272 - External Validation of Published RASI Cutoffs for Primary Aldosteronism Lateralization: Experience from a High-Volume Center

Purpose: The Relative Aldosterone Secretion Index (RASI) was proposed to quantify aldosterone hypersecretion during adrenal vein sampling (AVS)^1,2. Rossi et al.³ established diagnostic cutoffs of ≥2.5 for right lateralization and ≤0.96 for left lateralization. This study aimed to externally test these published RASI cutoffs using post-stimulation AVS data from a high-volume tertiary referral center.

Materials and Methods: A retrospective analysis of patients undergoing AVS between 2020 and 2025 was conducted. Only procedures with bilateral selectivity (post-stimulation selectivity index ≥5) were included. RASI was calculated as the left adrenal-to-IVC ratio of aldosterone-to-cortisol ratios. Lateralization was defined by lateralization index ≥4 and categorized as left, right, or bilateral disease⁴. Data were presented as median and interquartile range (IQR). Kruskal-Wallis test was used for comparisons. Diagnostic accuracy of published and institution-specific cutoffs was evaluated by sensitivity, specificity, and area under the curve (AUC).

Results: Data from 304 AVS procedures were collected. Only 290 were included in the final analysis: 86 with left lateralizing, 152 with non-lateralizing, and 52 with right lateralizing disease. Median RASI values differed significantly across groups (p < 0.001): left lateralizing 5.28 (3.59–7.77), non-lateralizing 3.10 (2.19–4.31), and right lateralizing 0.40 (0.17–0.98).

For right lateralization, a cutoff ≤0.96 demonstrated 75% sensitivity and 91% specificity. The optimal institution cutoff (≤1.46) showed higher sensitivity (90%) with high specificity (86%), and an AUC of 0.928. For left lateralization, the Rossi cutoff (≥2.5) showed high sensitivity (87%) but low specificity (33%). The institution-derived cutoff (≥3.89) provided better performance with 71% sensitivity and 70% specificity (AUC=0.748).

Conclusion: External RASI cutoffs demonstrated high specificity for right lateralization but poor specificity for left lateralization. Institution-derived cutoffs showed balanced sensitivity and specificity for both lateralization directions. These findings support the diagnostic utility of RASI and highlight the importance of center-level customization of cutoffs.