Rapid Cortisol Assay in Adrenal Vein Sampling

Primary aldosteronism (PA) is a common form of hypertension caused by excess aldosterone secretion: it involves over 11% of the patients referred to specialized hypertension centres [1], about one fifth of those with drug-resistant hypertension [2] and about 6% of the hypertensive patients seen in general practice [3].

PA is held to cause cardiovascular disease in excess of the degree of blood pressure elevation, which translates into a high rate of cardiovascular events. These ominous consequences can be prevented with a timely diagnosis. Once the diagnosis of PA has been made, the decision to proceed further with surgical or medical treatment depends on identification of the PA subtype. In fact, adrenalectomy, can be necessary in up to two thirds of the PA cases [1], and was shown to regress cardiovascular damage and prevent CV events at long-term [4]. Therefore, identification of PA followed by subtyping entails fundamental steps that can be particularly beneficial in some subgroups of patients as those with drug-resistant hypertension, who are at high cardiovascular risk of PA [2].

Adrenalectomy is indicated only once a lateralized aldosterone excess has been identified by adrenal vein sampling (AVS). This is a technically challenging procedure because placing the catheter's tip within the tiny and short right adrenal vein is difficult. Samples are, therefore, often obtained from at or near the orifice of the vein, which can lead to dilution of the adrenal hormones, and thus to non selective studies. Unfortunately, accomplishment bilateral selectivity was disappointingly low, particularly if high cutoff values of the selectivity index are used, as shown by the largest study ever performed worldwide. To overcome this limitation several methods have recently been proposed, including stimulation with cosyntropin or metoclopramide [5-10], use of alternative biomarkers, as for example, plasma metanephrines or androstenedione that have a step-up between adrenal vein and inferior vena cava (IVC) higher than cortisol [11], and the intra-procedural rapid cortisol assay (IRCA). The latter can improve the success rate of AVS, particularly at a stage when radiologists are gaining experience with the procedure. Accordingly, several centres have adopted this practice after its introduction [12] and reported anecdotal successes, but logistic problems have prevented widespread use of IRCA in clinical practice [13].

Recently a kit for the semi-quantitative IRCA has been developed [14], which lends itself to routine clinical use owing to its simplicity, ease, and speed of use. Yet, the advantages of exploiting an IRCA-based strategy for improving the success rate in achieving selectivity of AVS has been proven only in one randomized clinical trial (RCT) carried out in Japan, which involved 7 centres, most of which had no or, a limited experience with AVS [14].

RTCs represent the basis for evidence-based and high-grade class of recommendations in practice guidelines. Therefore, we plan to test the hypothesis that the IRCA during AVS could improve the rate of bilaterally selective AVS studies over that accomplished by a routine AVS protocol even when used at referral centres that routinely perform this diagnostic procedure.

Methods Consecutive patients with biochemical diagnosis of PA seeking surgical cure of PA will be recruited for this study, according to current guidelines [15]. These patients have a compelling indication to AVS before being referred or not for adrenalectomy [9]. The only exclusion criteria will be: a) refusal of the centre to participate in the study; b) refusal of the patient to undergo AVS , c) contraindications to the general anaesthesia that is required for laparoscopic adrenalectomy; d) cortisol co-secreting adenoma. The latter criterion is necessary because, given the cortisol-derived selectivity index (see below) primary endpoint chosen, inclusion of these tumours would introduce an obvious confounding bias on results [16].

Study design The outlay of this prospective randomized two-arm multicentre study, and its primary endpoint are summarized below. Randomization will be performed with a specific algorithm as provided in a the SPSSS statistical package. The study will last until 220 patients will be recruited and randomly assigned to either arm considering the possibility of a 10% dropout rate. Data collection will be performed by means of a specific form created ad hoc for the AVIS Study and modified as required [17]. The assignment to either arm will be coded and the data analysis for the primary end-point will be performed by investigators blindly to arm assignment.

