ORIGINAL
ARTICLE
Does Contralateral Suppression at Adrenal Venous
Sampling Predict Outcome Following Unilateral
Adrenalectomy for Primary Aldosteronism?
A Retrospective Study
Martin J. Wolley, Richard D. Gordon, Ashraf H. Ahmed, and Michael Stowasser
Endocrine Hypertension Research Centre, University of Queensland School of Medicine, Greenslopes and
Princess Alexandra Hospitals, Brisbane 4102, Australia
Context: In primary aldosteronism (PA), adrenal vein sampling (AVS) distinguishes unilateral and
bilateral disease by comparison of aldosterone/cortisol (A/F) ratios. There is controversy about the
criteria for interpretation, however, and in particular it is not clear whether contralateral suppression (CS) (defined as A/Fadrenal ⱕ A/Fperipheral on the unaffected side) is important. We therefore
performed a retrospective study to determine whether CS in surgically treated unilateral PA was
associated with blood pressure (BP) and biochemical outcomes.
Setting and Design: Patients who underwent unilateral adrenalectomy for PA after successful AVS
were included if the lateralization index (A/Fdominant:A/Fnondominant) was ⱖ2. Cases were reviewed
at 6 to 24 months follow-up for outcomes with respect to the presence and degree of CS.
Results: Sixty-six of 80 patients had CS. Baseline characteristics were similar. At postoperative
follow-up, those with CS had lower systolic BP (SBP) (128 mm Hg vs 144 mm Hg, P ⫽ .001), a greater
proportion with cure or improvement of hypertension (96% vs 64%, P ⫽ .0034), a greater proportion with biochemical cure of PA on fludrocortisone suppression testing (43 of 49 [88%] vs 4 of
9 [44%], P ⫽ .002) and were taking a lower median number of antihypertensive medications (0 vs
1.5, P ⫽ .0032). In a multivariate model, the degree of CS and preoperative SBP were both significantly correlated with postoperative SBP, but the lateralization index, sex, and age were not.
Conclusion: In this study, the presence of CS correlated with good BP and biochemical outcomes
from surgery. This finding suggests that CS should be a factor in deciding whether to offer surgery
for treatment of PA. (J Clin Endocrinol Metab 100: 1477–1484, 2015)
n primary aldosteronism (PA), adrenal vein sampling
(AVS) is recognized as the most reliable method of distinguishing unilateral from bilateral disease (1, 2). Making
this distinction is vital for treatment decisions, because
unilateral disease typically responds well to surgical treatment whereas bilateral disease is usually medically treated
(1).
AVS is not a standardized procedure around the world,
however, with various units using different protocols both
for the procedure itself and in the interpretation of results
(2, 3). Typically, successful cannulation of the adrenal
I
veins is confirmed by adequate adrenal vein/peripheral
vein cortisol ratios (the selectivity index) (1, 4, 5). Aldosterone/cortisol (A/F) ratios are then calculated to
correct aldosterone concentrations for dilution from
nonadrenal blood. A/F ratios are compared either between left and right or with peripheral levels or both to
decide whether the disease is bilateral or unilateral. Recent
consensus guidelines suggest using a lateralization index
(LI) (A/Fdominant:A/Fnondominant) of ⱖ4 if ACTH-stimulated AVS or ⱖ2 if non-ACTH-stimulated AVS values are
used as cutoff values for unilateral disease, with contralat-
ISSN Print 0021-972X ISSN Online 1945-7197
Printed in U.S.A.
Copyright © 2015 by the Endocrine Society
Received September 29, 2014. Accepted January 22, 2015.
First Published Online January 30, 2015
Abbreviations: A/F, aldosterone/cortisol; APA, aldosterone-producing adenoma; AUC,
area under the curve; AVS, adrenal vein sampling; BP, blood pressure; CI, confidence
interval; CT, computed tomography; FST, fludrocortisone suppression test; LV, left ventricular; ROC, receiver operating characteristic; SBP, systolic blood pressure.
doi: 10.1210/jc.2014-3676
J Clin Endocrinol Metab, April 2015, 100(4):1477–1484
jcem.endojournals.org
1477
1478
Wolley et al
Contralateral Suppression in AVS
eral suppression (the A/F ratio on the nondominant side
being lower than peripheral) being used as an optional
additional criterion (5).
Our longstanding protocol (for non-ACTH-stimulated
AVS) has been to accept a ⱖ2-fold ratio between adrenal
and peripheral A/F ratios on one side with the presence of
contralateral suppression as an indication of unilateral
disease, and surgery is generally recommended on these
grounds (4). In some cases, other factors such as a high LI
without complete contralateral suppression or other patient factors may also influence the decision to offer surgery (6). Notably, the degree of contralateral suppression
is variable, and it is not clear whether the presence or
degree of contralateral suppression influences blood pressure (BP) outcomes from surgery. We therefore performed
a retrospective study examining the outcomes from adrenalectomy for unilateral PA, with respect to the presence
and degree of contralateral suppression in an attempt to
answer this question.
