AVS) is the gold-standard test to assess for lateralization in primary aldosteronism (PA). In case of unsuccessful right adrenal vein (RAV) cannulation, our tertiary care centre uses a multinomial regression model to extrapolate the RAV cannulation results.

Objectives: 1) To determine the proportion of successful RAV cannulation during AVS at our hospital; 2) To evaluate the management and clinical evolution of patients with PA who underwent AVS, including those who were treated based on the multinomial regression model results.

Methods: This retrospective cohort study included patients aged ≥18 years with a diagnosis of PA who underwent AVS between January 2017 and September 2022 at our centre. Epidemiological data at diagnosis were collected, as well as data on AVS cannulation, AVS complications and PA clinical evolution. Successful RAV cannulation was determined by the interventional radiologist and by cortisol ratios. Results are presented as percentage or mean±SD.

Results: 39 patients were included (33,3% women; mean age 51,1±11,5 years; 89,7% with hypertension treated with a mean of 2,5±1,1 antihypertensive drugs; 74,4% with hypokalemia), for a total of 40 AVS procedures. 3 procedures (7,5%) selectively cannulated both adrenal veins. 3 procedures (7,5%) led to complications (subcutaneous hematoma). Of the 33 incomplete AVS procedures, the clinical management after 6 of these procedures was unavailable due to follow-up in other hospital centres. Among the remaining 27 patients, 5 (18,5%) had a second complete AVS in another tertiary care centre, 12 (44,4%) were medically treated, 1 (3,7%) had an adrenalectomy and 11 (40,7%) were treated according to the multinomial regression model results. The model lateralized aldosterone secretion in 5 patients, who then all had adrenalectomy, after which 4 were diagnosed adrenal adenomas and one was diagnosed adrenal hyperplasia on pathology. The 8 complete AVS procedures led to 6 diagnoses of unilateral disease and 2 diagnoses of bilateral disease. Applying the multinominal regression model to the initial AVS procedure of these 8 cases would have correctly predict aldosterone lateralization in 7 cases (87,5%).

Conclusion: AVS success rate in our tertiary care centre was low. The multinominal regression model appeared to adequately detect unilateral disease in our small sample. It is necessary to reevaluate our AVS technical protocol to increase our AVS successful cannulation rate.