Subtype differentiation in primary hyperaldosteronism: evaluation of alternative algorithms avoiding a controversial gold standard
Katharina Lang1, Stefanie Hahner1, O Vonend2, L C Rump2, M Quninkler3, Sven Diederich3, Katharina Maier1, Felix Beuschlein4, Martin Bidlingmaier4, S Endres4, C Engelke1, Ralph Kickuth1, Martin Fassnacht1, Martin Reincke4 & Bruno Allolio1
Context: Primary aldosteronism (PHA) is the most common cause of secondary hypertension. Distinguishing unilateral (UAH) from bilateral (BAH) and other causes of PHA is one of the greatest challenges in the diagnostic process. Adrenal venous sampling (AVS) is widely considered to be the gold standard for subtype differentiation.
Methods: Technical and diagnostic outcome of AVS was analysed retrospectively in 59 patients with PHA who underwent AVS at the University of Würzburg and was compared to imaging, clinical and laboratory data. As cut off for selective sampling during AVS a cortisol central/peripheral ratio of ≥2 was chosen.
Results: AVS: only 47.5% were bilaterally selective, 25.4% were unilaterally selective but still suitable for subtype differentiation. Of the informative AVS cases (n=43) 83.7% led to the correct diagnosis whereas in 16% the diagnosis remained uncertain or even turned out to be wrong. In 27.1% of cases AVS was not informative due to lack of selective sampling. Imaging: CT (n=42) or MRI (n=17) both revealed only moderate sensitivity (CT 67.8%, MRI 62.5%) and specificity (CT 70%, MRI 80%). Saline infusion test: 11% showed an increase of serum aldosterone (≥10%). Eighty percent of this subgroup had UAH. Aldosteronerenin-ratio at baseline did not differ significantly between UAH and BAH (UAH 120.2±86.7; BAH 83.0±60.2). Posture testing: 65.2% of the patients with UAH and none with BAH showed a fall of aldosterone or unchanged (increase <5%) serum aldosterone leading to a specificity and a positive predictive value of 100%.
Conclusion: AVS as current gold standard for subtype differentiation is invasive, often not informative and may even lead to the wrong diagnosis. In selected cases clinical and laboratory data in combination with CT may allow successful subtype differentiation without AVS. Diagnostic algorithms based on our observations will be evaluated prospectively with the aim to reduce the need of AVS.