Diagnostic Accuracy of Adrenal Iodine-131 6-Beta-Iodomethyl-19-Norcholesterol Scintigraphy for the Subtyping of Primary Aldosteronism.

Abstract

To evaluate the diagnostic accuracy of the 131I-6β-iodomethyl-19-norcholesterol (NP-59) adrenal scintigraphy for the subtyping diagnosis of primary aldosteronism (PA), considering as gold standard for the diagnosis of unilateral PA (UPA), either the results of the adrenal venous sampling (AVS) or the outcome after adrenalectomy. A retrospective multicenter study was performed on PA patients from 14 Spanish tertiary hospitals who underwent NP-59 scintigraphy with an available subtyping diagnosis. Patients were classified as UPA if biochemical cure was achieved after adrenalectomy or/and if an AVS lateralization index > 4 with ACTH stimulation or >2 without ACTH stimulation was observed. Patients were classified as having bilateral PA (BPA) if the AVS lateralization index was ≤4 with ACTH or ≤2 without ACTH stimulation or if there was evidence of bilateral adrenal nodules >1 cm in each adrenal gland detected by CT/MRI. A total of 86 patients with PA were included (70.9% (n = 61) with UPA and 29.1% (n = 25) with BPA). Based on the NP-59 scintigraphy results, 16 patients showed normal suppressed adrenal gland uptake, and in the other 70 cases, PA was considered unilateral in 49 patients (70%) and bilateral in 21 (30%). Based on 59-scintigraphy results, 10.4% of the patients with unilateral uptake had BPA, and 27.3% of the cases with bilateral uptake had UPA. The AUC of the ROC curve of the NP-59 scintigraphy for PA subtyping was 0.812 [0.707-0.916]. Based on the results of the CT/MRI and NP-59 scintigraphy, only 6.7% of the patients with unilateral uptake had BPA, and 24% of the cases with bilateral uptake had UPA. The AUC of the ROC curve of the model combining CT/MRI and 59-scintigraphy results for subtyping PA was 0.869 [0.782-0.957]. The results of NP-59 scintigraphy in association with the information provided by the CT/MRI may be useful for PA subtyping. However, their diagnostic accuracy is only moderate. Therefore, it should be considered a second-line diagnostic tool when AVS is not an option.