| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
From the Clinical Research Centers |
Urologic Oncology Branch/National Cancer Institute (J.L.P., M.M.W., W.R., A.A.B., W.M.L.), Hypertension-Endocrine Branch/National Heart, Lung and Blood Institute (J.R.G.), and Department of Radiology/Walter Magnusson Clinical Center/National Institutes of Health (P.L.C., J.L.D.), Bethesda, Maryland 20892-1501; and Department of Pharmacology (J.C.P.), Georgetown University, Washington, DC
Address correspondence and requests for reprints to: John L. Phillips, M.D., Urologic Cancer Institute, National Cancer Institute, Building 10, Room 2B47, Bethesda, Maryland 20892-1501.
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
APA and BAH patients can present with very similar signs and symptoms, serum chemistries, responses to medication, and nodules on computed tomograpy (CT) scans. The treatment of APA in most patients is surgical adrenalectomy, whereas that of BAH is medication and potassium-sparing diuretics (3, 4). A correct diagnosis is, therefore, highly important in the selection of patients for adrenalectomy. Unfortunately, no single test has been identified to fulfill this need. In practice, a number of tests, observations, and imaging modalities are used, each of which adds a subjective level of confidence to clinical judgement.
In general, all patients with hypertension and unexplained hypokalemia (serum K, <3.2 Meq/dL) should be tested for plasma aldosterone concentration (PAC) and PRA. Those patients who have hyperaldosteronism (PAC, >14 ng/dL) and low PRA (<2.0 ng/mL·h) should undergo further evaluation to determine whether there is autonomous PAC production due to either BAH or an aldosteronoma. Bedside testing has been used to evaluate the suppressibility of aldosterone by monitoring the response of plasma cortisol (PFC), aldosterone (PAC), and 18-hydroxycorticosterone (18-OH-B) to provocative testing (morning ambulation or saline infusion) in the sodium-replete state (5, 6). The results are best interpreted if a circadian drop in morning PFC is observed; this suggests that other associated hormonal changes are not the result of ACTH stimulation. Classically, when there is a fall or no change in PFC after 2 h of ambulation, an associated fall in PAC is suggestive of an APA and not BAH (7, 8). A fall in 18-OH-B under similar circumstances may be a more likely occurrence and is also suggestive of APA (6). A resting 18-OH-B level greater than 50 ng/dL has been considered as suggestive of an APA, whereas a PAC value less than 8.5 ng/dL has been used to rule out an APA (9, 10, 11). In practice, many patients do not demonstrate circadian falls in PFC, manifest the classic responses of PAC and 18-OH-B to provocative testing, or have low PAC or 18-OH-B values. Sensitivities and specificities of bedside testing range between 59% and 72% and 54% and 97%, respectively (11, 12).
Evaluation with either CT or magnetic resonance imaging is performed to
assess the presence of small, solitary, unilateral adrenal masses that
are highly suggestive of an APA (13, 14). Such a finding
is not conclusive because 2–5% of the general population have
incidental adrenal masses (15). In addition, patients with
BAH can present with asymmetric adrenal macronodules, whereas some APA
patients have tumors too small to visualize on CT (<5 mm) (Fig. 1
). Sensitivity and specificity of CT in
hyperaldosteronism ranges from 48–58% and 91–92%, respectively
(16). Thus, patients with bilateral nodularity or
normal-appearing adrenal glands on CT are usually referred for adrenal
vein sampling (AVS).
|
This study evaluates our most recent 7-yr experience in the workup of patients with PHA to determine the predictive value of the above tests for discriminating between BAH and APA patients.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Forty-nine patients with hypertension and hypokalemia were referred with the diagnosis of PHA based on elevations of PAC and suppression of PRA. All patients had hypertension greater than 2 yr. We performed endocrine evaluation after cessation of antihypertensive medication at least 2 weeks before admission to the Clinical Center, NIH. Informed consent was obtained, and the patients were thereafter maintained on a metabolic diet containing 109 mEq/day sodium.
Bedside testing
All 49 patients completed bedside testing. After 3 or more days, PAC, PFC, PRA, plasma 18-OH-B, and serum electrolytes were obtained on the morning of study during bed rest (supine) and again after 2 h of ambulation (upright). The tests were repeated again 3 or more days later for all, except one patient. The data presented are the means of the values of the two rounds of testing. A circadian "pattern" was defined as a decrease or no change in upright PFC compared with the supine level. A positive bedside test was defined as a fall, and a negative bedside test as a rise, in a given hormone level with ambulation, respectively.
