Increasing rates of referrals for investigation of primary aldosteronism at a tertiary centre
View Article PDF
Primary aldosteronism (PA) is a pathology of autonomous secretion of aldosterone from one or both adrenal glands, classically associated with hypertension and hypokalemia. PA appears to carry a greater burden of cardiovascular disease than essential hypertension and is independently associated with end organ damage, including cardiovascular disease, stroke, left ventricular hypertrophy, microalbuminuria and atrial fibrillation.[[1,2]] The prevalence of PA in hypertensive adults is at least 6%, with some estimates exceeding 20% in those with refractory hypertension.[[1,3–4]]
In Aotearoa New Zealand, screening is generally performed with a plasma aldosterone to renin ratio (ARR). PA is suggested by a high aldosterone and appropriately suppressed renin level, with thresholds differing depending on local guidelines, although the sensitivity and specificity of the ARR is variable depending on the assay and threshold used. The diagnosis is confirmed by dynamic testing. A saline infusion test (SIT) is most commonly performed in Aotearoa New Zealand, whereby two litres of normal saline is infused intravenously over 4 hours, and aldosterone levels are measured both before and after the infusion. The expansion in intravascular volume should suppress renin, and thus aldosterone production, in healthy individuals. In patients with PA, the aldosterone concentration remains inappropriately high at the 4-hour mark. Cortisol may also be measured, as a non-suppressed cortisol at 4 hours may indicate a confounding ACTH effect and produce false positive results. For patients in whom the SIT suggests inappropriate aldosterone secretion, adrenal vein sampling (AVS) may then be performed by an interventional radiologist. AVS may be used to identify unilateral disease in patients who are surgical candidates, as unilateral PA can be cured by surgically removing the pathologic adrenal gland. Unilateral disease is typically due to a functioning adenoma, while bilateral disease may be due to adrenal hyperplasia. Contemporary studies show a higher frequency of unilateral adenomas than reported historically with the availability of improved imaging and biochemical assessments, suggesting a higher proportion of unilateral disease in patients with confirmed PA.[[5]] Computed tomography (CT) can be used to identify an adenoma, but caution needs to be applied to interpreting the significance of an adenoma, particularly in older patients where incidental adrenal adenomas are common.[[6]] The exact role of AVS compared with CT to inform the decision to undertake adrenalectomy remains to be clarified.[[7]] Patients with bilateral disease or those who choose not to undertake surgery may be treated with mineralocorticoid antagonists (MRAs) such as spironolactone or eplerenone; however, eplerenone is not funded for this indication in Aotearoa New Zealand. Both surgical and medical therapies have been shown to be effective at reducing blood pressure, and there are likely additional benefits to cardiovascular health independent of blood pressure improvements.[[8–10]]
Despite a readily available screening test, PA has historically been underdiagnosed in Aotearoa New Zealand and internationally.[[11,12]] Lower frequency of screening may be due to the misconception that hypokalemia is a prerequisite for PA or that it would be managed similarly to essential hypertension once diagnosed.[[11,13]] The requirement for medication changes before dynamic testing also discourages clinicians from screening, especially since the ARR may produce frequent false results even when interfering medications have been removed.[[14,15]]
It is well documented in overseas cohorts that certain characteristics increase the likelihood of PA, including hypokalemia, a higher ARR and a greater antihypertensive requirement.[[8,10]] Other characteristics such as a shorter duration of hypertension, lower BMI and absence of end organ damage can predict the likelihood of success from adrenalectomy.[[16]] However, the predictive value of clinical characteristics in predicting diagnosis or response to treatment has not been studied in Aotearoa New Zealand.
In this audit, we described the characteristics of patients referred to our centre for confirmatory testing for PA. We analysed whether any of these characteristics predicted a diagnosis of PA after a positive screening test. For those with an eventual positive diagnosis, we described their surgical and medical outcomes and whether the Primary Aldosteronism Surgical Outcome (PASO) score was able to predict clinical success of surgery.[[16]] These relationships may aid clinicians in diagnosing PA, and among patients who have confirmed disease, may inform the decision of whether to offer surgical intervention.
Methods
Study population
This audit was reviewed and approved by Capital and Coast District Health Board (CCDHB) Clinical Audit and Research Committee. All patients who attended our unit for an SIT, AVS, adrenalectomy or initial post-intervention follow-up for PA between 1 January 2016 and 31 November 2021 were included. At the time of data collection, our unit covered the regions previously known as Capital and Coast, Hutt Valley and Wairarapa District Health Boards (DHBs), collectively referred to as 3DHB. We additionally performed AVS and adrenalectomies for patients in the mid to lower North Island (previously Hawke’s Bay, Whanganui and MidCentral DHBs). Investigations performed up to 19 April 2022 were included. Patients were identified with a searchable template for SIT and AVS results on the electronic medical record (EMR), and a clinician-recorded list of adrenalectomies was used. Email referrals were manually checked to ensure no eligible patients were missed. Patients referred locally for investigation of PA usually have an SIT performed after an ARR result above the local threshold of 30.5 pmol/mIU or high clinical suspicion, such as resistant hypertension and hypokalemia despite a normal ARR. Patients from other regions were typically referred for AVS after case confirmation, though occasionally referred for SIT or directly for adrenalectomy.
Data collection
Demographic information, clinical and biochemical characteristics and antihypertensive medications were preferentially collected from the referral letter and supplemented by the EMR. Ethnicity was analysed as total response ethnicity under Level 1 ethnic codes and compared to the ethnicity distribution of our catchment based on 2018 Census data.[[17]] AVS and SIT results were extracted from templates filled out in the EMR. Data was stored securely in REDCap on hospital servers.
Diagnosis, subtyping and treatment
Plasma aldosterone concentration for ARR, SIT and AVS was measured in EDTA plasma using the IDS-iSYS® Immunochemiluminometric assay with a reference range of 103 to 1,197 pmol/L when upright. Plasma renin concentration was measured in EDTA plasma using the IDS-iSYS® chemiluminescence immunoassay with a reference range of 5.3 to 99.1 mIU/L when upright. A screening ARR cutoff of 30.5 pmol/mIU is recommended by the manufacturer, with a quoted sensitivity of 98.9% and specificity of 78.9% when units are converted from the validation study.[[18]] Screening ARRs were taken by local laboratories and recommended to be tested while sitting and before 10 am. There was no rapid cortisol assay available during AVS procedures to assess successful cannulation.
Local SIT protocols aligned with Endocrine Society recommendations. MRAs were stopped 6 weeks prior to testing, and all other antihypertensive medications except alpha-blockers and calcium channel blockers were stopped 2 weeks prior, unless continuation was clinically necessary. The protocol for AVS procedures was the same but specified a 4-week washout for MRAs. The test was performed supine with potassium, creatinine and aldosterone measured at 0 minutes and 4 hours. Renin was additionally measured at 0 minutes for a baseline ARR. Protocols did not change over the study period. All SIT and AVS results were interpreted through discussion at a departmental meeting that included at least one consultant endocrinologist. Cortisol was not routinely measured as part of an SIT during the study period. Local SIT protocols use the following 4-hour aldosterone levels as thresholds for the diagnosis of PA:
• <140 pmol/L—aldosterone suppressed; PA excluded.
• 140–280 pmol/L—indeterminate; diagnosis based on renin, aldosterone and potassium levels over duration of the test. May be repeated after optimising interfering factors.
• >280 pmol/L—aldosterone not suppressed; PA confirmed.
Due to the increasing frequency of adrenal incidentalomas with increasing patient age, AVS was required in patients aged over 40 years. AVS were Synacthen-stimulated and performed by a single operator, except in three unsuccessful cases where it was documented that an alternative operator had attempted the procedure. A selectivity index >5:1 confirmed cannulation success and a lateralisation index of >4:1 confirmed unilateral secretion. In patients under 40 years old, radiological lateralisation was accepted without AVS. Based on lateralisation data and patient preference, either surgical or medical therapy was offered as clinically indicated. Patients outside of our catchment were included in surgical outcome analyses if they received adrenalectomy at our centre, and follow-up was available. Patients who were diagnosed and/or followed up at our centre were included in the medical treatment outcome analyses if follow-up was available. Follow-up data was preferentially collected from the EMR through clinic appointment letters 6–12 months after intervention, but interactions outside the window were used when this was unavailable. The primary care and electronic prescription records were used to supplement hospital records or as the main data source when clinic letters were unavailable.
Clinical and biochemical success criteria[[19,20]]
Defined daily dose (DDD) was used to quantify total antihypertensive medication requirement and determine clinical success. DDD was calculated using the World Health Organization ATC/DDD Index 2021.[[21]] Changes in DDD were significant if they were more than 0.5 times the pre-intervention DDD. Changes in blood pressure were significant if systolic blood pressure reduced by >20mmHg or diastolic by >10mmHg. Normotension was defined as office readings of <140/90mmHg or home readings of <135/85mmHg. Antihypertensive medications used for other indications such as alpha-blockers for prostatic hypertrophy or beta-blockers for atrial fibrillation were included in the DDD. Pre-treatment blood pressure readings and DDD were taken from the closest time to initiating an MRA or surgery as possible, typically the clinic appointment prior to intervention, but the referral blood pressure and DDD was used if unavailable.
For surgically treated patients, complete clinical success required normotension and no antihypertensive medication at follow-up, and partial clinical success required a reduction in blood pressure and/or DDD compared to pre-operative status. Patients treated with an MRA were deemed to have had complete clinical success if they remained normotensive on MRA therapy without other antihypertensive medication. Partial clinical success required a reduction in blood pressure and/or DDD after starting an MRA. Complete biochemical success required a normal ARR at follow-up without hypokalemia, while partial biochemical success required reduction in plasma aldosterone of >50% from pre-surgery with no hypokalemia.
PASO surgical outcome score
View Figures 1–3, Tables 1–4.
The PASO surgical outcome score can be used to predict likelihood of complete clinical success for surgically treated patients with unilateral disease, where a favorable outcome is more likely with a shorter duration of hypertension, female sex, lower BMI, lower antihypertensive medication requirement, absence of target organ damage (TOD) and larger adrenal lesion. A score greater than 16 predicts probable clinical success.
Onset of hypertension was determined by the earliest date of the patient taking antihypertensive medications or a recorded diagnosis of hypertension. Sex and BMI were typically provided in the referral letter. TOD was present if the patient had documented microalbuminuria or met criteria for left ventricular hypertrophy on ECG or echocardiogram, as documented in the EMR. When unavailable or when neither of these criteria were met, TOD was entered as absent while BMI, duration of hypertension and adrenal lesion diameter were entered as the median value.[[22]]
Analyses
When patient information was unavailable for a particular analysis, the patient was excluded from that analysis only. Except when discussing treatment outcomes and numbers of procedures performed, patients outside of our region were excluded from analyses as they typically had been diagnosed prior to referral and were heterogeneous to those identified locally.
Independent t-Tests were used to describe differences in means between groups unless results were non-parametric, in which case a Mann–Whitney U test was used. Paired t-Tests were used to describe mean changes before and after intervention unless results were non-parametric, in which case a Wilcoxon Rank-Sum Test was used. Differences in proportions were calculated using an n-1 Chi-squared test. Statistical significance was set at a p-value of 0.05.
Results
180 patients met the criteria for inclusion, 24 being out of our catchment. Among all eligible patients, from 14 October 2014 to 24 March 2022, we performed 173 SIT and 80 AVS procedures, as well as 27 adrenalectomies (Figure 2). This includes 4 SIT, 19 AVS and one adrenalectomy performed for patients out of catchment. Twenty patients had more than one SIT and 16 patients had more than one AVS performed at our centre.
156 eligible patients were identified within our catchment. Baseline characteristics are shown in Table 1. Most of these referrals came from general practitioners (57%), followed by cardiologists (9.6%), renal physicians (5.8%), general medicine physicians (5.8%), general surgeons (1.9%) and other specialists (1.3%). Eleven percent were identified by an endocrinologist. The remainder (7.7%) had unknown referrers. The reported ethnicity of referred patients in comparison to ethnicity rates in our region is shown in Figure 3.
Indications for referral
Forty patients out of 140 identified in our catchment with adequate referral information did not meet any Endocrine Society guidelines for PA screening (29%), as outlined in Table 2. This included two normotensive patients, both of whom were eventually diagnosed with normotensive PA. Three patients had an ARR below the referral threshold of 30.5 pmol/mIU, but there was high clinical suspicion of PA. Two of these patients went on to be diagnosed with PA.
One patient with a suppressed SIT was categorised as having PA due to a convincingly high baseline ARR and aldosterone. One patient with a non-suppressed SIT was determined not to have PA due to an inappropriately high 4-hour renin.
Subtypes of PA
Sixty-three eligible patients within our catchment (40%) eventually received a diagnosis of PA, 76 patients (49%) had the diagnosis excluded, and the remaining 17 patients were still under investigation at completion of data collection or had indeterminate investigations, so were excluded from Table 1. We performed 169 SIT for 150 unique patients within our catchment, contributing to 58 confirmed PA diagnoses, not including five patients diagnosed with PA without SIT, in keeping with Endocrine Society guidelines. This resulted in a positive SIT rate of 34%.
Of those who received a positive diagnosis, 22 had unilateral disease (35%), 20 had bilateral disease (32%) and 21 were unconfirmed (33%), where AVS was still being awaited (9.5%), not attempted (9.5%) or unsuccessful (14%). Six patients under 40 years old with unilateral findings on imaging were subtyped as unilateral PA and referred for surgery without AVS in keeping with local protocols.
Eighty AVS procedures were performed to subtype eligible patients who had confirmed or probable PA, including 14 second-attempt AVS and two third-attempt AVS. The success rate was 64% among patients’ first AVS procedures and 58% when including second and third procedures. Twenty-five of these procedures were performed for patients outside of our catchment, and three unsuccessful AVS attempts were documented as not being performed by the primary operator. AVS failure related to failure to successfully cannulate the right adrenal vein on all but two occasions. There were no significant AVS complications documented in the EMR over the study period.
Treatment outcomes
Clinical and biochemical treatment outcomes are shown in Table 4. Twenty-five patients out of 27 had follow-up available after adrenalectomy performed in our region (mean follow-up 256 days, range 16 to 1,456 days). Of these 25 with available follow-up information, 24 adrenalectomies were performed by a single endocrine surgeon, including three in private. Two additional patients who met eligibility criteria for inclusion in the study were excluded from analyses of treatment success as they had adrenalectomy performed outside of our centre. Twenty-four patients (96%) achieved clinical success, including seven patients (28%) with complete clinical success and 17 (68%) with partial clinical success. Only one patient (4%) had absent clinical success. Complete biochemical success occurred in 20 patients (80%), with one patient each having partial biochemical success and absent biochemical success (4% each). Three patients did not have a post-operative ARR available (12%).
A PASO surgical outcome score of >16 correctly predicted complete clinical success of adrenalectomy for unilateral PA with a sensitivity of 57% and a specificity of 89%. The positive predictive value was 67% and the negative predictive value was 84%, giving the model an accuracy of 80% in our sample.
Of the 20 patients who underwent documented medical treatment with an MRA and had follow-up information available, 14 (70%) achieved clinical success, including seven patients (35%) with complete clinical success and seven (35%) with partial clinical success. Five (25%) had absent clinical success. The remaining one patient had an increased BP but reduced antihypertensive medication dosage. Three additional patients had follow-up with inadequate information to determine clinical success.
Adverse events of surgical treatment
Of the 23 patients with available surgical information who underwent adrenalectomy under the care of an endocrine surgeon in our centre, 12 (52%) experienced self-limiting post-operative numbness. One patient experienced permanent sensory loss at the T12 dermatome. Another was readmitted for transient post-operative pain. One patient experienced a major adverse event (4%) with a post-operative acute kidney injury causing deterioration of chronic kidney disease. There were no recorded cases of significant intraoperative complications or perioperative mortality.
Adverse events of medical treatment
One patient of eight who tried eplerenone experienced severe vomiting, but no other patients had documented adverse effects. Thirteen out of 33 patients who tried spironolactone experienced adverse drug reactions (39%), specifically gynaecomastia (24%), low libido (6%), hypotension (6%), fatigue (3%), constipation (3%) and unspecified intolerance (6%). Two patients (6%) stopped spironolactone for conception. One additional patient not included in the above group denied MRA therapy due to the potential anti-androgenic side effects.
Discussion
We can expect that almost 8,000 people are living with PA in our catchment size of 409,260 individuals aged 20 years or over, given the national prevalence of hypertension of 31% and a conservative estimate that PA is present in 6% of adults with hypertension based on international data.[[17,23]] We are not aware of any recent data on the likely prevalence of PA in Aotearoa New Zealand. Over the 5-year study period we only had 63 confirmed diagnoses of PA in our catchment and 156 eligible patients identified. While this indicates that PA is likely to be heavily underdiagnosed in our region, the number of referrals and SITs undertaken have both increased five-fold from 2015 to 2021 with a clear upward trend, even with the interruptions to services due to COVID-19 in 2020 and 2021. Rates of diagnosis have also generally shown a positive trend over the same period.
Compared to the total response ethnicity of adults aged 20 and over based on 2018 Census data for the region, there were significantly more referrals than expected for patients of Asian ethnicity and less than expected for patients of European ethnicity. The proportion of referred patients who identified as Māori was similar to the proportion of Māori in the region. There are no other data describing ethnic differences in PA prevalence in Aotearoa New Zealand, including the effect of PA on Māori, which is particularly crucial to establish given the inequities in cardiovascular disease which affect Māori.[[24]]
We found that certain characteristics may offer value in predicting a diagnosis of PA independently of having a raised ARR. In our sample of patients referred in for assessment of potential PA, hypokalemia or potassium supplementation were not sensitive indicators of PA but were relatively specific. Other findings may increase the likelihood of PA, including higher antihypertensive requirement, ARR level and cardiovascular disease. Despite 29% of referred patients not having clear evidence of meeting the Endocrine Society criteria for PA screening, we support the use of these criteria as well as using clinical judgement for case detection, given that 18% of patients within our catchment with eventual PA did not clearly meet these criteria at referral.
We observed high rates of complete or partial clinical success in our cohort. However, 42% of male patients experienced gynaecomastia with spironolactone. PA patients may benefit from eplerenone being subsidised as it has lower rates of adverse effects, including in our cohort, but thought to have similar or only slightly lower efficacy in treating PA.[[25,26]] Currently eplerenone is not funded to use for PA in Aotearoa New Zealand, despite being funded for use in congestive heart failure, so would cost PA patients $300 per year at a typical dose of 50mg.[[27]]
Burrello et al. have validated a PASO score of >16 to predict complete clinical success of adrenalectomy with an accuracy of 79% (sensitivity of 71%, specificity 84%) with an original cohort of 380 patients, which largely matches what was observed in our sample.[[16]] Our sensitivity was somewhat lower (57%), although we had a very small sample size of only seven patients with complete clinical success.
Little has been published on PA in Aotearoa New Zealand, which is a barrier to generating awareness and understanding of this condition that requires disease-specific therapy to be managed effectively.[[11]] Our study has several strengths which aim to remedy this. It is the first in Aotearoa New Zealand to examine the increasing rates of referrals and testing for PA. Our findings reinforce the importance of identifying PA with targeted screening, including in patients with longstanding hypertension. On average, our cohort had almost 10 years of established hypertension before being diagnosed with PA, but most still received effective intervention.
Our study is also the first in Aotearoa New Zealand to discuss the predictive value of clinical and biochemical characteristics in the diagnosis of PA, as well as attempting to validate the PASO predictive score. Herd et al. have previously described the effectiveness of surgical treatment of PA in Aotearoa New Zealand, but did not go so far as to describe the characteristics that predict a positive diagnosis.[[28]] A handful of papers published in other journals have discussed case reports, the point prevalence of PA and role of dynamic testing in Aotearoa New Zealand, but none have described the entire course of diagnosis and treatment, medical treatment outcomes, or increasing referrals. This cohort is comparable with cohorts described in the international literature, noting similar rates of complete clinical success from surgery (28% in our cohort vs 37%), medical treatment (both 35%) and suppressed SITs (46% vs 44%).[[19,20,29]]
Our study has several limitations. Firstly, patients only became eligible when undertaking specialist investigation for PA, so the characteristics we described in Table 1 may therefore be less valid in a primary care setting. Secondly, the study was limited to a single centre and retrospective, which contributed to missing information. For example, 52% of patients did not have sufficient information to exclude or confirm presence of TOD for the PASO score, so the imputation of this value as absent reduces the predictive validity in our cohort. Follow-up occurred outside of the recommended 6–12 month window in 52% of surgically treated patients, which may distort rates of clinical success.[[30]] Our AVS success rates being lower than expected from international literature also reduced observed rates of unilateral disease.[[31]] Finally, we did not routinely measure cortisol as part of our SIT protocols, which may have introduced a confounding effect of aldosterone levels from ACTH.[[8]] We hope to eventually move to using mass spectrometry for these assays when it becomes available in Aotearoa New Zealand, given the inaccuracy of immunoassays in diagnosing PA.[[5]]
Future research should aim to overcome these limitations by describing the status of PA from a primary care point of view where patients are most frequently identified, which may also allow estimates of prevalence, including ethnic differences and inequities. Prospectively describing the medical and surgical outcomes of PA patients would more effectively describe the success rates and harms of treatment, including providing further evidence to support funding eplerenone.
Summary
Abstract
Aim
To describe the frequency and characteristics of patients referred for specialist investigation of primary aldosteronism (PA) in the lower North Island over a 5-year period, and the outcomes of those who received treatment.
Method
Patients who underwent confirmatory testing or treatment for PA at Wellington Regional Hospital were retrospectively identified and data were collected from electronic clinical records.
Results
There has been a five-fold increase in both referrals and confirmatory testing for PA in 2021 compared to 2015. Compared to patients without PA, those eventually diagnosed with PA had a higher ARR, serum sodium, antihypertensive requirement and cardiovascular disease prevalence, as well as lower serum renin, potassium and GFR (all p <0.05), but similar blood pressure. Complete or partial clinical success was achieved in 96% of surgically treated patients compared with 70% of medically treated patients. Thirty-nine percent of patients experienced minor adverse effects with spironolactone and only one significant adverse event was experienced perioperatively.
Conclusion
The rate of referrals and confirmatory testing for PA are increasing in our region. Adrenalectomy and mineralocorticoid antagonist therapy are both safe and effective treatments, although minor adverse effects were common with spironolactone.
Author Information
Acknowledgements
Correspondence
Correspondence Email
Competing Interests
1) Parasiliti-Caprino M, Lopez C, Prencipe N et al. Prevalence of primary aldosteronism and association with cardiovascular complications in patients with resistant and refractory hypertension. J Hypertens. 2020;38(9):1841-1848. doi: 10.1097/HJH.0000000000002441.
2) Milliez P, Girerd X, Plouin PF et al. Evidence for an increased rate of cardiovascular events in patients with primary aldosteronism. J Am Coll Cardiol. 2005;45(8):1243-1248. doi: 10.1016/j.jacc.2005.01.015.
3) Monticone S, Burrello J, Tizzani D et al. Prevalence and clinical manifestations of primary aldosteronism encountered in primary care practice. J Am Coll Cardiol. 2017;69(14):1811-1820. doi: 10.1016/j.jacc.2017.01.052.
4) Brown JM, Siddiqui M, Calhoun DA et al. The unrecognised prevalence of primary aldosteronism: a cross-sectional study. Ann Intern Med. 2020;173(1):10-20. doi: 10.7326/M20-0065.
5) Eisenhofer G, Kurlbaum M, Peitzsch M et al. The Saline Infusion Test for Primary Aldosteronism: Implications of Immunoassay Inaccuracy. J Clin Endocrinol Metab. 2022;107(5):e2027-e2036. doi: 10.1210/clinem/dgab924.
6) Jing Y, Hu J, Luo R et al. Prevalence and characteristics of adrenal tumors in an unselected screening population: A cross-sectional study. Ann Intern Med. 2022;175(10):1383-1391. doi: 10.7326/M22-1619.
7) Dekkers T, Prejbisz A, Kool LJS et al. Adrenal vein sampling versus CT scan to determine treatment in primary aldosteronism: an outcome-based randomised diagnostic trial. Lancet Diabetes Endocrinol. 2016;4(9):739-746. doi: 10.1016/S2213-8587(16)30100-0.
8) Funder JW, Carey RM, Mantero F et al. The management of primary aldosteronism: case detection, diagnosis, and treatment: an endocrine society clinical practice guideline. J Clin Endocrinol Metab. 2016;101(5):1889-1916. doi: 10.1210/jc.2015-4061.
9) Muth A, Ragnarsson O, Johannsson G et al. Systematic review of surgery and outcomes in patients with primary aldosteronism. Br J Surg. 2015;102(4):307-317. doi: 10.1002/bjs.9744.
10) Catena C, Colussi G, Nadalini E et al. Cardiovascular outcomes in patients with primary aldosteronism after treatment. Arch Intern Med. 2008;168(1):80-85. doi: 10.1001/archinternmed.2007.33.
11) van der Merwe W, van der Merwe V. May we at least have a civilised discussion about primary aldosteronism in New Zealand? N Z Med J. 2015;128(1419):63-64.
12) Mulatero P, Stowasser M, Loh KC et al. Increased diagnosis of primary aldosteronism, including surgically correctable forms, in centers from five continents. J Clin Endocrinol Metab. 2004;89(3):1045-1050. doi: 10.1210/jc.2003-031337.
13) Kaplan NM. Hypokalemia in the hypertensive patient, with observations on the incidence of primary aldosteronism. Ann Intern Med. 1967;66(6):1079-1090. doi: 10.7326/0003-4819-66-6-1079.
14) Nainani AK, Yang J, Peters S, Russell G. ‘I can’t understand why others don’t screen more’: a qualitative study exploring why Australian general practitioners screen for primary aldosteronism. BMJ Open. 2022;12(6):e061671. doi: 10.1136/bmjopen-2022-061671.
15) Stowasser M, Ahmed AH, Pimenta E et al. Factors affecting the aldosterone/renin ratio. Horm Metab Res. 2012;44(3):170-176. doi: 10.1055/s-0031-1295460.
16) Burrello J, Burrello A, Stowasser M et al. The Primary Aldosteronism Surgical Outcome Score for the Prediction of Clinical Outcomes After Adrenalectomy for Unilateral Primary Aldosteronism. Ann Surg. 2020;272(6):1125-1132. doi: 10.1097/SLA.0000000000003200.
17) Stats NZ. Subnational population estimates (DHB, DHB constituency), by age and sex, at 30 June 1996-2021 (2021 boundaries) [Internet]. Wellington (NZ): Statistics NZ [cited 2022 July 30]. Available from: https://nzdotstat.stats.govt.nz/wbos/Index.aspx?DataSetCode=TABLECODE7509.
18) Manolopoulou J, Fischer E, Dietz A et al. Clinical validation for the aldosterone-to-renin ratio and aldosterone suppression testing using simultaneous fully automated chemiluminescence immunoassays. J Hypertens. 2015;33(12):2500-2511. doi: 10.1097/HJH.0000000000000727.
19) Williams TA, Lenders JWM, Mulatero P et al. Outcomes after adrenalectomy for unilateral primary aldosteronism: an international consensus on outcome measures and analysis of remission rates in an international cohort. Lancet Diabetes Endocrinol. 2017;5(9):689-699. doi: 10.1016/S2213-8587(17)30135-3.
20) Lim YY, Shen J, Fuller PJ, Yang J. Current pattern of primary aldosteronism diagnosis: Delayed and complicated. Aust J Gen Pract. 2018;47(10):712-718. doi: 10.31128/AJGP-05-18-4587.
21) WHO Collaborating Centre for Drug Statistics Methodology. ATC/DDD Index 2022 [Internet]. Oslo (NO): World Health Organization; 2022 [cited 2022 Jan 1]. Available from: www.whocc.no/atc_ddd_index/.
22) Suurd DPD, Visscher WP, Vorselaars WMCM et al. A simplified primary aldosteronism surgical outcome score is a useful prediction model when target organ damage is unknown - Retrospective cohort study. Ann Med Surg (Lond). 2021;65:102333. doi: 10.1016/j.amsu.2021.102333.
23) McLean RM, Williams S, Mann JI et al. Blood pressure and hypertension in New Zealand: results from the 2008/09 Adult Nutrition Survey. N Z Med J. 2013;126(1372):66-79.
24) Selak V, Poppe K, Grey C et al. Ethnic differences in cardiovascular risk profiles among 475,241 adults in primary care in Aotearoa, New Zealand. N Z Med J. 2020;133(1521):14-27.
25) Endocrinology Subcommittee for the Pharmacology and Therapeutics Advisory Committee. Record of the Endocrinology Subcommittee of PTAC Meeting held on 30 March 2021. Item 12: Eplerenone for primary aldosteronism [Internet]. Wellington (NZ): Pharmac; 2021 [cited 2023 Mar 18]. Available from: https://pharmac.govt.nz/assets/2021-03-30-Endocrinology-Subcommittee-Records-Web-Version.pdf.
26) Parthasarathy HK, Ménard J, White WB et al. A double-blind, randomized study comparing the antihypertensive effect of eplerenone and spironolactone in patients with hypertension and evidence of primary aldosteronism. J Hypertens. 2011;29(5):980-90. doi: 10.1097/HJH.0b013e3283455ca5.
27) PHARMAC. Community Schedule [Internet]. Wellington (NZ): Pharmac; 2023 [cited 2023 Feb 20]. Available from: https://schedule.pharmac.govt.nz/ScheduleOnline.php.
28) Herd A, Harman R, Taylor E. Surgical outcomes following laparoscopic adrenalectomy for treatment of Conn’s syndrome (primary hyperaldosteronism) between 1999 and 2006. N Z Med J. 2010;123(1324):50-56.
29) Libianto R, Russell GM, Stowasser M et al. Detecting primary aldosteronism in Australian primary care: a prospective study. Med J Aust. 2022;216(8):408-412. doi: 10.5694/mja2.51438.
30) O’Toole SM, Hornby C, Sze WC et al. Performance evaluation of scoring systems for predicting post-operative hypertension cure in primary aldosteronism. Clin Endocrinol. 2021;95(4):576-586. doi: 10.1111/cen.14534.
31) Rossi GP, Rossitto G, Amar L et al. Clinical Outcomes of 1625 Patients With Primary Aldosteronism Subtyped With Adrenal Vein Sampling. Hypertension. 2019;74(4):800-808. doi: 10.1161/HYPERTENSIONAHA.119.13463.
For the PDF of this article,
contact nzmj@nzma.org.nz
Increasing rates of referrals for investigation of primary aldosteronism at a tertiary centre
View Article PDF
Primary aldosteronism (PA) is a pathology of autonomous secretion of aldosterone from one or both adrenal glands, classically associated with hypertension and hypokalemia. PA appears to carry a greater burden of cardiovascular disease than essential hypertension and is independently associated with end organ damage, including cardiovascular disease, stroke, left ventricular hypertrophy, microalbuminuria and atrial fibrillation.[[1,2]] The prevalence of PA in hypertensive adults is at least 6%, with some estimates exceeding 20% in those with refractory hypertension.[[1,3–4]]
In Aotearoa New Zealand, screening is generally performed with a plasma aldosterone to renin ratio (ARR). PA is suggested by a high aldosterone and appropriately suppressed renin level, with thresholds differing depending on local guidelines, although the sensitivity and specificity of the ARR is variable depending on the assay and threshold used. The diagnosis is confirmed by dynamic testing. A saline infusion test (SIT) is most commonly performed in Aotearoa New Zealand, whereby two litres of normal saline is infused intravenously over 4 hours, and aldosterone levels are measured both before and after the infusion. The expansion in intravascular volume should suppress renin, and thus aldosterone production, in healthy individuals. In patients with PA, the aldosterone concentration remains inappropriately high at the 4-hour mark. Cortisol may also be measured, as a non-suppressed cortisol at 4 hours may indicate a confounding ACTH effect and produce false positive results. For patients in whom the SIT suggests inappropriate aldosterone secretion, adrenal vein sampling (AVS) may then be performed by an interventional radiologist. AVS may be used to identify unilateral disease in patients who are surgical candidates, as unilateral PA can be cured by surgically removing the pathologic adrenal gland. Unilateral disease is typically due to a functioning adenoma, while bilateral disease may be due to adrenal hyperplasia. Contemporary studies show a higher frequency of unilateral adenomas than reported historically with the availability of improved imaging and biochemical assessments, suggesting a higher proportion of unilateral disease in patients with confirmed PA.[[5]] Computed tomography (CT) can be used to identify an adenoma, but caution needs to be applied to interpreting the significance of an adenoma, particularly in older patients where incidental adrenal adenomas are common.[[6]] The exact role of AVS compared with CT to inform the decision to undertake adrenalectomy remains to be clarified.[[7]] Patients with bilateral disease or those who choose not to undertake surgery may be treated with mineralocorticoid antagonists (MRAs) such as spironolactone or eplerenone; however, eplerenone is not funded for this indication in Aotearoa New Zealand. Both surgical and medical therapies have been shown to be effective at reducing blood pressure, and there are likely additional benefits to cardiovascular health independent of blood pressure improvements.[[8–10]]
Despite a readily available screening test, PA has historically been underdiagnosed in Aotearoa New Zealand and internationally.[[11,12]] Lower frequency of screening may be due to the misconception that hypokalemia is a prerequisite for PA or that it would be managed similarly to essential hypertension once diagnosed.[[11,13]] The requirement for medication changes before dynamic testing also discourages clinicians from screening, especially since the ARR may produce frequent false results even when interfering medications have been removed.[[14,15]]
It is well documented in overseas cohorts that certain characteristics increase the likelihood of PA, including hypokalemia, a higher ARR and a greater antihypertensive requirement.[[8,10]] Other characteristics such as a shorter duration of hypertension, lower BMI and absence of end organ damage can predict the likelihood of success from adrenalectomy.[[16]] However, the predictive value of clinical characteristics in predicting diagnosis or response to treatment has not been studied in Aotearoa New Zealand.
In this audit, we described the characteristics of patients referred to our centre for confirmatory testing for PA. We analysed whether any of these characteristics predicted a diagnosis of PA after a positive screening test. For those with an eventual positive diagnosis, we described their surgical and medical outcomes and whether the Primary Aldosteronism Surgical Outcome (PASO) score was able to predict clinical success of surgery.[[16]] These relationships may aid clinicians in diagnosing PA, and among patients who have confirmed disease, may inform the decision of whether to offer surgical intervention.
Methods
Study population
This audit was reviewed and approved by Capital and Coast District Health Board (CCDHB) Clinical Audit and Research Committee. All patients who attended our unit for an SIT, AVS, adrenalectomy or initial post-intervention follow-up for PA between 1 January 2016 and 31 November 2021 were included. At the time of data collection, our unit covered the regions previously known as Capital and Coast, Hutt Valley and Wairarapa District Health Boards (DHBs), collectively referred to as 3DHB. We additionally performed AVS and adrenalectomies for patients in the mid to lower North Island (previously Hawke’s Bay, Whanganui and MidCentral DHBs). Investigations performed up to 19 April 2022 were included. Patients were identified with a searchable template for SIT and AVS results on the electronic medical record (EMR), and a clinician-recorded list of adrenalectomies was used. Email referrals were manually checked to ensure no eligible patients were missed. Patients referred locally for investigation of PA usually have an SIT performed after an ARR result above the local threshold of 30.5 pmol/mIU or high clinical suspicion, such as resistant hypertension and hypokalemia despite a normal ARR. Patients from other regions were typically referred for AVS after case confirmation, though occasionally referred for SIT or directly for adrenalectomy.
Data collection
Demographic information, clinical and biochemical characteristics and antihypertensive medications were preferentially collected from the referral letter and supplemented by the EMR. Ethnicity was analysed as total response ethnicity under Level 1 ethnic codes and compared to the ethnicity distribution of our catchment based on 2018 Census data.[[17]] AVS and SIT results were extracted from templates filled out in the EMR. Data was stored securely in REDCap on hospital servers.
Diagnosis, subtyping and treatment
Plasma aldosterone concentration for ARR, SIT and AVS was measured in EDTA plasma using the IDS-iSYS® Immunochemiluminometric assay with a reference range of 103 to 1,197 pmol/L when upright. Plasma renin concentration was measured in EDTA plasma using the IDS-iSYS® chemiluminescence immunoassay with a reference range of 5.3 to 99.1 mIU/L when upright. A screening ARR cutoff of 30.5 pmol/mIU is recommended by the manufacturer, with a quoted sensitivity of 98.9% and specificity of 78.9% when units are converted from the validation study.[[18]] Screening ARRs were taken by local laboratories and recommended to be tested while sitting and before 10 am. There was no rapid cortisol assay available during AVS procedures to assess successful cannulation.
Local SIT protocols aligned with Endocrine Society recommendations. MRAs were stopped 6 weeks prior to testing, and all other antihypertensive medications except alpha-blockers and calcium channel blockers were stopped 2 weeks prior, unless continuation was clinically necessary. The protocol for AVS procedures was the same but specified a 4-week washout for MRAs. The test was performed supine with potassium, creatinine and aldosterone measured at 0 minutes and 4 hours. Renin was additionally measured at 0 minutes for a baseline ARR. Protocols did not change over the study period. All SIT and AVS results were interpreted through discussion at a departmental meeting that included at least one consultant endocrinologist. Cortisol was not routinely measured as part of an SIT during the study period. Local SIT protocols use the following 4-hour aldosterone levels as thresholds for the diagnosis of PA:
• <140 pmol/L—aldosterone suppressed; PA excluded.
• 140–280 pmol/L—indeterminate; diagnosis based on renin, aldosterone and potassium levels over duration of the test. May be repeated after optimising interfering factors.
• >280 pmol/L—aldosterone not suppressed; PA confirmed.
Due to the increasing frequency of adrenal incidentalomas with increasing patient age, AVS was required in patients aged over 40 years. AVS were Synacthen-stimulated and performed by a single operator, except in three unsuccessful cases where it was documented that an alternative operator had attempted the procedure. A selectivity index >5:1 confirmed cannulation success and a lateralisation index of >4:1 confirmed unilateral secretion. In patients under 40 years old, radiological lateralisation was accepted without AVS. Based on lateralisation data and patient preference, either surgical or medical therapy was offered as clinically indicated. Patients outside of our catchment were included in surgical outcome analyses if they received adrenalectomy at our centre, and follow-up was available. Patients who were diagnosed and/or followed up at our centre were included in the medical treatment outcome analyses if follow-up was available. Follow-up data was preferentially collected from the EMR through clinic appointment letters 6–12 months after intervention, but interactions outside the window were used when this was unavailable. The primary care and electronic prescription records were used to supplement hospital records or as the main data source when clinic letters were unavailable.
Clinical and biochemical success criteria[[19,20]]
Defined daily dose (DDD) was used to quantify total antihypertensive medication requirement and determine clinical success. DDD was calculated using the World Health Organization ATC/DDD Index 2021.[[21]] Changes in DDD were significant if they were more than 0.5 times the pre-intervention DDD. Changes in blood pressure were significant if systolic blood pressure reduced by >20mmHg or diastolic by >10mmHg. Normotension was defined as office readings of <140/90mmHg or home readings of <135/85mmHg. Antihypertensive medications used for other indications such as alpha-blockers for prostatic hypertrophy or beta-blockers for atrial fibrillation were included in the DDD. Pre-treatment blood pressure readings and DDD were taken from the closest time to initiating an MRA or surgery as possible, typically the clinic appointment prior to intervention, but the referral blood pressure and DDD was used if unavailable.
For surgically treated patients, complete clinical success required normotension and no antihypertensive medication at follow-up, and partial clinical success required a reduction in blood pressure and/or DDD compared to pre-operative status. Patients treated with an MRA were deemed to have had complete clinical success if they remained normotensive on MRA therapy without other antihypertensive medication. Partial clinical success required a reduction in blood pressure and/or DDD after starting an MRA. Complete biochemical success required a normal ARR at follow-up without hypokalemia, while partial biochemical success required reduction in plasma aldosterone of >50% from pre-surgery with no hypokalemia.
PASO surgical outcome score
View Figures 1–3, Tables 1–4.
The PASO surgical outcome score can be used to predict likelihood of complete clinical success for surgically treated patients with unilateral disease, where a favorable outcome is more likely with a shorter duration of hypertension, female sex, lower BMI, lower antihypertensive medication requirement, absence of target organ damage (TOD) and larger adrenal lesion. A score greater than 16 predicts probable clinical success.
Onset of hypertension was determined by the earliest date of the patient taking antihypertensive medications or a recorded diagnosis of hypertension. Sex and BMI were typically provided in the referral letter. TOD was present if the patient had documented microalbuminuria or met criteria for left ventricular hypertrophy on ECG or echocardiogram, as documented in the EMR. When unavailable or when neither of these criteria were met, TOD was entered as absent while BMI, duration of hypertension and adrenal lesion diameter were entered as the median value.[[22]]
Analyses
When patient information was unavailable for a particular analysis, the patient was excluded from that analysis only. Except when discussing treatment outcomes and numbers of procedures performed, patients outside of our region were excluded from analyses as they typically had been diagnosed prior to referral and were heterogeneous to those identified locally.
Independent t-Tests were used to describe differences in means between groups unless results were non-parametric, in which case a Mann–Whitney U test was used. Paired t-Tests were used to describe mean changes before and after intervention unless results were non-parametric, in which case a Wilcoxon Rank-Sum Test was used. Differences in proportions were calculated using an n-1 Chi-squared test. Statistical significance was set at a p-value of 0.05.
Results
180 patients met the criteria for inclusion, 24 being out of our catchment. Among all eligible patients, from 14 October 2014 to 24 March 2022, we performed 173 SIT and 80 AVS procedures, as well as 27 adrenalectomies (Figure 2). This includes 4 SIT, 19 AVS and one adrenalectomy performed for patients out of catchment. Twenty patients had more than one SIT and 16 patients had more than one AVS performed at our centre.
156 eligible patients were identified within our catchment. Baseline characteristics are shown in Table 1. Most of these referrals came from general practitioners (57%), followed by cardiologists (9.6%), renal physicians (5.8%), general medicine physicians (5.8%), general surgeons (1.9%) and other specialists (1.3%). Eleven percent were identified by an endocrinologist. The remainder (7.7%) had unknown referrers. The reported ethnicity of referred patients in comparison to ethnicity rates in our region is shown in Figure 3.
Indications for referral
Forty patients out of 140 identified in our catchment with adequate referral information did not meet any Endocrine Society guidelines for PA screening (29%), as outlined in Table 2. This included two normotensive patients, both of whom were eventually diagnosed with normotensive PA. Three patients had an ARR below the referral threshold of 30.5 pmol/mIU, but there was high clinical suspicion of PA. Two of these patients went on to be diagnosed with PA.
One patient with a suppressed SIT was categorised as having PA due to a convincingly high baseline ARR and aldosterone. One patient with a non-suppressed SIT was determined not to have PA due to an inappropriately high 4-hour renin.
Subtypes of PA
Sixty-three eligible patients within our catchment (40%) eventually received a diagnosis of PA, 76 patients (49%) had the diagnosis excluded, and the remaining 17 patients were still under investigation at completion of data collection or had indeterminate investigations, so were excluded from Table 1. We performed 169 SIT for 150 unique patients within our catchment, contributing to 58 confirmed PA diagnoses, not including five patients diagnosed with PA without SIT, in keeping with Endocrine Society guidelines. This resulted in a positive SIT rate of 34%.
Of those who received a positive diagnosis, 22 had unilateral disease (35%), 20 had bilateral disease (32%) and 21 were unconfirmed (33%), where AVS was still being awaited (9.5%), not attempted (9.5%) or unsuccessful (14%). Six patients under 40 years old with unilateral findings on imaging were subtyped as unilateral PA and referred for surgery without AVS in keeping with local protocols.
Eighty AVS procedures were performed to subtype eligible patients who had confirmed or probable PA, including 14 second-attempt AVS and two third-attempt AVS. The success rate was 64% among patients’ first AVS procedures and 58% when including second and third procedures. Twenty-five of these procedures were performed for patients outside of our catchment, and three unsuccessful AVS attempts were documented as not being performed by the primary operator. AVS failure related to failure to successfully cannulate the right adrenal vein on all but two occasions. There were no significant AVS complications documented in the EMR over the study period.
Treatment outcomes
Clinical and biochemical treatment outcomes are shown in Table 4. Twenty-five patients out of 27 had follow-up available after adrenalectomy performed in our region (mean follow-up 256 days, range 16 to 1,456 days). Of these 25 with available follow-up information, 24 adrenalectomies were performed by a single endocrine surgeon, including three in private. Two additional patients who met eligibility criteria for inclusion in the study were excluded from analyses of treatment success as they had adrenalectomy performed outside of our centre. Twenty-four patients (96%) achieved clinical success, including seven patients (28%) with complete clinical success and 17 (68%) with partial clinical success. Only one patient (4%) had absent clinical success. Complete biochemical success occurred in 20 patients (80%), with one patient each having partial biochemical success and absent biochemical success (4% each). Three patients did not have a post-operative ARR available (12%).
A PASO surgical outcome score of >16 correctly predicted complete clinical success of adrenalectomy for unilateral PA with a sensitivity of 57% and a specificity of 89%. The positive predictive value was 67% and the negative predictive value was 84%, giving the model an accuracy of 80% in our sample.
Of the 20 patients who underwent documented medical treatment with an MRA and had follow-up information available, 14 (70%) achieved clinical success, including seven patients (35%) with complete clinical success and seven (35%) with partial clinical success. Five (25%) had absent clinical success. The remaining one patient had an increased BP but reduced antihypertensive medication dosage. Three additional patients had follow-up with inadequate information to determine clinical success.
Adverse events of surgical treatment
Of the 23 patients with available surgical information who underwent adrenalectomy under the care of an endocrine surgeon in our centre, 12 (52%) experienced self-limiting post-operative numbness. One patient experienced permanent sensory loss at the T12 dermatome. Another was readmitted for transient post-operative pain. One patient experienced a major adverse event (4%) with a post-operative acute kidney injury causing deterioration of chronic kidney disease. There were no recorded cases of significant intraoperative complications or perioperative mortality.
Adverse events of medical treatment
One patient of eight who tried eplerenone experienced severe vomiting, but no other patients had documented adverse effects. Thirteen out of 33 patients who tried spironolactone experienced adverse drug reactions (39%), specifically gynaecomastia (24%), low libido (6%), hypotension (6%), fatigue (3%), constipation (3%) and unspecified intolerance (6%). Two patients (6%) stopped spironolactone for conception. One additional patient not included in the above group denied MRA therapy due to the potential anti-androgenic side effects.
Discussion
We can expect that almost 8,000 people are living with PA in our catchment size of 409,260 individuals aged 20 years or over, given the national prevalence of hypertension of 31% and a conservative estimate that PA is present in 6% of adults with hypertension based on international data.[[17,23]] We are not aware of any recent data on the likely prevalence of PA in Aotearoa New Zealand. Over the 5-year study period we only had 63 confirmed diagnoses of PA in our catchment and 156 eligible patients identified. While this indicates that PA is likely to be heavily underdiagnosed in our region, the number of referrals and SITs undertaken have both increased five-fold from 2015 to 2021 with a clear upward trend, even with the interruptions to services due to COVID-19 in 2020 and 2021. Rates of diagnosis have also generally shown a positive trend over the same period.
Compared to the total response ethnicity of adults aged 20 and over based on 2018 Census data for the region, there were significantly more referrals than expected for patients of Asian ethnicity and less than expected for patients of European ethnicity. The proportion of referred patients who identified as Māori was similar to the proportion of Māori in the region. There are no other data describing ethnic differences in PA prevalence in Aotearoa New Zealand, including the effect of PA on Māori, which is particularly crucial to establish given the inequities in cardiovascular disease which affect Māori.[[24]]
We found that certain characteristics may offer value in predicting a diagnosis of PA independently of having a raised ARR. In our sample of patients referred in for assessment of potential PA, hypokalemia or potassium supplementation were not sensitive indicators of PA but were relatively specific. Other findings may increase the likelihood of PA, including higher antihypertensive requirement, ARR level and cardiovascular disease. Despite 29% of referred patients not having clear evidence of meeting the Endocrine Society criteria for PA screening, we support the use of these criteria as well as using clinical judgement for case detection, given that 18% of patients within our catchment with eventual PA did not clearly meet these criteria at referral.
We observed high rates of complete or partial clinical success in our cohort. However, 42% of male patients experienced gynaecomastia with spironolactone. PA patients may benefit from eplerenone being subsidised as it has lower rates of adverse effects, including in our cohort, but thought to have similar or only slightly lower efficacy in treating PA.[[25,26]] Currently eplerenone is not funded to use for PA in Aotearoa New Zealand, despite being funded for use in congestive heart failure, so would cost PA patients $300 per year at a typical dose of 50mg.[[27]]
Burrello et al. have validated a PASO score of >16 to predict complete clinical success of adrenalectomy with an accuracy of 79% (sensitivity of 71%, specificity 84%) with an original cohort of 380 patients, which largely matches what was observed in our sample.[[16]] Our sensitivity was somewhat lower (57%), although we had a very small sample size of only seven patients with complete clinical success.
Little has been published on PA in Aotearoa New Zealand, which is a barrier to generating awareness and understanding of this condition that requires disease-specific therapy to be managed effectively.[[11]] Our study has several strengths which aim to remedy this. It is the first in Aotearoa New Zealand to examine the increasing rates of referrals and testing for PA. Our findings reinforce the importance of identifying PA with targeted screening, including in patients with longstanding hypertension. On average, our cohort had almost 10 years of established hypertension before being diagnosed with PA, but most still received effective intervention.
Our study is also the first in Aotearoa New Zealand to discuss the predictive value of clinical and biochemical characteristics in the diagnosis of PA, as well as attempting to validate the PASO predictive score. Herd et al. have previously described the effectiveness of surgical treatment of PA in Aotearoa New Zealand, but did not go so far as to describe the characteristics that predict a positive diagnosis.[[28]] A handful of papers published in other journals have discussed case reports, the point prevalence of PA and role of dynamic testing in Aotearoa New Zealand, but none have described the entire course of diagnosis and treatment, medical treatment outcomes, or increasing referrals. This cohort is comparable with cohorts described in the international literature, noting similar rates of complete clinical success from surgery (28% in our cohort vs 37%), medical treatment (both 35%) and suppressed SITs (46% vs 44%).[[19,20,29]]
Our study has several limitations. Firstly, patients only became eligible when undertaking specialist investigation for PA, so the characteristics we described in Table 1 may therefore be less valid in a primary care setting. Secondly, the study was limited to a single centre and retrospective, which contributed to missing information. For example, 52% of patients did not have sufficient information to exclude or confirm presence of TOD for the PASO score, so the imputation of this value as absent reduces the predictive validity in our cohort. Follow-up occurred outside of the recommended 6–12 month window in 52% of surgically treated patients, which may distort rates of clinical success.[[30]] Our AVS success rates being lower than expected from international literature also reduced observed rates of unilateral disease.[[31]] Finally, we did not routinely measure cortisol as part of our SIT protocols, which may have introduced a confounding effect of aldosterone levels from ACTH.[[8]] We hope to eventually move to using mass spectrometry for these assays when it becomes available in Aotearoa New Zealand, given the inaccuracy of immunoassays in diagnosing PA.[[5]]
Future research should aim to overcome these limitations by describing the status of PA from a primary care point of view where patients are most frequently identified, which may also allow estimates of prevalence, including ethnic differences and inequities. Prospectively describing the medical and surgical outcomes of PA patients would more effectively describe the success rates and harms of treatment, including providing further evidence to support funding eplerenone.
Summary
Abstract
Aim
To describe the frequency and characteristics of patients referred for specialist investigation of primary aldosteronism (PA) in the lower North Island over a 5-year period, and the outcomes of those who received treatment.
Method
Patients who underwent confirmatory testing or treatment for PA at Wellington Regional Hospital were retrospectively identified and data were collected from electronic clinical records.
Results
There has been a five-fold increase in both referrals and confirmatory testing for PA in 2021 compared to 2015. Compared to patients without PA, those eventually diagnosed with PA had a higher ARR, serum sodium, antihypertensive requirement and cardiovascular disease prevalence, as well as lower serum renin, potassium and GFR (all p <0.05), but similar blood pressure. Complete or partial clinical success was achieved in 96% of surgically treated patients compared with 70% of medically treated patients. Thirty-nine percent of patients experienced minor adverse effects with spironolactone and only one significant adverse event was experienced perioperatively.
Conclusion
The rate of referrals and confirmatory testing for PA are increasing in our region. Adrenalectomy and mineralocorticoid antagonist therapy are both safe and effective treatments, although minor adverse effects were common with spironolactone.
Author Information
Acknowledgements
Correspondence
Correspondence Email
Competing Interests
1) Parasiliti-Caprino M, Lopez C, Prencipe N et al. Prevalence of primary aldosteronism and association with cardiovascular complications in patients with resistant and refractory hypertension. J Hypertens. 2020;38(9):1841-1848. doi: 10.1097/HJH.0000000000002441.
2) Milliez P, Girerd X, Plouin PF et al. Evidence for an increased rate of cardiovascular events in patients with primary aldosteronism. J Am Coll Cardiol. 2005;45(8):1243-1248. doi: 10.1016/j.jacc.2005.01.015.
3) Monticone S, Burrello J, Tizzani D et al. Prevalence and clinical manifestations of primary aldosteronism encountered in primary care practice. J Am Coll Cardiol. 2017;69(14):1811-1820. doi: 10.1016/j.jacc.2017.01.052.
4) Brown JM, Siddiqui M, Calhoun DA et al. The unrecognised prevalence of primary aldosteronism: a cross-sectional study. Ann Intern Med. 2020;173(1):10-20. doi: 10.7326/M20-0065.
5) Eisenhofer G, Kurlbaum M, Peitzsch M et al. The Saline Infusion Test for Primary Aldosteronism: Implications of Immunoassay Inaccuracy. J Clin Endocrinol Metab. 2022;107(5):e2027-e2036. doi: 10.1210/clinem/dgab924.
6) Jing Y, Hu J, Luo R et al. Prevalence and characteristics of adrenal tumors in an unselected screening population: A cross-sectional study. Ann Intern Med. 2022;175(10):1383-1391. doi: 10.7326/M22-1619.
7) Dekkers T, Prejbisz A, Kool LJS et al. Adrenal vein sampling versus CT scan to determine treatment in primary aldosteronism: an outcome-based randomised diagnostic trial. Lancet Diabetes Endocrinol. 2016;4(9):739-746. doi: 10.1016/S2213-8587(16)30100-0.
8) Funder JW, Carey RM, Mantero F et al. The management of primary aldosteronism: case detection, diagnosis, and treatment: an endocrine society clinical practice guideline. J Clin Endocrinol Metab. 2016;101(5):1889-1916. doi: 10.1210/jc.2015-4061.
9) Muth A, Ragnarsson O, Johannsson G et al. Systematic review of surgery and outcomes in patients with primary aldosteronism. Br J Surg. 2015;102(4):307-317. doi: 10.1002/bjs.9744.
10) Catena C, Colussi G, Nadalini E et al. Cardiovascular outcomes in patients with primary aldosteronism after treatment. Arch Intern Med. 2008;168(1):80-85. doi: 10.1001/archinternmed.2007.33.
11) van der Merwe W, van der Merwe V. May we at least have a civilised discussion about primary aldosteronism in New Zealand? N Z Med J. 2015;128(1419):63-64.
12) Mulatero P, Stowasser M, Loh KC et al. Increased diagnosis of primary aldosteronism, including surgically correctable forms, in centers from five continents. J Clin Endocrinol Metab. 2004;89(3):1045-1050. doi: 10.1210/jc.2003-031337.
13) Kaplan NM. Hypokalemia in the hypertensive patient, with observations on the incidence of primary aldosteronism. Ann Intern Med. 1967;66(6):1079-1090. doi: 10.7326/0003-4819-66-6-1079.
14) Nainani AK, Yang J, Peters S, Russell G. ‘I can’t understand why others don’t screen more’: a qualitative study exploring why Australian general practitioners screen for primary aldosteronism. BMJ Open. 2022;12(6):e061671. doi: 10.1136/bmjopen-2022-061671.
15) Stowasser M, Ahmed AH, Pimenta E et al. Factors affecting the aldosterone/renin ratio. Horm Metab Res. 2012;44(3):170-176. doi: 10.1055/s-0031-1295460.
16) Burrello J, Burrello A, Stowasser M et al. The Primary Aldosteronism Surgical Outcome Score for the Prediction of Clinical Outcomes After Adrenalectomy for Unilateral Primary Aldosteronism. Ann Surg. 2020;272(6):1125-1132. doi: 10.1097/SLA.0000000000003200.
17) Stats NZ. Subnational population estimates (DHB, DHB constituency), by age and sex, at 30 June 1996-2021 (2021 boundaries) [Internet]. Wellington (NZ): Statistics NZ [cited 2022 July 30]. Available from: https://nzdotstat.stats.govt.nz/wbos/Index.aspx?DataSetCode=TABLECODE7509.
18) Manolopoulou J, Fischer E, Dietz A et al. Clinical validation for the aldosterone-to-renin ratio and aldosterone suppression testing using simultaneous fully automated chemiluminescence immunoassays. J Hypertens. 2015;33(12):2500-2511. doi: 10.1097/HJH.0000000000000727.
19) Williams TA, Lenders JWM, Mulatero P et al. Outcomes after adrenalectomy for unilateral primary aldosteronism: an international consensus on outcome measures and analysis of remission rates in an international cohort. Lancet Diabetes Endocrinol. 2017;5(9):689-699. doi: 10.1016/S2213-8587(17)30135-3.
20) Lim YY, Shen J, Fuller PJ, Yang J. Current pattern of primary aldosteronism diagnosis: Delayed and complicated. Aust J Gen Pract. 2018;47(10):712-718. doi: 10.31128/AJGP-05-18-4587.
21) WHO Collaborating Centre for Drug Statistics Methodology. ATC/DDD Index 2022 [Internet]. Oslo (NO): World Health Organization; 2022 [cited 2022 Jan 1]. Available from: www.whocc.no/atc_ddd_index/.
22) Suurd DPD, Visscher WP, Vorselaars WMCM et al. A simplified primary aldosteronism surgical outcome score is a useful prediction model when target organ damage is unknown - Retrospective cohort study. Ann Med Surg (Lond). 2021;65:102333. doi: 10.1016/j.amsu.2021.102333.
23) McLean RM, Williams S, Mann JI et al. Blood pressure and hypertension in New Zealand: results from the 2008/09 Adult Nutrition Survey. N Z Med J. 2013;126(1372):66-79.
24) Selak V, Poppe K, Grey C et al. Ethnic differences in cardiovascular risk profiles among 475,241 adults in primary care in Aotearoa, New Zealand. N Z Med J. 2020;133(1521):14-27.
25) Endocrinology Subcommittee for the Pharmacology and Therapeutics Advisory Committee. Record of the Endocrinology Subcommittee of PTAC Meeting held on 30 March 2021. Item 12: Eplerenone for primary aldosteronism [Internet]. Wellington (NZ): Pharmac; 2021 [cited 2023 Mar 18]. Available from: https://pharmac.govt.nz/assets/2021-03-30-Endocrinology-Subcommittee-Records-Web-Version.pdf.
26) Parthasarathy HK, Ménard J, White WB et al. A double-blind, randomized study comparing the antihypertensive effect of eplerenone and spironolactone in patients with hypertension and evidence of primary aldosteronism. J Hypertens. 2011;29(5):980-90. doi: 10.1097/HJH.0b013e3283455ca5.
27) PHARMAC. Community Schedule [Internet]. Wellington (NZ): Pharmac; 2023 [cited 2023 Feb 20]. Available from: https://schedule.pharmac.govt.nz/ScheduleOnline.php.
28) Herd A, Harman R, Taylor E. Surgical outcomes following laparoscopic adrenalectomy for treatment of Conn’s syndrome (primary hyperaldosteronism) between 1999 and 2006. N Z Med J. 2010;123(1324):50-56.
29) Libianto R, Russell GM, Stowasser M et al. Detecting primary aldosteronism in Australian primary care: a prospective study. Med J Aust. 2022;216(8):408-412. doi: 10.5694/mja2.51438.
30) O’Toole SM, Hornby C, Sze WC et al. Performance evaluation of scoring systems for predicting post-operative hypertension cure in primary aldosteronism. Clin Endocrinol. 2021;95(4):576-586. doi: 10.1111/cen.14534.
31) Rossi GP, Rossitto G, Amar L et al. Clinical Outcomes of 1625 Patients With Primary Aldosteronism Subtyped With Adrenal Vein Sampling. Hypertension. 2019;74(4):800-808. doi: 10.1161/HYPERTENSIONAHA.119.13463.
For the PDF of this article,
contact nzmj@nzma.org.nz
Increasing rates of referrals for investigation of primary aldosteronism at a tertiary centre
View Article PDF
Primary aldosteronism (PA) is a pathology of autonomous secretion of aldosterone from one or both adrenal glands, classically associated with hypertension and hypokalemia. PA appears to carry a greater burden of cardiovascular disease than essential hypertension and is independently associated with end organ damage, including cardiovascular disease, stroke, left ventricular hypertrophy, microalbuminuria and atrial fibrillation.[[1,2]] The prevalence of PA in hypertensive adults is at least 6%, with some estimates exceeding 20% in those with refractory hypertension.[[1,3–4]]
In Aotearoa New Zealand, screening is generally performed with a plasma aldosterone to renin ratio (ARR). PA is suggested by a high aldosterone and appropriately suppressed renin level, with thresholds differing depending on local guidelines, although the sensitivity and specificity of the ARR is variable depending on the assay and threshold used. The diagnosis is confirmed by dynamic testing. A saline infusion test (SIT) is most commonly performed in Aotearoa New Zealand, whereby two litres of normal saline is infused intravenously over 4 hours, and aldosterone levels are measured both before and after the infusion. The expansion in intravascular volume should suppress renin, and thus aldosterone production, in healthy individuals. In patients with PA, the aldosterone concentration remains inappropriately high at the 4-hour mark. Cortisol may also be measured, as a non-suppressed cortisol at 4 hours may indicate a confounding ACTH effect and produce false positive results. For patients in whom the SIT suggests inappropriate aldosterone secretion, adrenal vein sampling (AVS) may then be performed by an interventional radiologist. AVS may be used to identify unilateral disease in patients who are surgical candidates, as unilateral PA can be cured by surgically removing the pathologic adrenal gland. Unilateral disease is typically due to a functioning adenoma, while bilateral disease may be due to adrenal hyperplasia. Contemporary studies show a higher frequency of unilateral adenomas than reported historically with the availability of improved imaging and biochemical assessments, suggesting a higher proportion of unilateral disease in patients with confirmed PA.[[5]] Computed tomography (CT) can be used to identify an adenoma, but caution needs to be applied to interpreting the significance of an adenoma, particularly in older patients where incidental adrenal adenomas are common.[[6]] The exact role of AVS compared with CT to inform the decision to undertake adrenalectomy remains to be clarified.[[7]] Patients with bilateral disease or those who choose not to undertake surgery may be treated with mineralocorticoid antagonists (MRAs) such as spironolactone or eplerenone; however, eplerenone is not funded for this indication in Aotearoa New Zealand. Both surgical and medical therapies have been shown to be effective at reducing blood pressure, and there are likely additional benefits to cardiovascular health independent of blood pressure improvements.[[8–10]]
Despite a readily available screening test, PA has historically been underdiagnosed in Aotearoa New Zealand and internationally.[[11,12]] Lower frequency of screening may be due to the misconception that hypokalemia is a prerequisite for PA or that it would be managed similarly to essential hypertension once diagnosed.[[11,13]] The requirement for medication changes before dynamic testing also discourages clinicians from screening, especially since the ARR may produce frequent false results even when interfering medications have been removed.[[14,15]]
It is well documented in overseas cohorts that certain characteristics increase the likelihood of PA, including hypokalemia, a higher ARR and a greater antihypertensive requirement.[[8,10]] Other characteristics such as a shorter duration of hypertension, lower BMI and absence of end organ damage can predict the likelihood of success from adrenalectomy.[[16]] However, the predictive value of clinical characteristics in predicting diagnosis or response to treatment has not been studied in Aotearoa New Zealand.
In this audit, we described the characteristics of patients referred to our centre for confirmatory testing for PA. We analysed whether any of these characteristics predicted a diagnosis of PA after a positive screening test. For those with an eventual positive diagnosis, we described their surgical and medical outcomes and whether the Primary Aldosteronism Surgical Outcome (PASO) score was able to predict clinical success of surgery.[[16]] These relationships may aid clinicians in diagnosing PA, and among patients who have confirmed disease, may inform the decision of whether to offer surgical intervention.
Methods
Study population
This audit was reviewed and approved by Capital and Coast District Health Board (CCDHB) Clinical Audit and Research Committee. All patients who attended our unit for an SIT, AVS, adrenalectomy or initial post-intervention follow-up for PA between 1 January 2016 and 31 November 2021 were included. At the time of data collection, our unit covered the regions previously known as Capital and Coast, Hutt Valley and Wairarapa District Health Boards (DHBs), collectively referred to as 3DHB. We additionally performed AVS and adrenalectomies for patients in the mid to lower North Island (previously Hawke’s Bay, Whanganui and MidCentral DHBs). Investigations performed up to 19 April 2022 were included. Patients were identified with a searchable template for SIT and AVS results on the electronic medical record (EMR), and a clinician-recorded list of adrenalectomies was used. Email referrals were manually checked to ensure no eligible patients were missed. Patients referred locally for investigation of PA usually have an SIT performed after an ARR result above the local threshold of 30.5 pmol/mIU or high clinical suspicion, such as resistant hypertension and hypokalemia despite a normal ARR. Patients from other regions were typically referred for AVS after case confirmation, though occasionally referred for SIT or directly for adrenalectomy.
Data collection
Demographic information, clinical and biochemical characteristics and antihypertensive medications were preferentially collected from the referral letter and supplemented by the EMR. Ethnicity was analysed as total response ethnicity under Level 1 ethnic codes and compared to the ethnicity distribution of our catchment based on 2018 Census data.[[17]] AVS and SIT results were extracted from templates filled out in the EMR. Data was stored securely in REDCap on hospital servers.
Diagnosis, subtyping and treatment
Plasma aldosterone concentration for ARR, SIT and AVS was measured in EDTA plasma using the IDS-iSYS® Immunochemiluminometric assay with a reference range of 103 to 1,197 pmol/L when upright. Plasma renin concentration was measured in EDTA plasma using the IDS-iSYS® chemiluminescence immunoassay with a reference range of 5.3 to 99.1 mIU/L when upright. A screening ARR cutoff of 30.5 pmol/mIU is recommended by the manufacturer, with a quoted sensitivity of 98.9% and specificity of 78.9% when units are converted from the validation study.[[18]] Screening ARRs were taken by local laboratories and recommended to be tested while sitting and before 10 am. There was no rapid cortisol assay available during AVS procedures to assess successful cannulation.
Local SIT protocols aligned with Endocrine Society recommendations. MRAs were stopped 6 weeks prior to testing, and all other antihypertensive medications except alpha-blockers and calcium channel blockers were stopped 2 weeks prior, unless continuation was clinically necessary. The protocol for AVS procedures was the same but specified a 4-week washout for MRAs. The test was performed supine with potassium, creatinine and aldosterone measured at 0 minutes and 4 hours. Renin was additionally measured at 0 minutes for a baseline ARR. Protocols did not change over the study period. All SIT and AVS results were interpreted through discussion at a departmental meeting that included at least one consultant endocrinologist. Cortisol was not routinely measured as part of an SIT during the study period. Local SIT protocols use the following 4-hour aldosterone levels as thresholds for the diagnosis of PA:
• <140 pmol/L—aldosterone suppressed; PA excluded.
• 140–280 pmol/L—indeterminate; diagnosis based on renin, aldosterone and potassium levels over duration of the test. May be repeated after optimising interfering factors.
• >280 pmol/L—aldosterone not suppressed; PA confirmed.
Due to the increasing frequency of adrenal incidentalomas with increasing patient age, AVS was required in patients aged over 40 years. AVS were Synacthen-stimulated and performed by a single operator, except in three unsuccessful cases where it was documented that an alternative operator had attempted the procedure. A selectivity index >5:1 confirmed cannulation success and a lateralisation index of >4:1 confirmed unilateral secretion. In patients under 40 years old, radiological lateralisation was accepted without AVS. Based on lateralisation data and patient preference, either surgical or medical therapy was offered as clinically indicated. Patients outside of our catchment were included in surgical outcome analyses if they received adrenalectomy at our centre, and follow-up was available. Patients who were diagnosed and/or followed up at our centre were included in the medical treatment outcome analyses if follow-up was available. Follow-up data was preferentially collected from the EMR through clinic appointment letters 6–12 months after intervention, but interactions outside the window were used when this was unavailable. The primary care and electronic prescription records were used to supplement hospital records or as the main data source when clinic letters were unavailable.
Clinical and biochemical success criteria[[19,20]]
Defined daily dose (DDD) was used to quantify total antihypertensive medication requirement and determine clinical success. DDD was calculated using the World Health Organization ATC/DDD Index 2021.[[21]] Changes in DDD were significant if they were more than 0.5 times the pre-intervention DDD. Changes in blood pressure were significant if systolic blood pressure reduced by >20mmHg or diastolic by >10mmHg. Normotension was defined as office readings of <140/90mmHg or home readings of <135/85mmHg. Antihypertensive medications used for other indications such as alpha-blockers for prostatic hypertrophy or beta-blockers for atrial fibrillation were included in the DDD. Pre-treatment blood pressure readings and DDD were taken from the closest time to initiating an MRA or surgery as possible, typically the clinic appointment prior to intervention, but the referral blood pressure and DDD was used if unavailable.
For surgically treated patients, complete clinical success required normotension and no antihypertensive medication at follow-up, and partial clinical success required a reduction in blood pressure and/or DDD compared to pre-operative status. Patients treated with an MRA were deemed to have had complete clinical success if they remained normotensive on MRA therapy without other antihypertensive medication. Partial clinical success required a reduction in blood pressure and/or DDD after starting an MRA. Complete biochemical success required a normal ARR at follow-up without hypokalemia, while partial biochemical success required reduction in plasma aldosterone of >50% from pre-surgery with no hypokalemia.
PASO surgical outcome score
View Figures 1–3, Tables 1–4.
The PASO surgical outcome score can be used to predict likelihood of complete clinical success for surgically treated patients with unilateral disease, where a favorable outcome is more likely with a shorter duration of hypertension, female sex, lower BMI, lower antihypertensive medication requirement, absence of target organ damage (TOD) and larger adrenal lesion. A score greater than 16 predicts probable clinical success.
Onset of hypertension was determined by the earliest date of the patient taking antihypertensive medications or a recorded diagnosis of hypertension. Sex and BMI were typically provided in the referral letter. TOD was present if the patient had documented microalbuminuria or met criteria for left ventricular hypertrophy on ECG or echocardiogram, as documented in the EMR. When unavailable or when neither of these criteria were met, TOD was entered as absent while BMI, duration of hypertension and adrenal lesion diameter were entered as the median value.[[22]]
Analyses
When patient information was unavailable for a particular analysis, the patient was excluded from that analysis only. Except when discussing treatment outcomes and numbers of procedures performed, patients outside of our region were excluded from analyses as they typically had been diagnosed prior to referral and were heterogeneous to those identified locally.
Independent t-Tests were used to describe differences in means between groups unless results were non-parametric, in which case a Mann–Whitney U test was used. Paired t-Tests were used to describe mean changes before and after intervention unless results were non-parametric, in which case a Wilcoxon Rank-Sum Test was used. Differences in proportions were calculated using an n-1 Chi-squared test. Statistical significance was set at a p-value of 0.05.
Results
180 patients met the criteria for inclusion, 24 being out of our catchment. Among all eligible patients, from 14 October 2014 to 24 March 2022, we performed 173 SIT and 80 AVS procedures, as well as 27 adrenalectomies (Figure 2). This includes 4 SIT, 19 AVS and one adrenalectomy performed for patients out of catchment. Twenty patients had more than one SIT and 16 patients had more than one AVS performed at our centre.
156 eligible patients were identified within our catchment. Baseline characteristics are shown in Table 1. Most of these referrals came from general practitioners (57%), followed by cardiologists (9.6%), renal physicians (5.8%), general medicine physicians (5.8%), general surgeons (1.9%) and other specialists (1.3%). Eleven percent were identified by an endocrinologist. The remainder (7.7%) had unknown referrers. The reported ethnicity of referred patients in comparison to ethnicity rates in our region is shown in Figure 3.
Indications for referral
Forty patients out of 140 identified in our catchment with adequate referral information did not meet any Endocrine Society guidelines for PA screening (29%), as outlined in Table 2. This included two normotensive patients, both of whom were eventually diagnosed with normotensive PA. Three patients had an ARR below the referral threshold of 30.5 pmol/mIU, but there was high clinical suspicion of PA. Two of these patients went on to be diagnosed with PA.
One patient with a suppressed SIT was categorised as having PA due to a convincingly high baseline ARR and aldosterone. One patient with a non-suppressed SIT was determined not to have PA due to an inappropriately high 4-hour renin.
Subtypes of PA
Sixty-three eligible patients within our catchment (40%) eventually received a diagnosis of PA, 76 patients (49%) had the diagnosis excluded, and the remaining 17 patients were still under investigation at completion of data collection or had indeterminate investigations, so were excluded from Table 1. We performed 169 SIT for 150 unique patients within our catchment, contributing to 58 confirmed PA diagnoses, not including five patients diagnosed with PA without SIT, in keeping with Endocrine Society guidelines. This resulted in a positive SIT rate of 34%.
Of those who received a positive diagnosis, 22 had unilateral disease (35%), 20 had bilateral disease (32%) and 21 were unconfirmed (33%), where AVS was still being awaited (9.5%), not attempted (9.5%) or unsuccessful (14%). Six patients under 40 years old with unilateral findings on imaging were subtyped as unilateral PA and referred for surgery without AVS in keeping with local protocols.
Eighty AVS procedures were performed to subtype eligible patients who had confirmed or probable PA, including 14 second-attempt AVS and two third-attempt AVS. The success rate was 64% among patients’ first AVS procedures and 58% when including second and third procedures. Twenty-five of these procedures were performed for patients outside of our catchment, and three unsuccessful AVS attempts were documented as not being performed by the primary operator. AVS failure related to failure to successfully cannulate the right adrenal vein on all but two occasions. There were no significant AVS complications documented in the EMR over the study period.
Treatment outcomes
Clinical and biochemical treatment outcomes are shown in Table 4. Twenty-five patients out of 27 had follow-up available after adrenalectomy performed in our region (mean follow-up 256 days, range 16 to 1,456 days). Of these 25 with available follow-up information, 24 adrenalectomies were performed by a single endocrine surgeon, including three in private. Two additional patients who met eligibility criteria for inclusion in the study were excluded from analyses of treatment success as they had adrenalectomy performed outside of our centre. Twenty-four patients (96%) achieved clinical success, including seven patients (28%) with complete clinical success and 17 (68%) with partial clinical success. Only one patient (4%) had absent clinical success. Complete biochemical success occurred in 20 patients (80%), with one patient each having partial biochemical success and absent biochemical success (4% each). Three patients did not have a post-operative ARR available (12%).
A PASO surgical outcome score of >16 correctly predicted complete clinical success of adrenalectomy for unilateral PA with a sensitivity of 57% and a specificity of 89%. The positive predictive value was 67% and the negative predictive value was 84%, giving the model an accuracy of 80% in our sample.
Of the 20 patients who underwent documented medical treatment with an MRA and had follow-up information available, 14 (70%) achieved clinical success, including seven patients (35%) with complete clinical success and seven (35%) with partial clinical success. Five (25%) had absent clinical success. The remaining one patient had an increased BP but reduced antihypertensive medication dosage. Three additional patients had follow-up with inadequate information to determine clinical success.
Adverse events of surgical treatment
Of the 23 patients with available surgical information who underwent adrenalectomy under the care of an endocrine surgeon in our centre, 12 (52%) experienced self-limiting post-operative numbness. One patient experienced permanent sensory loss at the T12 dermatome. Another was readmitted for transient post-operative pain. One patient experienced a major adverse event (4%) with a post-operative acute kidney injury causing deterioration of chronic kidney disease. There were no recorded cases of significant intraoperative complications or perioperative mortality.
Adverse events of medical treatment
One patient of eight who tried eplerenone experienced severe vomiting, but no other patients had documented adverse effects. Thirteen out of 33 patients who tried spironolactone experienced adverse drug reactions (39%), specifically gynaecomastia (24%), low libido (6%), hypotension (6%), fatigue (3%), constipation (3%) and unspecified intolerance (6%). Two patients (6%) stopped spironolactone for conception. One additional patient not included in the above group denied MRA therapy due to the potential anti-androgenic side effects.
Discussion
We can expect that almost 8,000 people are living with PA in our catchment size of 409,260 individuals aged 20 years or over, given the national prevalence of hypertension of 31% and a conservative estimate that PA is present in 6% of adults with hypertension based on international data.[[17,23]] We are not aware of any recent data on the likely prevalence of PA in Aotearoa New Zealand. Over the 5-year study period we only had 63 confirmed diagnoses of PA in our catchment and 156 eligible patients identified. While this indicates that PA is likely to be heavily underdiagnosed in our region, the number of referrals and SITs undertaken have both increased five-fold from 2015 to 2021 with a clear upward trend, even with the interruptions to services due to COVID-19 in 2020 and 2021. Rates of diagnosis have also generally shown a positive trend over the same period.
Compared to the total response ethnicity of adults aged 20 and over based on 2018 Census data for the region, there were significantly more referrals than expected for patients of Asian ethnicity and less than expected for patients of European ethnicity. The proportion of referred patients who identified as Māori was similar to the proportion of Māori in the region. There are no other data describing ethnic differences in PA prevalence in Aotearoa New Zealand, including the effect of PA on Māori, which is particularly crucial to establish given the inequities in cardiovascular disease which affect Māori.[[24]]
We found that certain characteristics may offer value in predicting a diagnosis of PA independently of having a raised ARR. In our sample of patients referred in for assessment of potential PA, hypokalemia or potassium supplementation were not sensitive indicators of PA but were relatively specific. Other findings may increase the likelihood of PA, including higher antihypertensive requirement, ARR level and cardiovascular disease. Despite 29% of referred patients not having clear evidence of meeting the Endocrine Society criteria for PA screening, we support the use of these criteria as well as using clinical judgement for case detection, given that 18% of patients within our catchment with eventual PA did not clearly meet these criteria at referral.
We observed high rates of complete or partial clinical success in our cohort. However, 42% of male patients experienced gynaecomastia with spironolactone. PA patients may benefit from eplerenone being subsidised as it has lower rates of adverse effects, including in our cohort, but thought to have similar or only slightly lower efficacy in treating PA.[[25,26]] Currently eplerenone is not funded to use for PA in Aotearoa New Zealand, despite being funded for use in congestive heart failure, so would cost PA patients $300 per year at a typical dose of 50mg.[[27]]
Burrello et al. have validated a PASO score of >16 to predict complete clinical success of adrenalectomy with an accuracy of 79% (sensitivity of 71%, specificity 84%) with an original cohort of 380 patients, which largely matches what was observed in our sample.[[16]] Our sensitivity was somewhat lower (57%), although we had a very small sample size of only seven patients with complete clinical success.
Little has been published on PA in Aotearoa New Zealand, which is a barrier to generating awareness and understanding of this condition that requires disease-specific therapy to be managed effectively.[[11]] Our study has several strengths which aim to remedy this. It is the first in Aotearoa New Zealand to examine the increasing rates of referrals and testing for PA. Our findings reinforce the importance of identifying PA with targeted screening, including in patients with longstanding hypertension. On average, our cohort had almost 10 years of established hypertension before being diagnosed with PA, but most still received effective intervention.
Our study is also the first in Aotearoa New Zealand to discuss the predictive value of clinical and biochemical characteristics in the diagnosis of PA, as well as attempting to validate the PASO predictive score. Herd et al. have previously described the effectiveness of surgical treatment of PA in Aotearoa New Zealand, but did not go so far as to describe the characteristics that predict a positive diagnosis.[[28]] A handful of papers published in other journals have discussed case reports, the point prevalence of PA and role of dynamic testing in Aotearoa New Zealand, but none have described the entire course of diagnosis and treatment, medical treatment outcomes, or increasing referrals. This cohort is comparable with cohorts described in the international literature, noting similar rates of complete clinical success from surgery (28% in our cohort vs 37%), medical treatment (both 35%) and suppressed SITs (46% vs 44%).[[19,20,29]]
Our study has several limitations. Firstly, patients only became eligible when undertaking specialist investigation for PA, so the characteristics we described in Table 1 may therefore be less valid in a primary care setting. Secondly, the study was limited to a single centre and retrospective, which contributed to missing information. For example, 52% of patients did not have sufficient information to exclude or confirm presence of TOD for the PASO score, so the imputation of this value as absent reduces the predictive validity in our cohort. Follow-up occurred outside of the recommended 6–12 month window in 52% of surgically treated patients, which may distort rates of clinical success.[[30]] Our AVS success rates being lower than expected from international literature also reduced observed rates of unilateral disease.[[31]] Finally, we did not routinely measure cortisol as part of our SIT protocols, which may have introduced a confounding effect of aldosterone levels from ACTH.[[8]] We hope to eventually move to using mass spectrometry for these assays when it becomes available in Aotearoa New Zealand, given the inaccuracy of immunoassays in diagnosing PA.[[5]]
Future research should aim to overcome these limitations by describing the status of PA from a primary care point of view where patients are most frequently identified, which may also allow estimates of prevalence, including ethnic differences and inequities. Prospectively describing the medical and surgical outcomes of PA patients would more effectively describe the success rates and harms of treatment, including providing further evidence to support funding eplerenone.
Summary
Abstract
Aim
To describe the frequency and characteristics of patients referred for specialist investigation of primary aldosteronism (PA) in the lower North Island over a 5-year period, and the outcomes of those who received treatment.
Method
Patients who underwent confirmatory testing or treatment for PA at Wellington Regional Hospital were retrospectively identified and data were collected from electronic clinical records.
Results
There has been a five-fold increase in both referrals and confirmatory testing for PA in 2021 compared to 2015. Compared to patients without PA, those eventually diagnosed with PA had a higher ARR, serum sodium, antihypertensive requirement and cardiovascular disease prevalence, as well as lower serum renin, potassium and GFR (all p <0.05), but similar blood pressure. Complete or partial clinical success was achieved in 96% of surgically treated patients compared with 70% of medically treated patients. Thirty-nine percent of patients experienced minor adverse effects with spironolactone and only one significant adverse event was experienced perioperatively.
Conclusion
The rate of referrals and confirmatory testing for PA are increasing in our region. Adrenalectomy and mineralocorticoid antagonist therapy are both safe and effective treatments, although minor adverse effects were common with spironolactone.
Author Information
Acknowledgements
Correspondence
Correspondence Email
Competing Interests
1) Parasiliti-Caprino M, Lopez C, Prencipe N et al. Prevalence of primary aldosteronism and association with cardiovascular complications in patients with resistant and refractory hypertension. J Hypertens. 2020;38(9):1841-1848. doi: 10.1097/HJH.0000000000002441.
2) Milliez P, Girerd X, Plouin PF et al. Evidence for an increased rate of cardiovascular events in patients with primary aldosteronism. J Am Coll Cardiol. 2005;45(8):1243-1248. doi: 10.1016/j.jacc.2005.01.015.
3) Monticone S, Burrello J, Tizzani D et al. Prevalence and clinical manifestations of primary aldosteronism encountered in primary care practice. J Am Coll Cardiol. 2017;69(14):1811-1820. doi: 10.1016/j.jacc.2017.01.052.
4) Brown JM, Siddiqui M, Calhoun DA et al. The unrecognised prevalence of primary aldosteronism: a cross-sectional study. Ann Intern Med. 2020;173(1):10-20. doi: 10.7326/M20-0065.
5) Eisenhofer G, Kurlbaum M, Peitzsch M et al. The Saline Infusion Test for Primary Aldosteronism: Implications of Immunoassay Inaccuracy. J Clin Endocrinol Metab. 2022;107(5):e2027-e2036. doi: 10.1210/clinem/dgab924.
6) Jing Y, Hu J, Luo R et al. Prevalence and characteristics of adrenal tumors in an unselected screening population: A cross-sectional study. Ann Intern Med. 2022;175(10):1383-1391. doi: 10.7326/M22-1619.
7) Dekkers T, Prejbisz A, Kool LJS et al. Adrenal vein sampling versus CT scan to determine treatment in primary aldosteronism: an outcome-based randomised diagnostic trial. Lancet Diabetes Endocrinol. 2016;4(9):739-746. doi: 10.1016/S2213-8587(16)30100-0.
8) Funder JW, Carey RM, Mantero F et al. The management of primary aldosteronism: case detection, diagnosis, and treatment: an endocrine society clinical practice guideline. J Clin Endocrinol Metab. 2016;101(5):1889-1916. doi: 10.1210/jc.2015-4061.
9) Muth A, Ragnarsson O, Johannsson G et al. Systematic review of surgery and outcomes in patients with primary aldosteronism. Br J Surg. 2015;102(4):307-317. doi: 10.1002/bjs.9744.
10) Catena C, Colussi G, Nadalini E et al. Cardiovascular outcomes in patients with primary aldosteronism after treatment. Arch Intern Med. 2008;168(1):80-85. doi: 10.1001/archinternmed.2007.33.
11) van der Merwe W, van der Merwe V. May we at least have a civilised discussion about primary aldosteronism in New Zealand? N Z Med J. 2015;128(1419):63-64.
12) Mulatero P, Stowasser M, Loh KC et al. Increased diagnosis of primary aldosteronism, including surgically correctable forms, in centers from five continents. J Clin Endocrinol Metab. 2004;89(3):1045-1050. doi: 10.1210/jc.2003-031337.
13) Kaplan NM. Hypokalemia in the hypertensive patient, with observations on the incidence of primary aldosteronism. Ann Intern Med. 1967;66(6):1079-1090. doi: 10.7326/0003-4819-66-6-1079.
14) Nainani AK, Yang J, Peters S, Russell G. ‘I can’t understand why others don’t screen more’: a qualitative study exploring why Australian general practitioners screen for primary aldosteronism. BMJ Open. 2022;12(6):e061671. doi: 10.1136/bmjopen-2022-061671.
15) Stowasser M, Ahmed AH, Pimenta E et al. Factors affecting the aldosterone/renin ratio. Horm Metab Res. 2012;44(3):170-176. doi: 10.1055/s-0031-1295460.
16) Burrello J, Burrello A, