Key words renal function - primary aldosteronism - adrenalectomy - hypoaldosteronism
Introduction
Hypertension affects between 10 and 40% of the general population and is a
leading risk factor for cardiovascular death [1 ]. Primary aldosteronism (PA) is the most common cause of endocrine
hypertension [2 ]
[3 ]
[4 ].
The prevalence of PA ranges from 10% to 25% in the hypertensive
population [5 ]. Unilateral PA accounts for
approximately 40% of the cases and is preferentially treated with unilateral
adrenalectomy [2 ]. Although almost all PA
patients improve blood pressure control after surgery, complete clinical success
after biochemical cure has been reported only in about 50% (ranging from 35
to 80%) of patients with unilateral PA after adrenalectomy [2 ]
[6 ]
[7 ]
[8 ].
It has been well established that aldosterone excess is associated with an increased
risk of cardiovascular and cerebrovascular complications in patients with PA [9 ]
[10 ].
Aldosterone and salt induce renal inflammation and fibrosis in hypertensive rats
through production of reactive oxygen species and activation of mitogen-activated
protein kinase [11 ]. PA patients showed more
renal impairment and a high prevalence of proteinuria when compared to hypertensive
matched individuals [12 ]
[13 ]. Moreover, high aldosterone levels in PA
patients can promote glomerular hyperfiltration and, consequently, false elevation
of the estimated glomerular filtration rate (eGFR) before PA treatment [13 ]. In agreement with this hypothesis, few
studies reported a decrease in renal function following unilateral adrenalectomy
[14 ]
[15 ]
[16 ]. Pre-operative aldosterone
levels, age and potassium levels were predictors of eGFR decline after unilateral
adrenalectomy. High systolic preoperative blood pressure was associated with eGFR
decline after 12 months of unilateral adrenalectomy in a single study [14 ]. Moreover, Kobayashi et al. [17 ] demonstrated that a greater acute fall in
eGFR was associated with a smaller long-term decline in renal function after
mineralocorticoid treatment in PA patients, but this association was not observed in
PA patients after adrenalectomy.
Besides the kidney function, the rate and effect of postoperative hypoaldosteronism
after adrenalectomy for unilateral PA were evaluated only by a few studies [18 ]
[19 ]
[20 ]. The presentation of
postoperative hypoaldosteronism might be transient hyperkalemia and very
low/undetectable aldosterone levels or persistent hyperkalemia associated or
not with hypotension. Recently, Wada et al. [21 ] demonstrated that prolonged hypoaldosteronism is likely to develop in
patients with aldosteronoma after adrenalectomy who are older and have impaired
renal function. Fludrocortisone replacement is the proposed treatment for persistent
hyperkalemia, hyponatremia and/or hypotension until the transient
hypoaldosteronism has been resolved [19 ]
[20 ].
The aim of our study was to investigate the evolution of kidney function at different
time points after unilateral adrenalectomy in unilateral PA and to determine
predictive factors for the eGFR decline. In addition, we evaluated the incidence and
effect of biochemical and clinical hypoaldosteronism after surgery.
Patients and Methods
We performed a retrospective cohort study to evaluate PA patients with unilateral
disease submitted to unilateral adrenalectomy between 2010 and 2022 in a
quaternary academic center. We have included all consecutive patients with PA
that underwent unilateral adrenalectomy. The study was approved by the Ethics
Committees of the Hospital das Clínicas, University of São Paulo
Medical School. This is an observational study and complied with the STROBE
guidelines.
Clinical, biochemical and imaging data were collected from patient records. Serum
aldosterone, plasma direct renin concentration (DRC), eGFR, and electrolytes as
sodium and potassium levels were determined at the diagnosis, preoperative
period, and 1 week, 1, 3 and 6 months after laparoscopic unilateral
adrenalectomy. The eGFR was calculated based on the following formula: eGFR
(ml/min/1.73 m2 )=186×serum
creatinine–1.154 ×age–0.203
(×0.742, if female) [22 ].