J Clin Endocrinol Metab, April 2015, 100(4):1477–1484
excluding an acute increase in ACTH preventing suppression of
aldosterone.
AVS
AVS was performed between 8:00 and 11:00 AM after overnight recumbency and without ACTH stimulation. Sequential
cannulation of both adrenal veins was performed by 3 experienced radiologists. Usually a short time (⬍15 minutes) elapsed
between sampling of the left and right adrenal veins. Gradients
of at least 3 between adrenal and peripheral venous cortisol concentrations were taken to indicate adequate sampling. If the A/F
ratio on one side was at least 2 times the simultaneously collected
peripheral ratio and on the other side was the same as or less than
the peripheral ratio (contralateral suppression), the study was
considered to demonstrate lateralization of aldosterone production. Adrenal venous A/F ratios higher than peripheral ratios on
both sides were taken to indicate bilateral aldosterone production even if the LI was ⬎2. When results of AVS were inconclusive, repeat AVS was offered to the patient. Computed tomography (CT) scanning of the adrenal glands was always performed
before AVS to identify any adrenal lesions and to localize adrenal
veins and thus assist in successful cannulation.
Outcomes
Subjects and Methods
Setting and subjects
This study was performed in the Hypertension Unit of the
Princess Alexandra Hospital, Brisbane, Australia. Ethical approval for the study was granted by the relevant authorities. A
prospectively updated database was used to identify patients
who underwent adrenalectomy for PA between 2000 and 2014
(inclusive). Inclusion criteria included successful AVS (bilateral
selectivity index of ⱖ3), LI of ⱖ2, and at least 6 months of
follow-up. BP outcomes were taken from the 6- to 24-month
follow-up period on the basis that after this time period patients
with persistent hypertension postadrenalectomy are likely to undergo more aggressive BP medication titration. Postoperative
echocardiographic and urine albumin/creatinine ratio results
were taken from the period of 6 to 24 months postsurgery.
Diagnosis of PA
The diagnosis of PA was established according to our previously published criteria (4, 7). Interfering medications (diuretics,
␤-blockers, angiotensin-converting enzyme inhibitors, angiotensin II receptor antagonists, and dihydropyridine calcium channel
antagonists) were stopped, if possible, at least 4 weeks before
testing (6 weeks in the case of diuretics). Patients with an elevated
aldosterone/renin ratio on at least 2 occasions (⬎70 with plasma
aldosterone in picomoles per liter and plasma active renin in
milliunits per liter) underwent a fludrocortisone suppression test
(FST) to confirm PA. FST results were regarded as positive if the
aldosterone level at 1000 hours after at least 2 hours of upright
posture was ⱖ165 pmol/L after 4 days of oral salt loading with
concurrent fludrocortisone administration (0.1 mg every 6
hours), providing that upright renin was suppressed to ⬍8.4
mU/L, plasma potassium was within the normal range (sufficient
oral potassium chloride for 6 hours to keep the plasma potassium
level measured 3 times daily as close as possible to 4.0 mmol/L),
and plasma cortisol was lower at 10:00 AM than at 8:00 AM,
Patients were assessed for BP and biochemical outcomes between 6 and 12 months of regular postoperative follow-up.
Seated BP was recorded at clinic visits using the BPTru automated oscillometric monitor (VSM MedTech). After patients
had been seated alone in a room for 5 minutes, 4 readings were
taken; the first was discarded, and the last 3 seated blood pressures were averaged.
Hypertension was defined as “cured” if patients were normotensive (BP of ⱕ140/90 mm Hg) without taking antihypertensive medications and “improved” if fewer medications were
needed to maintain or decrease the baseline BP (provided the
dosage of none was increased) or if the dosage of one or more was
at a reduced level (provided no additional medications were
used). Patients were regarded as “not improved” if more or the
same number of antihypertensive medications were needed to
maintain similar BP or the patient’s BP was higher than at
baseline.
Biochemical outcome was assessed by a postoperative FST
(8) usually performed between 3 and 6 months after surgery.
Biochemical cure was defined as negative postoperative FST
results.
Statistical methods
Statistical significance was assigned at a P value of ⬍.05.
Comparisons between groups were made using the Wilcoxon
rank sum test or the Fisher exact test where appropriate. In addition, corrected P values were calculated using the method of
Benjamini, Hochberg, and Yekutieli to allow for multiple comparisons (9).
For outcome in terms of postoperative BP, we constructed a
multivariate linear model. Systolic blood pressure (SBP) was
modeled as the dependent variable, and contralateral suppression, the A/F ratio of the dominant adrenal gland, preoperative
BP, the LI, the presence of hypokalemia, the presence of a lesion
on CT scanning, sex, age, and the presence of comorbidities were
included as covariates in the initial model. Continuous variables
that were not normally distributed were log transformed, and
doi: 10.1210/jc.2014-3676
jcem.endojournals.org
1479
To review the sensitivity and specificity of contralateral suppression and the
LI in predicting biochemical cure of PA
and BP outcomes, we constructed receiver operating characteristic (ROC)
analyses. Confidence intervals (CIs) for
the area under the curve (AUC) were
computed with 2000 stratified bootstrap
replicates.
Statistical analysis was performed in
R (version 3.1.0; R Foundation for Statistical Computing) with testing of the
linear model using the MASS package
(10) and ROC analysis using the pROC
package (11).
Results
Figure 1. Patient disposition.
normality was confirmed by the Shapiro-Wilk test. Covariates
were removed from the multivariate model in backward stepwise
fashion using the exact Akaike information criterion method to
arrive at the final model.
Table 1.
Patient characteristics before
surgery
A total of 113 records were reviewed; of these, 80 patients met all
inclusion criteria including having
an LI of ⱖ2 (Figure 1). The summary
characteristics of the 2 subgroups
showing or not showing contralateral suppression on AVS are compared in Table 1. Of the
80 patients available for analysis, 66 had contralateral
suppression. Compared with those without contralateral
Baseline (Preadrenalectomy) Patient Characteristics
P Value
No. of patients
Male, no. (%)
Age, y
BMI, kg/m2
Diabetic, no. (%)
Chronic kidney disease, no. (%)
Obstructive sleep apnea, no. (%)
Hypokalemic (⬍3.5), no. (%)
Definite adrenal adenoma on CT, no. (%)
Definite adrenal adenoma on histology, no. (%)
Systolic BP, mm Hg
Diastolic BP, mm Hg
No. of antihypertensive drugs
Lateralization index
A/Fadrenal:A/Fperipheral
Dominant
Nondominant
LV mass index
No.
Value, g/m2
Urinary ACR
No.
Value, mg/mmol
Contralateral
Suppression
No Contralateral
Suppression
Unadjusted
Adjusted
66
30 (45.5)
50.7 (16 –75)
29 (21– 45)
5 (7.6)
3 (4.5)
6 (9.1)
51 (77)
46 (70)
57 (86)
145 (119 –190)
92 (63–123)
3 (0 – 6)
14.5 (2.27–190)
14
6 (43)
57.6 (40 –73)
31.9 (20 –36)
2 (14.3)
2 (14.3)
2 (14.3)
9 (64)
5 (36)
8 (57)
153 (120 –180)
90 (76 –113)
3 (2–5)
8.49 (2.14 –16.3)
.90
.24
.86
.60
.28
.62
.32
.029
.02
.36
.59
.14
.0065
.90
.51
.90
.74
.56
.74
.59
.087
.07
.59
.74
.35
.028
6.04 (1.3–33)
0.36 (0.07– 0.99)
12.6 (3.1–100)
1.34 (1.01– 6.96)
.021
⬍.0005
.07
.015
35
111 (73–162)
7
115 (61–152)
.71
.79
29
1.9 (0.22–12.3)
5
2.1 (1.1– 47.9)
.44
.60
Abbreviations: ACR, albumin/creatinine ratio; BMI, body mass index; LVMI, left ventricular mass index (left ventricular mass indexed to body
surface area). Group data are shown as median (range) unless otherwise stated.
1480
Wolley et al
Contralateral Suppression in AVS
suppression, more patients with contralateral suppression
had evidence of an adenoma on a CT scan or by histology,
but a similar proportion were hypokalemic. Preoperative
BP, left ventricular (LV) mass index (when available), and
the albumin/creatinine ratio (when available) were similar
for those with and without contralateral suppression. The
LI was significantly higher in the group with contralateral
suppression. The A/Fadrenal:A/Fperipheral (nondominant)
(ie, activity of the contralateral gland, defined as suppression if ⬍1) was significantly higher in the patients without
contralateral suppression than in those with contralateral
suppression.
Outcomes
After surgical adrenalectomy, those without contralateral suppression had significantly higher postoperative
SBP with a lesser drop from preoperative levels and required a larger median number of antihypertensive medications than those with contralateral suppression. When
classified by BP outcome, a smaller proportion of the
group without contralateral suppression had cure or improvement of hypertension (Table 2). Preoperative and
postoperative biochemistry data are presented in Table 3.
