Bilateral Inferior Petrosal Sinus Sampling Without Lateralization Is Less Accurate for the Diagnosis of Cushing Disease

Abstract

Context: During bilateral inferior petrosal sinus sampling (BIPSS), the side-to-side adrenocorticotropic hormone (ACTH) ratio, referred to as sampling lateralization, was used to predict pituitary adenoma localization.

Objective: To investigate the potential different diagnostic accuracy of BIPSS for differentiating Cushing disease (CD) and ectopic ACTH secretory syndrome (EAS) patients with low lateralization (inferior petrosal sinus [IPS]:IPS ≤ 1.4) and high lateralization (IPS:IPS > 1.4).

Methods: This single-center retrospective study (2011-2021) included (all patients had BIPSS results and confirmed pathologic diagnoses) 220 consecutive CD patients (validation set), 30 EAS patients, and 40 of the CD patients who had digital subtraction angiography (DSA) videos (discovery set).

Results: In the discovery set, the low-lateralization CD group (n = 11) had a higher median plasma ACTH concentration (62.2, IQR 44.7-181.0 ng/L) than the high-lateralization CD group (n = 29) (33.0, IQR 18.5-59.5, P = .013). Lower IPS to peripheral ratios were observed in the low-lateralization group during BIPSS, both before and after stimulation (P = .013 and P = .028). The sensitivity of BIPSS before stimulation in differentiating CD from EAS was lower in the low-lateralization group than the high-lateralization group (54.6% vs 93.1%, P = .003), as validated in the validation set. DSA videos revealed higher vascular area difference visible in the 2 sides of the pituitary in low lateralization (median 1.2 × 10⁵ pixels, IQR 0.5-1.8) than the high-lateralization group (0.4 × 10⁵ pixels, IQR 0.1-0.7, P = .008). The vascular area ratio of the 2 sides was also significantly higher in low (1.55, IQR 1.31-2.20) than high lateralization (1.19, IQR 1.07-1.35, P = .010).

Conclusion: Our study suggested that low lateralization in CD patients may reduce the diagnostic sensitivity of BIPSS, which might be potentially associated with peripituitary vascular anatomy.

Keywords: Cushing disease; digital subtraction angiography; ectopic ACTH syndrome; inferior petrosal sinus sampling; pituitary.

Figure 1.

Flowchart of the patient inclusion process incorporating the discovery set and validation set. Abbreviations: ACTH, adrenocorticotropic hormone; CS, Cushing syndrome; CD, Cushing disease; EAS, ectopic ACTH secretory syndrome.

Figure 2.

The discovery set DSA images in the BIPSS operation were analyzed by first converting the dynamic images into static images, followed by circling the peripituitary area (the box) and performing the calculation of the contrasted area at the input side and the contrasted area at the output side of the other box.

Figure 3.

ROC for max unilateral IPS:P before or after stimulation in discovery set. A, the low-lateralization group (n = 11) vs EAS group(n = 30); B, the high-lateralization group (n = 29) vs EAS group(n = 30).

Figure 4.

The reconstructed peripituitary vascular image of (A) the low-lateralization group (n = 11) and (B) the high-lateralization group (n = 29) in the discovery set.