In the IRCA arm, the plasma obtained simultaneously from each adrenal vein and the IVC will be tested with typical cortisol assay and in case of non selective results catheters will be repositioned until selective results will be obtained until a maximum of 3 attempts. In the IRCA-sham group only the IVC plasma will be tested and no catheter's repositioning will be undertaken.

The AVS will be performed according to the protocol recently described with bilaterally simultaneous sampling but without metoclopramide stimulation [5]. The IRCA will be performed using the procedure on-going at each participating centre. This implies using either an in-house made, or the commercially available semi-quantitative kit (Trust Medical Support Co., Ltd. Fukuoka, Japan).

All procedures will be according to good clinical practice and to the Declaration of Helsinki and the study will started after approval of the Ethics Committee of the University of Padua and, if necessary, of the participating centres.

Primary endpoint Selectivity will be determined on both side by using a value of the selectivity index > 2.00 as defined in an Expert Consensus Statement of AVS [9]. The definition of the selectivity index (SI) has been already reported [16]; briefly, this is the ratio between plasma cortisol concentration in each adrenal vein and in the infrarenal inferior vena cava blood. The selectivity index will be determined based on measurement of cortisol in the central laboratory of each center. For quality control purposes aliquots of plasma will be collected in heparina and EDTA for metanephrine and androstenedione measurement in the core lab of the coordinating center.

Sample size calculation and statistical analysis We have calculated that with 100 patients submitted to AVS with a normal procedure (Group A) and 100 patients undergoing AVS with IRCA (Group B), the study will have a 82% power of detecting a significant difference of 18% at a two-sided 0.05 significance level, assuming that the SI in Group A is 67% and in Group B is 85 %.

Discussion Currently randomized clinical trials entail the best methodology to provide high-grade class of recommendation and high level of evidence. Therefore, they represent an inevitable step for introducing novel diagnostic and therapeutic strategies where uncertainties still exist as the subtyping of PA by means of adrenal vein sampling.

In fact, the recent publication of SPARTACUS, a randomized clinical trial in the field of a subtyping of PA, has challenged the Endocrine Society Practice guidelines recommendation of performing AVS in all patients seeking surgical cure of PA, who are reasonable candidate for adrenalectomy in general anaesthesia,according to guidelines. In fact, it showed no outcome differences between an AVS- and a computed tomography- based strategy for PA subtyping [18], thus adding a good deal of fuel to the controversy on whether to submit or not all PA to AVS before referring them for adrenalectomy. Although the study was widely criticized and, moreover was not adequately powered, it is altogether evident that in current clinical practice the diagnostic performance of AVS is far from optimal worldwide as shown in the AVIS Study.

Hence, efforts aimed at proving the usefulness of novel strategies for improving the rate of selective AVS studies entail a much worth effort, which should be undertaken by exploiting the RCT methodology. We therefore expect that studies, as the one herein described, being adequately powered from the statistical standpoint, can provide a definitive answer to the question of the whether IRCA can increase the rate of selectivity of AVS. Specifically, we expect to be able to detect a difference of about 18%, e.g. from 67% to 85%, in the rate of bilateral selectivity between the arms.

To date, only one study using a single commercially available kit for semiquantitative IRCA has exploited use of a randomized design to prove the usefulness of this strategy for improving the success rate of AVS in Japan [14]. Unfortunately, most recruited centres and radiologists were not proficient in performing the procedure, which might explain the remarkable advantage seen with IRCA. Hence, it remains to be demonstrated if use of the IRCA can provide similar advantages in different settings as referral centres endowed with radiologists that are proficient in performing AVS. Once verified, or disproved, this hypothesis will have an impact on widespread adoption or abandoning of the IRCA in clinical practice.

In summary, testing the usefulness of an IRCA for increasing the success rate of AVS in a multi-centre randomized clinical trial as that herein described is expected to fill an important gap of information in the field of subtyping of PA. This increased knowledge will ultimately improve the diagnostic use of AVS, which remain the key test for referring PA patients for adrenalectomy.