The primary endpoint was the evaluation of renal function during follow-up (pre-
and post-adrenalectomy) and the factors associated with a decline in
postoperative eGFR. We have excluded 3 cases that did not have renal function at
the first week after adrenalectomy. The secondary endpoint was the incidence of
postoperative biochemical hypoaldosteronism (aldosterone during first week after
surgery ≤4 ng/dl and suppressed renin) and its effect on
renal function and electrolytes. During the first week post-surgery,
nonsteroidal anti-inflammatory drugs are not used for pain relief after
laparoscopic adrenalectomy due to the risk of worsening renal function. In our
protocol, the patients received dipyrone and opioids (usually tramadol or
codeine) for pain relief.
The algorithm for PA investigation followed the 2016 Endocrine Society Guideline
for PA management [2 ]. In our Institution,
a positive screening for PA, defined as an aldosterone/DRC
ratio+>+2 ng/dl/μIU/ml
(or 55.5 pmol/l/μIU/ml) with concomitant
aldosterone levels+>+10 ng/dl (or 277
pmol/l), has a true positive rate of 95% for PA diagnosis [23 ].
Confirmatory test was not performed in PA patients with spontaneous hypokalemia,
suppressed renin levels plus plasma aldosterone
concentration+>+20 ng/dl (554
pmol/l). All other patients performed at least one confirmatory test. We
have employed the following confirmatory tests: 1) Saline infusion test
(infusion of 2 liter of 0.9% saline over 4 h):
aldosterone+>+10 ng/dl (277
pmol/l) at 4 hours confirmed PA; 2) Captopril challenge test
(oral intake of 50 mg captopril after 1 h in seated position):
absence of aldosterone suppression+>+30%
confirmed PA; 3) Furosemide upright test (intravenous injection of 40 mg
furosemide in upright position maintained for 2 h): plasma renin
activity+<+2 ng/ml/h
(DRC <10 μIU/ml) confirmed PA [24 ].
Sequential adrenal venous sampling (AVS) was performed under cosyntropin
continuous infusion by an experienced interventional radiologist. Successful
catheterization was defined by a selectivity index+≥+5.
Unilateral disease was defined by a lateralization
index+≥+4. Bilateral PA was defined by bilateral
aldosterone excess in AVS [25 ]. Unilateral
disease was determined by biochemical PA cure after unilateral adrenalectomy.
Undetermined lateralization included patients that did not undergo AVS
and/or did not present biochemical PA cure after unilateral
adrenalectomy [26 ].
Resistant hypertension is defined as blood pressure that remains above goal in
spite of the concurrent use of 3 antihypertensive agents of different classes
(ideally, one of the 3 agents should be a diuretic and all agents should be
prescribed at optimal dose amounts). Patients with blood pressure at goal while
taking four or more antihypertensive medications are considered to have
controlled resistant hypertension [27 ]
[28 ].
Clinical and biochemical success after adrenalectomy for unilateral disease was
evaluated according to the Primary Aldosteronism Surgical Outcome (PASO)
criteria [29 ]. Biochemical cure was
defined as correction of hypokalemia and normalization of the
aldosterone-to-renin ratio after 6 months of surgery. In patients with a raised
aldosterone-to-renin ratio post-surgery, aldosterone secretion should be
suppressed in a confirmatory test. Complete clinical success was defined as a
blood pressure+<+140×90 mmHg without
anti-hypertensive drugs after 6 months of follow-up. The cut-off of blood
pressure+≥+140×90 mmHg was used to
define stage 1 hypertension in both European and Brazilian guidelines for the
management of arterial hypertension [27 ]
[28 ].
Aldosterone and DRC were measured using an automated chemiluminescence-based
assay (LIAISON kit, DiaSorin, Salugia, Italy) in all patients. Aldosterone
concentration was measured in serum and DRC in plasma with
ethylenediaminetetraacetic acid. Functional sensitivity (lowest concentration at
which the analyte can be reliably detected) was 3 ng/dl (83
pmol/l) for aldosterone and 4 μIU/ml for DRC. DRC normal
range in seated position varies from 4.6 to 46 μU/ml.
