Clinical Research

DepartmenUniversity, Co

CorrespondHospital, ColuFort Washingjes9061@nyp.o

Ann Vasc Surghttp://dx.doi.or� 2014 Elsevi

Manuscript re

2013; publishe

The Vascular Surgeon’s Experience withAdrenal Venous Sampling for the Diagnosisof Primary Hyperaldosteronism

Jeffrey J. Siracuse, Heather L. Gill, Irene Epelboym, Noelle C. Clarke, Nii-Kabu Kabutey,

In-Kyong Kim, James A. Lee, and Nicholas J. Morrissey, New York, New York

Background: Adrenal venous sampling (AVS) is used to distinguish between bilateral idio-pathic hyperplasia and a functional adrenal tumor in patients with hyperaldosteronism. Success-ful sampling from both adrenal veins is necessary for lateralization and may require more than 1procedure. AVS has traditionally been performed by interventional radiologists; however, ourgoal was to examine the outcomes when performed by a vascular surgeon.Methods: All patients with a diagnosis of hyperaldosteronism were referred for AVS regardlessof imaging findings. Cortisol and aldosterone levels were measured in blood samples from bothadrenal veins. Postoperative analysis of intraoperative laboratory values before and after cosyn-tropin administration determined successful cannulation and sampling of each vein.Results: Between 2007 and 2012, 53 patients underwent AVS by one vascular surgeon. Theaverage age was 54 and 63% were men. Our success rate increased with experience, becauseduring the earlier years (2007e2010) primary and secondary success rates were 58% and 68%,respectively compared with later years (2011e2012) when primary and secondary success rateswere 82% and 95%, respectively (P < 0.05). Results of AVS altered localization of diseasecompared with what had been anticipated based on preoperative imaging and thus influencedsurgical decision making in 47% of cases.Conclusions: AVS is an important procedure in the work up of hyperaldosteronism to helpidentify and localize metabolically active tumors. It is an additional area in medicine where avascular surgeon can lend expertise. Success with the procedure improves with experienceand should be performed by high volume surgeons.

INTRODUCTION

Primary hyperaldosteronism results from abnor-

mally high secretion of aldosterone that is not sup-

pressed by the renineangiotensin pathway and is

t of Surgery, New York-Presbyterian Hospital, Columbiallege of Physicians and Surgeons, New York, NY.

ence to: Jeffrey J. Siracuse, MD, New York-Presbyterianmbia University, College of Physicians and Surgeons, 161ton Avenue, New York, NY 10032, USA; E-mail:rg

2014; -: 1–5g/10.1016/j.avsg.2013.10.009er Inc. All rights reserved.

ceived: August 1, 2013; manuscript accepted: October 9,

d online: ---.