With use of the same strict criteria as those used to
diagnose PA, postoperative FSTs were performed in 58 of
80 patients to determine whether PA had been biochemTable 2.
J Clin Endocrinol Metab, April 2015, 100(4):1477–1484
ically cured. Clear results obtained in 52 procedures
showed that 88% of patients with contralateral suppression were cured, compared with 44% of those without
contralateral suppression (P ⫽ .002) (Table 2). In the other
6 procedures, the results were not clear because of unsatisfied strict criteria for potassium or cortisol levels on day
4 of the test. A further comparison of BP outcomes was
made between those with (n ⫽ 43) and without (n ⫽ 4)
contralateral suppression for whom a postoperative FST
had demonstrated biochemical cure of PA. Those with
contralateral suppression had a median SBP of 125 mm Hg
with a median of 0 medications, compared with a median
SBP of 141 mm Hg (P ⫽ .067) with a median of 1 medication (P ⫽ .068) in the group without contralateral suppression. These differences did not quite reach statistical
significance, possibly because of the small patient numbers. Diastolic blood pressures were similar (83 mm Hg vs
84 mm Hg, P ⫽ .76).
In those for whom preoperative and postoperative
echocardiogram results were available, neither the median
postoperative LV mass index nor the change compared
with the preoperative value differed significantly between
the 2 groups. In those in whom preoperative and postoperative albumin/creatinine ratios were available, the postoperative urinary albumin/creatinine ratio was signifi-
Postadrenalectomy Outcomes
P value
No. of patients
Follow-up for clinical outcomes, mo
BMI, kg/m2
Systolic BP, mm Hg
Diastolic BP, mm Hg
No. of antihypertensive drugs
Hypertension outcomes, no. (%)
Cured
Improved
Not improved
FST outcomes, no. (%)
No.
Cure
Not cured
Inconclusive
Echocardiographic outcomes
No.
LVMI
Change vs pre-ADX
Post-ADX, mo
Urinalysis outcomes
No.
Urinary ACR
Change vs pre-ADX
Post-ADX, mo
Contralateral
Suppression
No Contralateral
Suppression
Unadjusted
Adjusted
66
12 (6 to 24)
30.7 (19 to 44)
128 (100 –186)
85 (56 to 117)
0 (0 to 4)
14
11.5 (6 to 22)
31.9 (19 to 37)
144 (121 to 170)
89 (75 to 114)
1.5 (0 to 3)
.38
.89
.001
.1
.0032
.9
.89
.015
.27
.017
27 (41)
36 (55)
3 (4.5)
2 (14)
7 (50)
5 (36)
.0034
.017
49
43 (88)
1 (2)
5 (10)
9
4 (44)
4 (44)
1 (12)
0.0021
.016
35
96 (56 to 137)
⫺12 (⫺57 to ⫹ 13)
18 (6 to 24)
7
104 (68 –130)
⫺6 (⫺44 to ⫹ 13)
18 (6 to 24)
.24
.44
.66
.51
.60
.76
29
0.59 (0.07 to 2.1)
⫺1.2 (⫺11 to ⫹ 0.28)
11 (6 to 12)
5
3.4 (0.9 to 13)
⫺0.63 (⫺43 to ⫹ 12)
10 (6 to 12)
.0016
.39
.38
.016
.59
.59
Abbreviations: ACR, albumin/creatinine ratio; ADX, adrenalectomy; BMI, body mass index; LVMI, left ventricular mass index (left ventricular mass
indexed to body surface area).
doi: 10.1210/jc.2014-3676
Table 3.
jcem.endojournals.org
1481
Biochemistry Data Preoperatively and Postoperatively
P Value
No. of patients
Preoperative serum potassium, mmol/L
Postoperative serum potassium, mmol/L
Preoperative aldosterone, pmol/L
Postoperative aldosterone, pmol/L
Preoperative renin, mU/L
Postoperative renin, mU/L
Preoperative aldosterone/renin ratio
Postoperative aldosterone/renin ratio
Contralateral
Suppression
No Contralateral
Suppression
Unadjusted
Adjusted
66
3.3 (2.2– 4.8)
4.0 (3.5– 4.9)
801 (220 –2920)
156 (70 –1330)
4.5 (1.2–29)
13 (1.6 –110)
230 (11–1355)
14.4 (1.5–142)
14
3.5 (2.6 – 4.3)
4.0 (3.5– 4.8)
900 (329 –2950)
278 (70 –700)
5 (2–17)
12.5 (2–34)
185 (19.6 – 433)
25 (3.9 –116)
.42
.70
.97
.028
.041
.54
.19
.064
.67
.8
.97
.164
.164
.72
.38
.17
Group data are shown as medians (range) unless otherwise stated. Note that before 2002, renin was measured as plasma renin activity and has
been converted to the approximate direct renin concentration by multiplying by 8.4.
cantly lower in the group who had shown contralateral
suppression, but the median reduction was not statistically
different (Table 2).