Aldosterone normal range in seated position varies from 2.5 to
39.2 ng/dl (69 to 1.087 pmol/l). The inter-assay
coefficient of variation for the aldosterone assay ranged from 12% at
lower concentrations to 6% at higher concentrations. The inter-assay
coefficient of variation for the renin assay was 5.5%. The intra-assay
coefficient of variation was up to 5.3% for aldosterone and up to
11.7% for DRC.
Statistical analysis
Descriptive statistics were presented as absolute and relative frequencies for
qualitative variables, and as means with standard deviations or medians with
interquartile ranges for quantitative variables. The correlation between eGFR at
1 week post-surgery and age, aldosterone, and renin concentration was estimated
using Pearson’s correlation coefficient. The comparison of the mean eGFR
at 1 week post-surgery among different qualitative and quantitative variables,
including sex, aldosterone levels at diagnosis, hypertension severity, PA cure
after adrenalectomy, hypokalemia, number of anti-hypertensive drugs, and renin,
was performed using the t-test, assuming the normality assumption was satisfied.
The normality assumption was assessed using the Shapiro–Francia
test.
The main analysis involved fitting univariable and multivariable linear
regression models considering eGFR at 1 week post-surgery as outcome adjusted by
eGFR 3 months before surgery and the following covariates: sex, age, aldosterone
at diagnosis, renin, hypertension severity, PA cure after adrenalectomy, and the
number of anti-hypertensive drugs. The variable eGFR 3 months before surgery was
used as an adjustable covariable, model known as the ANCOVA test [30 ]. Multiple imputation analyses were
performed as described by Van Buuren [31 ].
For the linear regression model, missing data were imputed with 20 replicates
using MICE (Multiple Imputation with Chained Equations) with predictive mean
matching.
All hypotheses were two-sided, and statistical significance was considered if the
p-value was less than 0.05. The statistical analyses were conducted using the R
statistical software R (R Core Team, v4.1.2).
Results
A total of 94 patients with PA who underwent unilateral adrenalectomy were included
([Table 1 ]). Forty patients
(42.5%) were men and 54 (57.5%) were women, with a mean age of
diagnosis of 48 years (+±+11.8). Women were younger than men
at the diagnosis of PA (45+±+11.52 vs.
52.2+±+11.12; p=0.03). At diagnosis, the mean
aldosterone and renin levels were 29.35 ng/dl [interquartile range
(IQR) 23.13; 44.53] and 4.0 μIU/ml (IQR 4.00; 4:00), respectively.
Hypokalemia at diagnosis was evidenced in 79 out of 94 (84%) of the
patients. Biochemical cure of PA after unilateral adrenalectomy was achieved in 84
out of 94 cases (89.36%). AVS was performed more often in men than in women
(67.5% vs. 48.15%; p=0.034), but PA cure and complete
clinical success were not different among the sexes ([Table 1 ]).
Table 1 Demographic and clinical characteristics of the
patients submitted to unilateral adrenalectomy for primary
aldosteronism.
Variable
Total (n=94)
Male (n=40)
Female (n=54)
p-Value*
Age (years) (mean+±+SD)
48.06+±+11.84
52.20+±+11.12
45.00+±+11.52
0.003#
Resistant hypertension n (%)
79 (84.04)
33 (82.50)
46 (85.19)
0.780
Hypokalemia n (%)
79 (84.04)
33 (82.50)
46 (85.19)
0.780
+≥+3 Antihypertensive drugs n (%)
75 (79.79)
32 (80.00)
43 (79.63)
0.999
Renin at diagnosis (μUI/ml) (median: IQR)
4.00 (4.00; 4.00)
4.00 (4.00; 4.00)
4.00 (4.00; 4.00)
0.158
Aldosterone at diagnosis (ng/dl) (median: IQR)
29.35 (23.13; 44.53)
27.35 (22.10; 42.10)
29.80 (23.33; 45.32)
0.513
Aldosterone-to-renin ratio (median: IQR)
7.34 (5.18; 10.67)
6.62 (5.04; 10.19)
7.45 (5.80; 11.33)
0.426
Adrenal venous sampling n (%)
53 (56.38)
27 (67.50)
26 (48.15)
0.034
PA cure after unilateral adrenalectomy n (%)
84 (89.36)
38 (95.00)
46 (85.119)
0.132
Complete clinical success n (%)
37 (39.36)
13 (32.50)
24 (44.44)
0.285
SD denotes standard deviation. IQR: Interquartile range.
* Fisher exact test; # T-test, U
Mann–Whitney test.
All patients were under spironolactone treatment 3 months before surgery. In our
cohort, 37 patients (40%) had normal blood pressure without any
anti-hypertensive treatment (complete clinical success) after adrenalectomy.
Fifty-seven patients (60%) had an improvement in blood pressure control with
less anti-hypertensive drugs. During the first month after surgery, these 57
patients were treated with amlodipine (5–10 mg/day)
and/or hydralazine (50 mg bid or tid), because these two medications
do not interfere in renal function. Angiotensin receptor blockers and
angiotensin-converting enzyme inhibitors were not used during the first month
because of the risk of worsening hyperkalemia and inducing hypoaldosteronism.
During follow-up, the eGFR reduced in the first postoperative week compared to 3
months before surgery ([Table 2 ]). During the
first 6 months after adrenalectomy, eGFR remained stable at similar levels to the
first week after surgery ([Fig. 1 ]). Age
(r=–0.33, p=0.001), aldosterone levels 3 months before
surgery (r=–0.26, p=0.021) and eGFR 3 months before surgery
(r=0.75, p+<+0.001) had a significant correlation
with eGFR during first postoperative week ([Fig.
2 ]). Renal function during the first week post-surgery did not correlate
with preoperative renin levels.
Fig. 1 Estimated glomerular filtration rate (eGFR) before unilateral
adrenalectomy and at 1 week, 1, 3 and 6 months after surgery.
Fig. 2 Correlations between age (r=–0.352,
p=0.001), aldosterone levels 3 months before surgery
(r=–0.281, p=0.015) and eGFR 3 months before surgery
(r=0.750, p+<+0.0001) with eGFR during first
postoperative week. eGFR: Estimated glomerular filtration rate.
Table 2 Renal function and biochemical evolution before
unilateral adrenalectomy and at 1 week, 1, 3 and 6 months after
surgery.
3 Months before surgery
First week after surgery
1 Month after surgery
3 Months after surgery
6 Months after surgery
Creatinine (mg/dl)
1.15±0.05
1.37±0.068
1.36±0.094
1.32±0.087
1.26±0.081
eGFR (ml/min/1.73 m2 )
53.6±1.71
46.4±1.79
46.1±2.66
47.1±2.52
48.2±2.49
Aldosterone (ng/dl)
47.6±4.03
8.04±0.95
7.83±1.19
9.85±2.04
12.9±4.11
Renin (μUI/ml)
25.1±7.07
9.05±2.14
23±6.67
18.1±3.66
24.4±5.72
Mean+±+standard error. eGFR: Estimated glomerular
filtration rate.
Following adrenalectomy, renin levels decreased at the first week of surgery compared
to preoperative renin levels, which were within the normal range under
spironolactone treatment ([Table 2 ]). Renin
levels increased significantly after 3 months of adrenalectomy when compared to
immediate postoperative levels. Aldosterone concentration decreased significantly
during the first week of surgery and remained at lower limits of the normal range
during the 6 months after surgery.
Next, we performed univariate and multivariate linear regression analysis to search
for predictors of eGFR decrease after 1 week of surgery ([Table 3 ]). Sex, age, renin level, hypokalemia,
biochemical cure of PA after surgery and the number of antihypertensives were not
predictors of eGFR during the first week post-surgery. High aldosterone levels at
diagnosis were statistically correlated with a decline in renal function in the
univariate model (p=0.022). In the multivariate analysis, aldosterone levels
at diagnosis had a tendency to be an independent predictor of renal function after
surgery (p=0.059).
Table 3 Linear regression for risk factors associated with
first week postoperative estimated glomerular filtration rate
(eGFR).