generally caused by an aldosterone-secreting adrenal

adenoma, bilateral adrenal hyperplasia, or unilateral

gland hyperplasia in rare cases.1 It is now recognized

as the most common cause of secondary hyperten-

sion and can eventually lead to severe hypokalemia

if left untreated.1,2 Therefore, detection of primary

hyperaldosteronism as a cause of hypertension is

particularly important when the patients with pri-

mary hyperaldosteronism are at increased risks for

cardiovascular events, independent of their elevated

blood pressure.3

Traditionally,whenprimaryhyperaldosteronismis

suspectedclinically, the initial screening test is periph-

eral venous aldosteroneerenin ratio. If this is found to

be elevated, cross-sectional abdominal imaging (usu-

ally computed tomography [CT]) is performed in an

1

2 Siracuse et al. Annals of Vascular Surgery

attempt to identify a functional adenoma.2 However,

imaging modalities may not be sufficiently sensitive

or specific todrivemanagementdecisions alone: small

adenomas can be missed and incidental nonfunc-

tional adenomas may be falsely implicated as the

cause of symptoms in patients with bilateral adrenal

hyperplasia.2 Defining the etiology and determining

the laterality of disease are therefore essential in this

setting, as treatment differs substantially; bilateral dis-

ease is managed pharmacologically, whereas patients

with unilateral disease are offered surgical resec-

tion.2,4 Surgical resection can often be done laparos-

copically and is successful in resolving symptoms in

manyof the cases, thusmaking accurate lateralization

of the aldosterone secretion, crucial.5 Adrenal venous

sampling (AVS) is an adjunct diagnostic modality to

cross-sectional imaging; in some scenarios, it can be

more specific and sensitive than for distinguishing

the etiology of primary hyperaldosteronism and ulti-

mately help guide treatment.5e7

AVS can be technically challenging, and often the

rate-limiting step to a successful procedure is the

cannulation and biochemical sampling of the right

adrenal vein. The left adrenal vein exits the superior

surface of the left renal vein near its midportion,

making the cannulation fairly straight forward. In

contrast, the origin of the right adrenal vein is

directly off the inferior vena cava in a posterior

lateral direction without an intervening medium-

sized vein to guide catheter placement. The right ad-

renal vein often lies between the 11th and 12th ribs,

and this can be used as a guide. Also, the identifica-

tion of the kidney on fluoroscopy can help localize

the approximate location. Generally, the catheter

is placed superiorly to the vein facing the contralat-

eral/left side and then it is posteriorly rotated while

withdrawn to seat the catheter in the right adrenal

vein. Then, a gentle small injection of contrast is

used to confirm location while avoiding possible

hematoma.

Given these technical considerations and the sig-

nificant diagnostic value of AVS, it is essential that

this procedure be performed by a highly skilled

physician to ensure optimal circumstances for a

successful outcome. At many institutions, AVS is

performed by interventional radiologists with vary-

ing degrees of success (8e96%). Lowvolume centers

have been shown to have poor results.8,9 At

Columbia University Medical Center, the vascular

surgeons have become the primary referral option

for the endocrinologists and endocrine surgeons.10,11

Interestingly, there is no reported series that de-

scribes the AVS outcomes when performed by

vascular surgeons. Therefore, our specific aim was

to present our experience to demonstrate that our

technical expertise is invaluable in the successful

performance of AVS and that this is an additional

area in medicine where vascular surgeons can lend

expertise.

METHODS

All patients who presented to either the endocrinol-

ogists and/or the endocrine surgeons at Columbia

University Medical Center with suspected primary

hyperaldosteronism were referred to a single

vascular surgeon for AVS regardless of preoperative

imaging. A retrospective review of all cases and out-

comes was performed. Because of diurnal variation,

AVS is always performed as the first procedure of the

day.11 Vascular access is initially obtained through

the right femoral vein. Baseline aldosterone and

cortisol levels are measured in the inferior vena

cava.We used a SIM 1 (CookMedical, Bloomington,

IN) for the right adrenal vein and a SIM 2 (Cook

Medical) catheter for the left adrenal vein. In cases

where we are unsuccessful with these catheters, a

Mickelson catheter (Cook Medical) can be used for

the right side and a 4-French straight glide catheter

for the left side. We use a floppy .03500 guidewire as

our standard wire for this procedure. We confirmed

cannulation with a limited angiogram with a small

amount of contrast. Baseline cortisol and aldosterone

levels were sent from each vein. Gravity drainage

was used to avoid collapsing the vein. Cosyntropin

(250 mg) was administered as a bolus via peripheral

injection over 3e4 minutes. The left and right adre-

nal veins were then resampled. Postoperative anal-

ysis of intraoperative laboratory values before and

after cosyntropin administration determines success-

ful cannulation and sampling of each vein. The ratio

of cortisol concentrations in the adrenal vein to the

inferior vena cava is referred to as the ‘‘cortisol

step-up’’ or the ‘‘sensitivity index’’ (SI) and the

A/C ratios from the adrenal vein samples, higher

(dominant) over lower (nondominant) is called the

‘‘A/C gradient’’ or the ‘‘lateralization index’’ (LI).

As per accepted criteria at most major centers, we

use an SI of >3:1 prestimulation and >5:1 poststim-

ulation to determine successful cannulation. An LI of

>4:1 is necessary to establish laterality, which, if pre-

sent, guides the management algorithm toward sur-

gical resection of a presumed metabolically active

tumor.5,10 Success was primary if there was success-

ful cannulation for the first time and secondary if

an additional procedure was required. We divided

the patients into 2 time periods to determine the

improvement in our success rates with time: 2007e2010 and 2011e2012.

Fig. 1. Primary and secondary success rates of adrenal venous sampling from 2007 to 2010 compared with adrenal

venous sampling from 2011 to 2012.