Because a number of publications (12–14) have advocated an LI of ⱖ3 as an indicator of unilateral disease, this
cutoff was arbitrarily applied to the entire cohort and excluded only 1 patient from each group. This made no significant change to the outcome analysis (data not shown)
nor did application of a cutoff LI of ⱖ4, which excluded
4 patients from each group, although the levels of statistical significance decreased (data not shown).
Results of multivariate modeling
For the outcome of postoperative SBP, we constructed
a multivariate linear model. Contralateral suppression
(P ⫽ .0061) and preoperative SBP (P ⫽ .019) were significantly correlated with postoperative SBP, and age was
borderline significant in the model (P ⫽ .1409), but the LI,
sex, the A/F ratio in the dominant adrenal gland, hypokalemia, BMI, and comorbidities were not (multiple R2 ⫽
0.19, P ⫽ .001). In a model comparing the change in SBP
before and after surgery, contralateral suppression (P ⫽
.006) and preoperative SBP (P ⬍ .0005) were again significantly correlated, but other covariates were not (multiple R2 ⫽ 0.48, P ⬍ .0005).
ROC characteristic analysis of LI and contralateral
suppression
We constructed a ROC curve using AVS results as a
predictor of biochemical cure in the 52 patients in whom
a postoperative FST was performed, and a conclusive result was obtained. When the LI was used as a predictor of
biochemical cure at the postoperative FST, the AUC was
61% (95% CI, 37– 85) and the highest combination of
specificity and sensitivity was obtained at an LI of 17.1
(specificity, 100%; sensitivity, 36.2%). When contralateral suppression (as A/Fadrenal:A/Fperipheral [nondominant]) was examined, the AUC was 92.8% (95% CI, 80 –
100), and the highest combination of specificity and
sensitivity was obtained at a ratio of 1.3 (specificity, 80%;
sensitivity, 95.7%). These results are illustrated graphically in Figure 2.
When AVS results were assessed as a predictor of BP
outcome (assessing cure or improvement of hypertension
vs no improvement in the whole cohort of patients), the
AUC for the LI was 61.8% (95% CI, 46.6 –77) and the
highest combination of specificity and sensitivity was seen
at an LI of 16.4 (specificity, 100%; sensitivity, 43%) (Figure 3). When contralateral suppression was used as a predictor, the AUC was 82% (95% CI 66 –97) and the highest
combination of sensitivity and specificity was obtained at
Figure 2. ROC curves for contralateral suppression (CS; A/Fadrenal:A/
Fperipheral [nondominant]) and LI as predictors of biochemical cure of
primary aldosteronism on postoperative FSTs. The points of maximal
specificity and sensitivity for CS or LI as predictors of biochemical cure
are noted on the graph. Note that all patients have a LI of ⱖ2.
1482
Wolley et al
Contralateral Suppression in AVS
cutoffs of 1.1 (specificity, 62.5%; sensitivity, 88.9%) and
0.6 (specificity, 87.5%; sensitivity, 63.9%) (Figure 3)
Discussion
In this study we found that among patients with apparent
unilateral PA treated with adrenalectomy, patients with
contralateral suppression had better BP outcomes than
those without. Despite having similar baseline characteristics, patients with contralateral suppression had lower
postoperative BPs and required a lesser number of antihypertensive drugs, and a greater proportion had biochemical evidence of cure by a postoperative FST. In a
multivariate model, contralateral suppression and preoperative SBP were significantly correlated with postoperative SBP, whereas the LI, age, sex, and the presence of
comorbidities were not. When compared with the LI,
contralateral suppression was also a better predictor of
biochemical cure on a postoperative FST and cure or
improvement of clinical BPs when assessed by ROC
analysis.
Patients who showed contralateral suppression on AVS
were more likely to have a confirmed adenoma by macroscopic and microscopic examination despite a lower median A/Fadrenal:A/Fperipheral ratio in the dominant gland.
Measures of disease severity such as hypokalemia and LV
mass index were similar in the 2 subgroups.
Figure 3. ROC curves for contralateral suppression (CS; A/Fadrenal:A/
Fperipheral [nondominant]) and the LI as predictors of cure or
improvement of hypertension. The points of maximal specificity and
sensitivity for CS or LI as predictors of cure or improvement of
hypertension are noted on the graph. Note that all patients have an LI
of ⱖ2.
J Clin Endocrinol Metab, April 2015, 100(4):1477–1484
The interpretation of AVS results is to some extent arbitrary and based largely on empirical evidence (3, 12, 15,
16). The first question is adequacy of the sample collected.
In a theoretical ideal AVS, blood would be drawn from
each adrenal vein at the same time and at the same “dilution” by extraadrenal blood to allow direct comparison
of aldosterone concentrations. Because this is not feasible,
correcting for the dilution of aldosterone by nonadrenal
blood using a simultaneously collected cortisol concentration has become standard. An acceptable adrenal to
peripheral cortisol gradient provides some confidence that
the samples are of adrenal vein origin. However, the precise level representing a minimum acceptable gradient is
still debated and varies among centers, largely based on
their past experience. We have found a cortisol gradient of
ⱖ3.0 to be the most satisfactory but do not ignore or
completely discount altogether gradients of ⱖ2.0. Other
investigators have also favored a ratio of ⱖ3.0 (17, 18).
The second question regards how to compare the 2
sides to recognize unilateral overproduction of aldosterone. The response to this has varied widely, depending on
unit preference, and there is no clear evidence to suggest
which method best predicts outcome (1, 5, 17). We have
favored the concept of contralateral suppression, which
rests on our understanding of the usual physiology of the
renin-angiotensin-aldosterone system, assuming that in
the case of unilateral autonomous aldosterone production, suppression of renin will cause the contralateral
“normal” gland to minimize secretion of aldosterone but
not cortisol (19, 20). Adrenal venous effluent from such a
suppressed gland should have a lower aldosterone concentration (corrected for cortisol concentration) than a
sample of peripheral blood (14). A more popular method
relies on calculation of a “lateralization index” (LI), which
relies on a comparison of the aldosterone/cortisol ratios in
the 2 adrenal veins, resulting in the terms “dominant” and
“nondominant” (or “contralateral”) sides. The focus is on
dominance of aldosterone production, rather than the
“normality” of aldosterone production by the unaffected
“normal” gland. The present study suggests that the normality of the contralateral adrenal gland (ability to markedly reduce aldosterone production when renin is suppressed) is very important. The implication is that a
contralateral gland whose aldosterone production is not
suppressed, regardless of any comparison with the other
side, will predictably have residual unsuppressible aldosterone production and a poorer BP outcome from surgery. It may be that if an aldosterone-producing adenoma
(APA) has developed unilaterally, but there is bilaterally
abnormal adrenal tissue, removal of the gland with the
“dominant” APA will leave behind abnormal contralat-
doi: 10.1210/jc.2014-3676
eral adrenal tissue capable of autonomous aldosterone secretion (21).
There have been various recent attempts to analyze
grouped AVS results to determine the best way to identify
unilateral disease. Rossi et al (22) assessed 151 consecutive
AVS results to determine the best values to use for selectivity and lateralization and found that an LI of 2 gave the
most acceptable sensitivity and specificity for identifying
unilateral APA. Follow-up in this study included demonstration of a normal aldosterone/renin ratio and resolution of hypokalemia but not a formal aldosterone suppression test seeking evidence for unsuppressible,
autonomous aldosterone production. Webb et al (18) reviewed AVS results in 108 cases of PA and compared a
variety of lateralization criteria, concluding that ACTHstimulated AVS using an LI of ⬎4 was most accurate in
determining unilateral disease but had limited meaning
because no criteria successfully predicted clinical outcome. In a multicenter examination by Monticone et al
(14) of 234 AVS procedures in patients considered to have
unilateral PA, contralateral suppression was present in
82%, but its presence or absence did not correlate with
outcome, because patients with or without contralateral
suppression experienced a 52% cure rate (14). The 8 separate centers used a variety of protocols to diagnose PA
and unilateral PA, and patient heterogeneity was significant between centers. Patient biochemistry and clinical
characteristics at baseline appear to be similar in the 2
studies (eg, baseline SBP of 152–169 mm Hg, with use of
2.2–3 drugs depending on the country in the Monticone
multicenter study compared with baseline SBP of 150 mm
Hg with use of 3 drugs in this study). Monticone et al
reported that all 93 patients who underwent postoperative
saline suppression testing were cured by this criterion.
This compares with an overall cure rate of 90% (judged by
normal postoperative FST results) in this study. This difference may be explained by the recumbent saline suppression test being a less sensitive test for the diagnosis of
autonomous aldosterone production than the FST with
upright samples on the final day, which was suggested by
a recently published, prospective pilot study of saline suppression test in the seated position (23). Alternatively, it
could reflect underlying differences between patient
populations.
In more than one third of patients in the current study
who lacked contralateral suppression, hypertension was
not improved by adrenalectomy after 6 to 24 months,
suggesting that it is not desirable to strongly recommend
surgery in the absence of contralateral suppression, despite the 64% whose hypertension was indeed cured or
improved. The results of medical treatment of PA are often
satisfactory but occasionally are not, and we have exam-
jcem.endojournals.org
1483
ined and reported on our experience after unilateral adrenalectomy in a carefully selected small group of patients
with bilateral PA (6). In that observational study of 40
patients, 19 had an LI of ⱖ2 but without contralateral
suppression and would fit into the category of unilateral
disease if considered by this criterion alone but not if contralateral suppression were mandatory. Hypertension was
cured or improved by surgery in this subgroup in proportions similar to those of the current study when contralateral suppression was absent. Given the lower rate of improvement in hypertension and biochemical cure of PA in
patients without contralateral suppression, our preference
is to offer surgery to those patients only in highly selected
cases and only after careful and detailed discussion. These
have included patients with (1) other parameters suggesting unilateral PA such as a particularly high LI (⬎5), a
clear-cut mass lesion on CT scanning within the dominant
aldosterone-producing gland, florid biochemical disturbance (eg, with markedly elevated plasma aldosterone
and/or hypokalemia), and/or lack of responsiveness of aldosterone to upright posture, who have indicated a preference for surgery over medical treatment and/or (2) poor
tolerance or suboptimal responses to specific medical
treatment of PA and/or (3) an adrenal mass lesion large
enough to warrant consideration of surgery based on malignant potential.
A strength of the present study is that BP outcomes were
usually supported by postoperative testing for residual autonomous aldosterone production, which is highly likely
to be a more reliable outcome measure for cure of PA than
BP with its many contributing factors and innate variability. One weakness of the present study is its retrospective
and observational nature. Another is that patients without
contralateral suppression who nevertheless went on to
have surgery were a relatively selected group, simply because they did not satisfy this unit’s usual criteria. However, they had similar baseline characteristics and satisfied
the LI criteria used by many groups with large and ongoing
experience. A third weakness is that patients were not
studied both before and after administration of ACTH,
which could, with advantage, be addressed in a future
prospective study. At present, firm evidence on whether
stimulation of aldosterone production in the contralateral, normal adrenal might reduce or abolish contralateral
suppression and adversely affect the LI is lacking, with
conflicting reports (14, 22, 24, 25). Because of relatively
small numbers, the ROC curve analysis of the present
study results can only be regarded as suggestive and requires confirmation. It is reasonable to suggest that contralateral suppression is more likely to occur and to be
more pronounced in patients with greater degrees of autonomous unilateral aldosterone production and with
1484
Wolley et al
Contralateral Suppression in AVS
more prolonged hyperaldosteronism, which might predispose at least some of them to residual hypertension after
surgery. Cure or improvement of hypertension after surgery for APA might depend not only on accurate lateralization of hyperaldosteronism but also on other factors
such as age, duration of hypertension, degree of arteriosclerotic disease, and kidney function. Therefore, prediction of improvement of postoperative BP in individual
patients not only requires the assessment of contralateral
suppression at AVS but also must take into account multiple other factors that can influence BP outcomes.
In conclusion, although the most accurate way to interpret AVS results remains uncertain, the presence of contralateral suppression of aldosterone production correlated well in the present study with both good BP and
biochemical outcomes. Although this finding requires
confirmation, preferably in a prospective study, it suggests
that contralateral suppression should be a factor in the
decision to offer surgery for apparently unilateral PA.
J Clin Endocrinol Metab, April 2015, 100(4):1477–1484
7.
8.
9.
10.
11.
12.
13.
14.
15.
Acknowledgments
Address all correspondence and requests for reprints to:
Michael Stowasser, Endocrine Hypertension Research Centre,
University of Queensland School of Medicine, Princess Alexandra Hospital, Ipswich Road, Woolloongabba, Brisbane 4102,
Australia. E-mail: [email protected].
M.W. is supported by a postgraduate scholarship from the
Princess Alexandra Hospital Research Foundation.
Disclosure Summary: The authors have nothing to disclose.
References
1. Funder JW, Carey RM, Fardella C, et al. Case detection, diagnosis,
and treatment of patients with primary aldosteronism: an endocrine
society clinical practice guideline. J Clin Endocrinol Metab. 2008;
93:3266 –3281.
2. Kempers MJE, Lenders JW, van Outheusden L, et al. Systematic
review: diagnostic procedures to differentiate unilateral from bilateral adrenal abnormality in primary aldosteronism. Ann Intern
Med. 2009;151:329 –371.
3. Rossi GP, Barisa M, Allolio B, et al. The Adrenal Vein Sampling
International Study (AVIS) for identifying the major subtypes of
primary aldosteronism. J Clin Endocrinol Metab. 2012;97:1606 –
1614.
4. Stowasser M, Taylor PJ, Pimenta E, Ahmed AH, Gordon RD. Laboratory investigation of primary aldosteronism. Clin Biochem Rev.
2010;31:39 –56.
5. Rossi GP, Auchus RJ, Brown M, et al. An expert consensus statement on use of adrenal vein sampling for the subtyping of primary
aldosteronism. Hypertension. 2014;63:151–160.
6. Sukor N, Gordon RD, Ku YK, et al. Role of unilateral adrenalectomy
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
in bilateral primary aldosteronism: a 22-year single center experience. J Clin Endocrinol Metab. 2009;94:2437–2445.
Wolley M, Gordon RD, Pimenta E, et al. Repeating adrenal vein
sampling when neither aldosterone/cortisol ratio exceeds peripheral
yields a high incidence of aldosterone-producing adenoma. J Hypertens. 2013;31:2005–2009.
Rutherford JC, Taylor WL, Stowasser M, Gordon RD. Success of
surgery for primary aldosteronism judged by residual autonomous
aldosterone production. World J Surg. 1998;22:1243–1245.
Benjamini Y, Yekutieli D. The control of the false discovery rate in
multiple testing under dependency. Ann Stat. 2001;29:1165–1181.
Venables WN. Modern Applied Statistics With S. 4th ed. New York,
NY: Springer; 2002.
Robin X, Turck N, Hainard A, Tiberti N, Lisacek F, Sanchez JC,
Müller M. pROC: an open-source package for R and S⫹ to analyze
and compare ROC curves. BMC Bioinformatics. 2011;12:7717.
Monticone S, Viola A, Rossato D, et al. Adrenal vein sampling in
primary aldosteronism: towards a standardised protocol [published
online ahead of print May 12, 2004]. Lancet Diabetes Endocrinol.
Sacks BA, Brook OR, Brennan IM. Adrenal venous sampling: promises and pitfalls. Curr Opin Endocrinol Diabetes Obes. 2013;20:
180 –185.
Monticone S, Satoh F, Viola A, et al. Aldosterone suppression on
contralateral adrenal during adrenal vein sampling does not predict
blood pressure response after adrenalectomy. J Clin Endocrinol
Metab. 2014;99:4158 – 4166.
Stowasser M, Gordon RD, Rutherford JC, Nikwan NZ, Daunt N,
Slater GJ. Diagnosis and management of primary aldosteronism. J
Renin Angiotensin Aldosterone Syst. 2001;2:156 –169.
Vaughan NJ, Jowett TP, Slater JD, et al. The diagnosis of primary
hyperaldosteronism. Lancet. 1981;1:120 –125.
Mulatero P, Bertello C, Rossato D, et al. Roles of clinical criteria,
computed tomography scan, and adrenal vein sampling in differential diagnosis of primary aldosteronism subtypes. J Clin Endocrinol
Metab. 2008;93:1366 –1371.
Webb R, Mathur A, Chang R, et al. What is the best criterion for the
interpretation of adrenal vein sample results in patients with primary
hyperaldosteronism? Ann Surg Oncol. 2012;19:1881–1886.
Shibata H, Ogishima T, Mitani F, et al. Regulation of aldosterone
synthase cytochrome P-450 in rat adrenals by angiotensin II and
potassium. Endocrinology. 1991;128:2534 –2539.
Young DB, Smith MJ, Jackson TE, Scott RE. Multiplicative interaction between angiotensin II and K concentration in stimulation of
aldosterone. Am J Physiol. 1984;247(3 Pt 1):E328 –E335.
Gomez-Sanchez CE, Rossi GP, Fallo F, Mannelli M. Progress in
primary aldosteronism: present challenges and perspectives. Horm
Metab Res. 2010;42:374 –381.
Rossi GP, Pitter G, Bernante P, Motta R, Feltrin G, Miotto D. Adrenal vein sampling for primary aldosteronism: the assessment of
selectivity and lateralization of aldosterone excess baseline and after
adrenocorticotropic hormone (ACTH) stimulation. J Hypertens.
2008;26:989 –997.
Ahmed AH, Cowley D, Wolley M, et al. Seated saline suppression
testing for the diagnosis of primary aldosteronism: a preliminary
study. J Clin Endocrinol Metab. 2014;99:2745–2753.
Nakamura Y, Satoh F, Morimoto R, et al. 18-Oxocortisol measurement in adrenal vein sampling as a biomarker for subclassifying
primary aldosteronism. J Clin Endocrinol Metab. 2011;96:E1272–
E1278.
Carr CE, Cope C, Cohen DL, Fraker DL, Trerotola SO. Comparison
of sequential versus simultaneous methods of adrenal venous sampling. J Vasc Interv Radiol. 2004;15:1245–1250.