Variable
Univariable model
Multivariable model
Estimate
SE
Statistic
p
Estimate
SE
Statistic
p
Sex
Male
REF
REF
Female
–1.189
2.474
–0.480
0.632
–0.145
2.660
–0.054
0.957
Age
1-unit
0.047
0.118
0.400
0.690
0.016
0.125
0.129
0.898
Aldosterone at diagnosis (ng/dl)
1-unit
–0.076
0.035
–2.168
0.033
–0.070
0.037
–1.913
0.059
Aldosterone 3 months before surgery (ng/dl)
1-unit
–0.042
0.036
–1.163
0.249
Renin at diagnosis (μUI/ml)
1-unit
–0.298
2.005
–0.149
0.882
Renin 3 months before surgery (μUI/ml)
1-unit
0.010
0.020
0.521
0.604
Resistant hypertension
Yes
REF
REF
No
4.737
3.268
1.450
0.151
3.445
3.344
1.030
0.306
Hypokalemia
Yes
REF
No
4.737
3.268
1.450
0.151
PA cure after surgery
Yes
REF
REF
No
–1.929
4.294
–0.449
0.654
–3.363
4.438
–0.758
0.451
Number of antihypertensives
+<+3
REF
+≥+3
2.548
3.157
0.807
0.422
Missing data were imputed with 20 replicates using Multiple Imputation with
Chained Equations (MICE) with predictive mean matching was performed. SE:
Standard error. REF: Reference category.
Postoperative biochemical hypoaldosteronism (aldosterone
≤4 ng/dl and suppressed renin) was diagnosed in 48%
of the cases during the first week after adrenalectomy ([Table 4 ]). Sodium and potassium levels were
not different between patients with or without hypoaldosteronism. Age at diagnosis
and renin levels before surgery did not correlate with postoperative
hypoaldosteronism. Interestingly, the patients with hypoaldosteronism had lower
levels of aldosterone at diagnosis
(31.5+±+21.1 ng/dl vs.
50.4+±+44.4 ng/dl, respectively;
p=0.008) and higher eGFR after 1 week of adrenalectomy
(54+±+15 ng/dl vs.
40+±+17.5 ml/min/1.73 m2 ,
respectively; p+<+0.001). Mild transient hyperkalemia
(5.0–5.5 mEq/l) was evidenced in 10 patients (11.2%) during
the first week after adrenalectomy. Four out of 89 PA patients (4.5%), older
than 60 years and with resistant hypertension, presented biochemical
hypoaldosteronism associated with persistent hyperkalemia and hypotension. These
four cases were treated with fludrocortisone replacement therapy (0.1 mg)
for a period varying from 6 months to 2 years.
Table 4 Comparison of renal function and electrolytes in
patients with and without postoperative biochemical
hypoaldosteronism.
Total (n=89)
Aldosterone after 1 week of surgery
P*
≤4 (n=43)
≥ 5 (n=46)
Renin before adrenalectomy (μUI/ml)
Mean+±+SD
26+±+66.26
21+±+49.32
30+±+78.34
0.629
Median (IQR)
4.0 (4, 11)
4 (4, 11)
4 (4, 10)
Range
4.0, 433
4.0, 275
4.0, 433
eGFR after 1 week of adrenalectomy
(ml/min/1.73
m
2
)
+<+0.001
Mean
46+±+17.7
54+±+14.95
40+±+17.52
Median (IQR)
47 (33.7, 59)
53 (42.45, 62)
37 (27.1, 48)
Range
10.8, 91
18.2, 91
10.8, 91
Sodium after 1 week of adrenalectomy (mEq/l)
0.519
Mean
141+±+3.19
141+±+3.69
141+±+2.68
Median (IQR)
141 (139, 142)
140 (139, 142)
142 (139, 142)
Range
133, 160
136, 160
133, 145
Potassium after 1 week of adrenalectomy
(mEq/l)
0.216
Mean+±+SD
4+±+0.55
4+±+0.55
4+±+0.56
Median (IQR)
4 (3.9, 5)
4 (3.75, 5)
4 (4, 5)
Range
2.6, 6
3.5, 6
2.6, 6
eGFR: Estimated glomerular filtration rate. IQR: Interquartile range.
* Wilcoxon rank sum test.
Discussion
In this study, we evaluated the evolution of renal function and the hypoaldosteronism
risk after unilateral adrenalectomy for PA. Few reports [14 ]
[15 ]
[16 ] previously evaluated renal
function at single time point (at 1 month, 6 months or 12 months) after
adrenalectomy for unilateral PA, but our study investigated the evolution of renal
function and risk of hypoaldosteronism during the first 6 months after adrenalectomy
at consecutive time points (1 week, 1 month, 3 months and 6 months) after
surgery.
The eGFR had a reduction during the first postoperative week compared to eGFR 3
months before surgery. High systolic preoperative blood pressure was associated with
eGFR decline after 12 months of unilateral adrenalectomy in a single study [14 ]. In our study, aldosterone levels at
diagnosis correlated with postoperative renal function at the first week after
adrenalectomy. Indeed, preoperative aldosterone levels, age and potassium levels
were previously defined as predictors of eGFR decline after unilateral adrenalectomy
[15 ]
[16 ].
Recently, Mermejo et al. [32 ] showed that
postoperative suppressed renin levels became detectable between fifteen and thirty
days after unilateral adrenalectomy in 13 patients with unilateral PA. Evaluating a
large cohort of 94 PA patients, we demonstrated that renin levels increased during
the first month of adrenalectomy, but renin concentrations became higher when
compared to immediate postoperative levels only after 3 months of adrenalectomy.
Similar to renin levels, aldosterone concentration decreased significantly during
the first week of surgery, but remained at lower limits at the normal range during
the 6 postoperative months.
Prolonged hypoaldosteronism accompanied by hyperkalemia was observed in
5–7% of PA patients in two previous studies [19 ]
[20 ].
Although the frequency of prolonged hyperkalemia was low, transient hyperkalemia
occurred in 11–13% of the cases [19 ]
[20 ]. Older age (≥53
years), longer duration of hypertension (>10 years), and impaired
preoperative renal function (eGFR <+58.2 ml/min)
were associated with prolonged postoperative hyperkalemia, and mineralocorticoid
receptor antagonist use did not prevent postoperative hyperkalemia [20 ]. Similar to these previous findings, we
demonstrated a rate of transient and prolonged hyperkalemia of 11.2% and
4.5% in PA patients after adrenalectomy, respectively. In our cohort, all
cases with prolonged hyperkalemia and hypoaldosteronism were older than 60 years old
and had a past history of resistant hypertension longer than 10 years.
Interestingly, preoperative renin levels were not different between patients with
and without postoperative hypoaldosteronism.
All previous studies evaluated the percentage decrease of eGFR after surgical
treatment as the main outcome [14 ]
[15 ]
[16 ].
However, it should be emphasized that the use of percentage change from baseline as
an outcome is a statistically inefficient method [30 ]. The analysis of covariance (ANCOVA) with baseline score as a
covariate has the highest statistical power. Therefore, one strength of this study
included the ANCOVA analysis of the postoperative eGFR using the preoperative eGFR
as a covariate instead of using percentage change as the main outcome. Another
strong point was the large number of PA patients. One limitation of our study is the
retrospective design. However, the database was established by collecting
information on consecutive patients with PA who underwent unilateral
adrenalectomy.
In conclusion, our findings demonstrated a decrease in renal function after
unilateral adrenalectomy for PA using analysis of covariance instead of percentage
eGFR decrease. Additionally, aldosterone levels at diagnosis correlated with
postoperative renal function at the first week after adrenalectomy. Although
biochemical hypoaldosteronism and transient hyperkalemia were frequent, clinical
hypoaldosteronism occurred in 5% of our cohort. Therefore, this evidence
supports the clinical relevance of a close monitoring of renal function after
adrenalectomy for unilateral PA to improve patient outcome and avoid complications
related to renal insufficiency, hyperkalemia and/or hypoaldosteronism.