Vol. -, No. -, - 2014 Surgeon’s experience with adrenal venous sampling 3

RESULTS

Between 2007 and 2012, 53 patients underwent

AVS. In the first time period (2007e2010), there

were 37 patients and there were 16 from 2011e2012. The average age in this cohort was 54 and

63% were men. Preoperative imaging was normal

in 20% of patients, 65% of patients had suspicious

unilateral findings e either thickening or a sus-

pected mass, and 15% had suspicious findings bilat-

erally. All patients were hypertensive, and 67% had

been diagnosed with hypokalemia.

Our success rate increased with experience. Dur-

ing the earlier years (2007e2010), primary and

secondary success rates were 58% and 68%, respec-

tively, compared with later years (2011e2012)

when primary and secondary success rates were

82% and 95% (P < 0.05), respectively (Fig. 1). In

all cases, procedural failure was secondary to

inability to cannulate the right adrenal vein. Results

of AVS altered localization of disease compared with

preoperative imaging, and thus influenced surgical

decision making in 47% of cases (Fig. 2). All the

cases, where the cross-sectional imaging had identi-

fied no adrenal gland abnormalities (normal) and

bilateral abnormalities, had their laterality changed

after AVS. In the group of patients who had preop-

erative imaging suspicious for unilateral disease,

AVSwas discordant 27%of the time: 71%had bilat-

eral hyperaldosteronism and 29% had a lesion

contralateral to the side suggested by imaging. For

all patients with suspected bilateral disease on

imaging who had successful cannulation, AVS

demonstrated unilateral disease. In addition, func-

tional disease was identified by AVS in all patients

with normal preoperative imaging who had a suc-

cessful cannulation: 16% had bilateral and 83%

had unilateral disease.

As of most recent follow-up, 91% of patients who

underwent AVS had an operation. In this group,

88% had an adrenocortical adenoma on final pa-

thology and the remainder nodular hyperplasia.

Ten patients whose final pathology demonstrated

adenoma had preoperative imaging that did not

demonstrate a lateral lesion. All patients who had

resection had resolution of their hypokalemia and

resolution or improvement in hypertension. There

was no postoperative complication from the AVS,

and all patients were discharged home the same day.

DISCUSSION

AVS is an essential procedure in the work up of pri-

mary hyperaldosteronism, because it aids the identi-

fication and localization of metabolically active

tumors with a high degree of sensitivity, specificity,

and accuracy. Our study demonstrates that the re-

sults of AVS altered care in nearly half of the patients

when compared with what would be done based

on preoperative imaging alone. In our hands,

procedure-related complication rate was 0, and we

were able to perform AVS with excellent primary

and secondary success rates in the last 2 years. This

was significantly better than the first part of the

study period (2007e2010). As it is the case with

Fig. 2. Preoperative cross-sectional imaging and subse-

quent alteration of lateralization after adrenal venous

sampling (AVS). Cross-sectional imaging lateralization

in parentheses. All the cases, where the cross-sectional

imaging had identified no adrenal gland abnormalities

(normal) and bilateral abnormalities, had their laterality

changed after AVS.

4 Siracuse et al. Annals of Vascular Surgery

many technically demanding procedures, there is a

clear volume-outcome relationship with respect to

successful cannulation, and thus, we advocate that

it should be performed by a high number of sur-

geons. This is also consistent with previous reports

of low volume centers having success rates as low

as 8%.8 Our primary and overall successful bilateral

adrenal vein cannulation in the later period is com-

parable with that reported by higher volume inter-

ventional radiologists.8,9,11e15

In some instances, where bilateral cannulation

was not performed and which was related to not

cannulating the right adrenal vein in all cases, we

noted the presence of an inferior accessory hepatic

vein (in some cases, as close as 4 mm to the right ad-

renal vein) as a possible culprit; this has been previ-

ously demostrated.16 One possible way to improve

on this would be to have a rapid assay for intraoper-

ative assessment of cortisol levels and thereby have

real-time confirmation of correct catheter position.

Indeed, some groups have demonstrated improve-

ment in right adrenal vein cannulation from 73%

to as high as 97% using this technique.16 We plan

on adopting this method in the near future, because

it would allow us to attempt right-sided cannulation

immediately should we find that initial catheteriza-

tion was unsuccessful, instead of subjecting the pa-

tient to a repeat procedure. The discordant results

between preoperative imaging (CT/magnetic reso-

nance imaging) and the AVS,most likely, are related

to the inability of preoperative imaging to distin-

guish between a functional and nonfunctional

adenoma and to the inability to detect very small

metabolically active tumors.

Our study suggests that AVS can be safely and

successfully performed by an experienced vascular

surgeon; however, it is limited by its retrospective

nature. Patient selection was based on referral by

endocrinologists and endocrine surgeons, which

potentially introduced bias as this population was

more likely to have suspected surgical versus

nonsurgical disease. This, however, does not affect

the probability of technical success and is therefore

less relevant to our final conclusions. All procedures

were performed by a single surgeon, thus poten-

tially limiting the generalizability of our findings.

However, we demonstrated at a high level of signif-

icance that success increases with experience, which

is applicable in a broad practice setting.

AVS is a crucial step in the work up of primary

hyperaldosteronism because it allows the physician

to distinguish between surgically treatable unilat-

eral disease and bilateral disease that should be

managed pharmacologically. As vascular surgeons,

we were able to successfully contribute to the care

of these patients; however, success is dependent

on experience and that should ultimately guide

referral to the specialist performing the procedure.

REFERENCES

1. Carey RM. Primary aldosteronism. J Surg Oncol 2012;106:

575e9.

2. Funder J, Carey R, Fardella C, et al. Case detection, diagnosis,

and treatment of patients with primary aldosteronism: an

Vol. -, No. -, - 2014 Surgeon’s experience with adrenal venous sampling 5

Endocrine Society clinical practice guideline. Clin Endocrinol

Metab 2008;93:3266e81.3. Milliez P, Girerd X, Plouin PF, et al. Evidence for an

increased rate of cardiovascular events in patients with pri-

mary aldosteronism. J Am Coll Cardiol 2005;45:1243e8.4. Nwariaku FE, Miller BS, Auchus R, et al. Primary hyperal-

dosteronism: effect of adrenal vein sampling on surgical

outcome. Arch Surg 2006;141:497e502.

5. Waldmann J, Maurer L, Holler J, et al. Outcome of surgery

for primary hyperaldosteronism. World J Surg 2011;35:

2422e7.

6. Kempers MJ, Lenders JW, van Outheusden L, et al. System-

atic review: diagnostic procedures to differentiate unilateral

from bilateral adrenal abnormality in primary aldoste-

ronism. Ann Intern Med 2009;151:329e37. Review.

7. Salem V, Hopkins TG, El-Gayar H, et al. Adrenal venous

sampling as a diagnostic procedure for primary hyperaldos-

teronism: experience from a tertiary referral centre. Hor-

mones (Athens) 2012;11:151e9.

8. Vonend O, Ockenfels N, Gao X, et al., German Conn’s Reg-

istry. Adrenal venous sampling: evaluation of the German

Conn’s registry. Hypertension 2011;57:990e5.

9. Young WF, Stanson AW. What are the keys to successful

adrenal venous sampling (AVS) in patients with primary

aldosteronism? Clin Endocrinol (Oxf) 2009;70:14e7.

10. Burshteyn M, Cohen DL, Fraker DL, et al. Adrenal venous

sampling for primaryhyperaldosteronism inpatientswith con-

current chronic kidney disease. J Vasc Interv Radiol 2013;24:

726e33. http://dx.doi.org/10.1016/j.jvir.2013.01.496.

11. Harvey A, Pasieka JL, Kline G, et al. Modification of the pro-

tocol for selective adrenal venous sampling results in both a

significant increase in the accuracy and necessity of the pro-

cedure in the management of patients with primary hyper-

aldosteronism. Surgery 2012;152:643e9. http://dx.doi.org/10.1016/j.surg.2012.07.007. discussion 649-51.

12. Young WF, Stanson AW, Thompson GB, et al. Role for adre-

nal venous sampling in primary aldosteronism. Surgery

2004;136:1227e35.13. Oh EM, Lee KE, Yoon K, et al. Value of adrenal venous sam-

pling for lesion localization in primary aldosteronism. World

J Surg 2012;36:2522e7.14. Trerotola SO, Smoger DL, Cohen DL, et al. The inferior

accessory hepatic vein: an anatomic landmark in adrenal

vein sampling. J Vasc Interv Radiol 2011;22:1306e11.

15. Auchus RJ, Michaelis C, Wians FH Jr, et al. Rapid cortisol as-

says improve the success rate of adrenal vein sampling for

primary aldosteronism. Ann Surg 2009;249:318e21.

16. Iacobone M, Citton M, Viel G, et al. Unilateral adrenal

hyperplasia: a novel cause of surgically correctable primary

hyperaldosteronism. Surgery 2012;152:1248e55.