Imaging
Forty-eight patients underwent CT (5-mm sections) that were read preoperatively by a radiologist (J.L.D. or P.L.C.) blinded to the bedside data. One patient had undergone adrenalectomy elsewhere, and outside preoperative films were reviewed. A positive CT was defined as the presence of a solitary or clearly delineated nodule associated with a normal-appearing contralateral gland. A negative CT was defined as no nodule seen or an equivocal study (bilateral masses, or increased thickness of one adrenal associated with a questionable mass in the contralateral gland).
AVS
Forty-eight patients underwent AVS as described previously (20). Briefly, PAC and PFC levels were obtained from the femoral vein and the right and left adrenal veins before and 15 min after the administration of a 250-pg ACTH bolus and a 0.5-pg ACTH/mL·min saline infusion. The A/C differential was defined as the larger A/C ratio of one side divided by the A/C ratio of the other side.
Diagnosis
A diagnosis of APA was made when there was evidence of unilateral suppression as defined by an A/C differential more than 5:1 before or after ACTH on AVS. If the A/C differential was less than 5:1, suppression was still diagnosed if a contralateral unaffected side A/C was less than peripheral A/C. BAH was diagnosed in the patients who had no suppression on AVS post-ACTH. Surgical specimens were evaluated by a pathologist blinded to the preoperative diagnosis.
Therapy
For patients in whom an APA was diagnosed, an adrenalectomy was performed exclusively via a flank incision (until 1994) or transperitoneal laparoscopically thereafter (21). Those patients with BAH or patients with an APA who were unable to undergo surgery were treated with potassium-sparing diuretics (e.g. amiloride or triamterene), potassium supplementation, and calcium channel blockade (22). Follow-up for the surgical group was at 1 month postoperatively and every 6 months thereafter or every 6 months for the medical group.
Statistics
Interactive web-based statistical programs were all accessed at members.aol.com/johnp71/javastat.html (use Netscape v2 or MS Internet Explorer v3 and above) and included parametric or nonparametric tests where appropriate. Base 2 logarithmic transformation of nonnormal continuous data were performed for uni- and multivariate logistic regression.
Bedside, invasive, and CT imaging test results were first evaluated
with univariate techniques to ascertain significant differences between
APA and BAH patients (significance, P < 0.05). The
bedside tests included supine and upright PAC and 18-OH-B and percent
change in PAC and 18-OH-B with ambulation. The differences in means of
transformed parametric variables were assessed by the matched pairs
signed-rank test or the Mann-Whitney U test, as appropriate.
Continuous data are presented as the means ±
SD, where indicated. Categorical variables were
assessed with two-way 
2 tests or Fisher’s
exact tests as appropriate. Additional categorical variables for
bedside tests were derived to establish potentially predictive cut-off
values: 15 ng/dL for PAC (<15 = 0, >15 = 1), 60 ng/dL for
18-OH-B (<60 = 0, >60 = 1), a 60% increase in PAC with
ambulation (<60 = 0, >60 = 1), and a 80% increase in 18-OH
with ambulation (<80 = 0, >80 = 1).
Significant covariates from the univariate analyses were entered into step-wise multivariate logistic regression models using a diagnosis of BAH vs. APA as the dichotomous outcome.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
From 1991–1998 we evaluated 49 patients (19 females and 30 males) 53.4 ± 9.8 yr of age with potassium values of 2.9 ± 0.3 mEq/dL (normal, 3.5–5.0), suppressed PRA (0.25 ± 0.24 ng/mL·h) after standing 2 h (upright) (normal, >2.0 ng/mL·h), and aldosterone values of 49.2 ± 76.2 ng/dL after bed rest of 2 or more hours (supine) (normal value, <14 ng/dL). The remainder of the serum electrolytes were normal. Four patients had baseline potassium values of 2.6 Meq/dL or lower and required supplementation. Three patients had normal serum potassium during testing.
Bedside testing
As shown in Fig. 2 and Table 1, supine values for PAC and 18-OH-B in
the APA group covered a broad range; in contrast, the BAH group had a
narrower range with values that were more normally distributed. With
ambulation, all eight BAH patients showed an increase in both PAC
(245%) and 18-OH-B (150%) values (Table 1). The APA group had two
clear subpopulations: those who responded to ambulation with expected
falls in PAC (12 patients or 30%) and 18-OH-B (14 patients or 46%),
and those who responded with a rise in PAC (29 patients or 70%) and
18-OH-B (26 patients or 54%) (Fig. 3, A
and B). The large proportion of patients with APA who had negative
tests (e.g. the hormone values rose) is reflected in Table 1. There was no difference in the proportion of APA patients with
negative tests after selecting for patients who had circadian falls in
plasma cortisol (3 with BAH and 21 with an APA; P >
0.05).
|
|
|
We then compared only the patients whose hormone values rose with
ambulation (e.g. all of the BAH patients and the 29 or 26
APA patients with negative tests for PAC or 18-OH-B, respectively).
Supine PAC and 18-OH-B levels were significantly lower in the BAH group
than in the APA subgroup (P < 0.01; Table 2). The percent increase of PAC in the
BAH group was 
2.5-fold that of the APA subgroup (P =
0.007) although there were no differences in the percent increase of
18-OH-B (Table 2). More patients with BAH had supine PAC and 18-OH-B
values less than 15 ng/dL and 60 ng/dL, respectively, than did patients
in the APA subgroup (P = 0.049 and 0.024; Table 3). A change in PAC greater than 60% was
also more prevalent in the BAH population than in the APA subgroup
(P = 0.011), although percent changes in 18-OH levels
did not discriminate between BAH and APA patients.
|
|
CT studies identified a solitary unilateral adrenal nodule in 34
patients: 4 of 8 (50%) BAH patients and 30 of 41 (73%) APA patients
(P = 0.187). Of the 30 patients with APA who had
nodules, 20 (66%) had negative tests for PAC; 17 of 29 (59%) had
negative tests for 18-OH-B. There was no difference in the prevalence
of a nodule comparing BAH patients with APA patients who had negative
bedside tests (P > 2; Table 3
). Of the 12 patients
with APA who had positive bedside tests for PAC (PAC went down with
ambulation), nodules were seen in 10 (83%). Similarly, of the 14
patients with APA who had positive bedside tests for 18-OH-B, nodules
were seen in 12 (86%).
The 11 APA patients who had negative CT scans (no nodule in 4 and bilateral nodules or equivocal studies in 7) underwent adrenalectomy based on the finding of lateralization on AVS. Pathology confirmed the diagnosis of APA in 11 (100%). Of the four patients with no nodules on CT, pathology revealed two adenomas of 1 cm, one of 0.7 cm, and one of 3 cm. All four BAH patients, two of whom had unilateral nodularity, showed neither lateralization nor suppression on post-ACTH AVS, were diagnosed as BAH, and were treated medically.
AVS
Forty-eight patients underwent AVS as described above. Table 4 represents the mean A/C differentials
for BAH vs. APA patients before and after ACTH stimulation.
Baseline values for APA were greater than those for BAH and suggests a
higher activity of adenomatous tissue (P = 0.024).
After ACTH administration, the A/C differential tended to increase in
APA patients and suggests a unilateral difference in the response to
stimulation (i.e. an adenoma) (P < 0.0001).
In contrast, after ACTH stimulation, values for BAH patients decreased
by half; this reflects the similarity in responsiveness of the two
glands and a trend toward parity.
|
, A–C. Figure 4A
reveals that in BAH, one adrenal gland can exhibit "dominance" over
the other at baseline sampling. After ACTH stimulation, there was
little to no suppression with both sides having a mean activity above
peripheral values. Figure 4B shows that for left-sided tumors some
right adrenal glands (the unaffected side) had A/C ratios higher than
peripheral; after ACTH stimulation, these right-sided glands showed
suppression with mean A/C ratios below peripheral values. The
right-sided tumors were different in that the left glands tended to be
suppressed before as well as after ACTH administration (Fig. 4C).
|
demonstrates that both BAH and
APA patients shared low (<3:1) and high (>5:1) differentials before
ACTH administration. After ACTH, however, no adrenal A/C ratio in a BAH
patient was more than 3-fold the contralateral side. In contrast, 28 of
41 APA adrenal A/C ratios were 7-fold higher than the unaffected side
after ACTH stimulation.
|
APA patients with positive bedside tests were excluded from
regression analysis because no patient with BAH had a positive test for
either PAC or 18-OH-B. Therefore, in the 37 patients with negative
bedside tests for PAC (8 with BAH and 29 with APA), supine PAC, a
cut-off of PAC at 15 ng/dL, the percent change in PAC, and a cut-off in
the percent change at 60% were all predictive (<0.05) (Table 6). In the 34 patients (8 with BAH and 26
with APA) with negative bedside tests for 18-OH-B, a cut-off at 60
ng/dL was predictive (P = 0.04), but continuous supine
18-OH-B values (P = 0.07) or the percent increase in
18-OH-B were not (P = .115). CT failed to have a
significant model fit and was not entered into regression modeling
(2, P = 0.101). An A/C
differential greater than 5:1 was predictive for adenoma with odds
ratios of 6.5 and 7.2 after a negative PAC test (P =
0.02) or a negative 18-OH-B test (P = 0.031),
respectively.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Bedside testing revealed that fewer than half of our pathologically
confirmed APA patients experienced a "classic" fall in PAC (30%)
or 18-OH-B (35%) with ambulation (a positive result). Almost half of
the patients with APA did not demonstrate a circadian fall in PFC,
making evaluation of other hormone changes difficult. Excluding this
group from analysis revealed that 60% of our patients, based on
either PAC or 18-OH-B response, may represent angiotensin-responsive
(AII-R) APAs. The previously reported prevalence of AII-R APAs is
10–30% (5, 11, 23, 24). AII-R patients tend to be
difficult to study noninvasively because their PAC or 18-OH-B levels
tend to rise with provocation, similar to patients with BAH
(25). In our series, the lower observed baseline values in
AII-R than in AII-unresponsive patients have been reported by others
and may reflect two different histological subtypes or even a zona
fasciculata origin of AII-R cells (26, 27). We cannot
exclude the possibility that our study group had a higher prevalence of
AII responsiveness because of a referral bias toward difficult cases or
toward cases that had equivocal bedside testing. Perhaps an indication
of such a possibility is our low percentage (about 50%) of patients
with falls in morning cortisol.
Ten of 12 (83%) patients with positive tests for PAC had a nodule on CT compared with 24 of 38 (63%) patients with negative bedside tests, including 4 patients with BAH. Therefore, CT was helpful in those patients whose hormone levels fell with ambulation diagnosing the side of disease as confirmed by AVS.
Importantly, all of our patients with positive bedside testing for PAC
or 18-OH-B (i.e. a fall in hormone levels with ambulation)
had an APA. Two patients who had negative PAC responses had positive
18-OH-B responses, an observation first used to advocate testing with
both PAC and 18-OH-B (6). In those patients with positive
tests who also had a nodule on CT, AVS was merely confirmatory for
diagnosis and location (Fig. 5). It
seems, therefore, that positive bedside testing (i.e. fall
in hormone value) with a positive CT may be sufficient to diagnose and
locate an APA. We could not test this hypothesis for predictability,
however, because no patient with BAH had a positive bedside test.
|
Baseline hormone levels tended to be higher in our AII-R APA patients than in BAH patients and the increase in PAC or 18-OH-B in the AII-R APA patients with ambulation was one half to one third the increase seen in the BAH patients. Comparing AII-R APA and BAH patients, supine PAC values and the percent increase in PAC with ambulation were all predictive. These data imply that the lower the supine PAC value, especially when less than 15 ng/dL, or the greater the increase in PAC with ambulation, especially when more than 80%, the more likely is the diagnosis BAH and not APA. Because a normal PAC is generally less than 14 ng/dL, a predictive cut-off of 15 ng/dL is illustrative of the observation that some patients, especially those with BAH, may have low to normal baseline aldosterone levels, depending on the degree of hypokalemia, renal function, and diet at the time of testing.
The base 2 log-transformed data used in our regression analysis is
easily interpreted. For example, a patient with a supine PAC level
(odds ratio, 4.8178) of 80 ng/dL has 4.8 times the risk of having an
APA as someone with half the value of 80 ng/dL, or 40 ng/dL.
Alternatively, the dichotomous data imply that a patient with a supine
18-OH-B value greater than 60 ng/dL, for example, has 10.5 times the
risk of having an APA as a patient with a value less than 60 ng/dL
(Table 6).
The multivariate analysis revealed that, as clinically suspected, strongly lateralizing baseline AVS (A/C differential, >5:1) contributed significantly to the predictability of the supine hormone tests. A patient with a supine PAC greater than 15 ng/dL and a baseline A/C differential greater than 5:1 on AVS, for example, has about 180 times the odds of having an adenoma compared with the patient with PAC less than 15 ng/dL and an A/C differential less than 5:1.
In the workup of the hyperaldosteronemic patient, the prevalence of AII-R aldosteronomas must be considered. We continue to recommend that patients with hypertension and PHA undergo sodium-repleted bedside testing and CT imaging with 5-mm cuts. Patients with a positive bedside test and a solitary nodule on CT may be recommended for adrenalectomy. For the patient with a negative bedside test, with or without a solitary nodule on CT, or for the patient with a positive bedside test and a negative CT, AVS is mandatory for diagnosis, localization, and optimal management.
|
Footnotes |
|---|
Received May 18, 2000.
Revised August 28, 2000.
Accepted September 8, 2000.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  J. W. Funder, R. M. Carey, C. Fardella, C. E. Gomez-Sanchez, F. Mantero, M. Stowasser, W. F. Young Jr., and V. M. Montori Case Detection, Diagnosis, and Treatment of Patients with Primary Aldosteronism: An Endocrine Society Clinical Practice Guideline J. Clin. Endocrinol. Metab., September 1, 2008; 93(9): 3266 - 3281. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 V.C. Wu, S.C. Chueh, H.W. Chang, W.C. Lin, K.L. Liu, H.Y. Li, Y.H. Lin, K.D. Wu, and B.S. Hsieh Bilateral aldosterone-producing adenomas: differentiation from bilateral adrenal hyperplasia QJM, January 1, 2008; 101(1): 13 - 22. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. J. Auchus, D. W. Chandler, S. Singeetham, N. Chokshi, F. E. Nwariaku, B. L. Dolmatch, S. A. Holt, F. H. Wians Jr., S. C. Josephs, C. K. Trimmer, et al. Measurement of 18-Hydroxycorticosterone during Adrenal Vein Sampling for Primary Aldosteronism J. Clin. Endocrinol. Metab., July 1, 2007; 92(7): 2648 - 2651. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Authors/Task Force Members:, G. Mancia, G. De Backer, A. Dominiczak, R. Cifkova, R. Fagard, G. Germano, G. Grassi, A. M. Heagerty, S. E. Kjeldsen, et al. 2007 Guidelines for the Management of Arterial Hypertension: The Task Force for the Management of Arterial Hypertension of the European Society of Hypertension (ESH) and of the European Society of Cardiology (ESC) Eur. Heart J., June 11, 2007; (2007) ehm236v1. [Full Text] [PDF]
|
|
|
|
 |
|
 N. Daunt Adrenal Vein Sampling: How to Make It Quick, Easy, and Successful RadioGraphics, October 1, 2005; 25(suppl_1): S143 - S158. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P.-F. Plouin, L. Amar, and G. Chatellier Trends in the prevalence of primary aldosteronism, aldosterone-producing adenomas, and surgically correctable aldosterone-dependent hypertension Nephrol. Dial. Transplant., April 1, 2004; 19(4): 774 - 777. [Full Text] [PDF]
|
|
|
|
 |
|
 R. K. Lingam, S. A. Sohaib, I. Vlahos, A. G. Rockall, A. M. Isidori, J. P. Monson, A. Grossman, and R. H. Reznek CT of Primary Hyperaldosteronism (Conn's Syndrome): The Value of Measuring the Adrenal Gland Am. J. Roentgenol., September 1, 2003; 181(3): 843 - 849. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
E. A. Espiner, D. G. Ross, T. G. Yandle, A. M. Richards, and P. J. Hunt Predicting Surgically Remedial Primary Aldosteronism: Role of Adrenal Scanning, Posture Testing, and Adrenal Vein Sampling J. Clin. Endocrinol. Metab., August 1, 2003; 88(8): 3637 - 3644. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Ganguly Distinguishing Unilateral Aldosteronoma from Bilateral Disease J. Clin. Endocrinol. Metab., August 1, 2001; 86(8): 4004 - 4005. [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
). Note the wide spread of the APA group
and the narrow range of the BAH group. Black bars, mean
values.
) and those with a rise in PAC (
). Some of the latter group
of APA patients may have been AII-R. In contrast, all of the values in
BAH patients rose (
). B, The 18-OH-B response with ambulation was
similar. Y-values are ng/dL, mean ± SE.
