Article Text

Download PDFPDF
Success in adrenal venous sampling between two protocols: experience at a tertiary centre
  1. Mohammad Ali Husainy1,
  2. Cheng Fang1,
  3. Ana Nicolescu1,
  4. Dylan Lewis1,
  5. Pauline Kane1,
  6. Royce P Vincent2
  1. 1 Department of Radiology, King's College Hospital NHS Foundation Trust, London, UK
  2. 2 Department of Clinical Biochemistry (Viapath), King's College Hospital NHS Foundation Trust, London, UK
  1. Correspondence to Dr Mohammad Ali Husainy, Department of Radiology, King's College Hospital NHS Foundation Trust, Denmark Hill, London SE5 9RS, UK; m.husainy{at}

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Adrenal venous sampling (AVS) is currently the best available confirmatory test for lateralising aldosterone production in proven primary aldosteronism (PA).1 AVS is an invasive procedure, which carries a small risk of complications, but requires skilled radiologist to obtain biochemically selective samples.2 The technical success rate of AVS has been cited up to 96% in specialised centres,3 whereas unsuccessful AVS is primarily due to failure in cannulating the right adrenal vein, because of its challenging anatomy. Furthermore, in spite of ‘radiologically successful’ cannulation of both adrenal veins, the samples obtained can still be non-diagnostic due to dilution and lack of selectivity.

Following a revision of our AVS protocol (both radiological and biochemical) in 2012, we assessed its impact on the success rate of the procedure. We carried out a retrospective review of all AVS performed in the interventional radiology department since 2008. Prior to 2012 (pre-2012 protocol), three interventional radiologist (IR) performed AVS in randomly allotted slots without cosyntropin stimulation using a size 5 French (Fr) SIM2 Glidecath (Terumo, Somerset, New Jersey) as the preferred catheter for cannulation. As per the revised protocol (post-2012 protocol), two IR (combined experience of >24 years) were designated to perform all AVS procedures and patients were booked as the first case in the morning list. All received cosyntropin bolus intravenously (250 μg in 5 mL of 0.9% saline) an hour before the procedure. The adrenal vein was cannulated with a 4 Fr Cobra 2 (C2) catheter with side and end holes mounted on a 0.035″ Storq (Cordis, Miami Lakes, Florida) wire. A biochemist attended these procedures to cross verify sample collection and handling as per both protocols.

Cannulation of the adrenal vein is deemed successful if the selectivity index (SI) (cortisol adrenal vein:cortisol peripheral vein) is >2 without/>3 with cosyntropin stimulation.4 Furthermore, without cosyntropin, a lateralisation index (LI) ≥2 (cortisol-corrected aldosterone ratio) from high side to low side ((A:Chigh/A:Clow)) is suggestive of unilateral disease. With cosyntropin, a LI of ≥4 is suggestive of unilateral disease, whereas ≤3 indicates bilateral hypersecretion.4

Cortisol concentrations were measured by ADVIA Centaur analyser (Siemens Healthcare Diagnostics, Frimley, UK). The limit of detection (LoD) was 30 nmol/L and coefficient of variation, 4.22%–4.98%. High results were re-run using 1:2 on-board dilutions with further dilutions done manually when required. Aldosterone concentrations were measured by Liaison analyser (DiaSorin, Dartford, UK). The LoD was 50 pmol/L and the assay was linear up to 2770 pmol/L. All results >2770 pmol/L were reanalysed after 1:10 manual dilution using ‘endocrinology diluent (319133)’ provided by the manufacturer.

Fisher's exact test was performed to obtain the p value using SPSS software V.21. A p<0.05 was considered to be significant.

In total, 45 patients were identified through our database system who had 47 episodes of AVS between 2008 and 2015. Median age was 51 (range 31–70) years, out of which 71% were males; 30 out of 47 had the AVS using the new protocol.

Using the post-2012 protocol, the success rate achieved increased to 76.7% (n=23/30) from 29.0% (5/17) and to 93.3% (28/30) from 64.7% (11/17) for the right and left adrenal veins, respectively versus the pre-2012 protocol (both p<0.02). The overall success rate for bilateral cannulation improved to 72% from 19% (p<0.02). One patient who had two failed AVS using the pre-2012 protocol eventually had a successful outcome using the post-2012 protocol.

We reviewed the CT scans of 27 (5 pre-2012 and 22 post-2012) cases with bilateral determinant SI to see if it correlates to the side of adrenal adenoma (tables 1 and 2). Overall, CT and biochemical findings were concordant in 66.7% (18/27) of the time (pre-2012: 40% (2/5); post-2012: 73% (16/22)). In cases where CT demonstrated unilateral disease, the concordance was 25% (1/4) and 83% (15/18) for pre-2012 and post-2012 protocol, respectively.

Table 1

CT findings versus adrenal venous sampling (AVS) using the pre-2012 protocol

Table 2

CT findings versus adrenal venous sampling (AVS) using the post-2012 protocol

A prerequisite to AVS success is operator experience.5 We found in our study, designating two experienced IR and scheduling it as the first case in the morning contributed significantly in improving the success rate of adrenal vein cannulation. We experienced some difficultly in cannulating the right adrenal vein using a 5 Fr SIM2 Glidecath. The discrepancy in the calibre of the catheter and the vein and not having side holes caused the vein to collapse and not yield sufficient sample volumes even though the cannulation was achieved. In majority, cannulation of the right adrenal vein has been successfully overcome by using a 4 Fr C2 catheter with side holes. It facilitates cannulation and allows the sample to be taken through the side holes if the tip is lying against the wall. We note our observation is contrary to a study where they found that using catheters with side hole causes dilution of the adrenal hormones.6

Before 2012, we performed AVS without cosyntropin stimulation. For the 17 procedures performed, the operator was confident in 12 cases that the right adrenal vein has been cannulated. However, the assay demonstrated SI of ≥2 in only 5 (29.4%) patients. This is in concurrence with a study where technical success rate was around 31% without cosyntropin.7 This implies that either the samples were diluted or the background cortisol concentrations in the adrenal veins were too low. Cosyntropin infusion or bolus has been reported to improve AVS success rates by increasing cortisol secretion and therefore the SI.5 However, its effect on aldosterone secretion is unclear and has been even reported to decrease LI. Thus, use of cosyntropin infusion prior to AVS remains an area of debate.

The limitation of our study is that it was a single centre observational study with relatively small cohort. Although the sample size is comparable to the existing published studies, we acknowledge the absence of head-to-head comparison of AVS with and without cosyntropin stimulation.5 ,8 The purpose of our study was to assess how we are performing the AVS procedures and whether change in strategy would prove beneficial to the clinicians as well as the patients. In our experience, close working between radiology and biochemistry departments helps to optimise the AVS procedure.

In conclusion, a combination of designating IR to perform AVS procedures, allowing sufficient dedicated time, a selection of small catheters (with side holes) and cosyntropin stimulation significantly improves the technical success rate. The protocol we have described here can be adopted by centres offering AVS for the diagnosis of primary aldosteronism.


View Abstract


  • Handling editor Tahir Pillay

  • Contributors MAH, PK and RPV equally contributed in design of the study, manuscript writing. CF, DL and AN collected the data and helped in the critical review of the paper.

  • Competing interests None declared.

  • Provenance and peer review Not commissioned; externally peer reviewed.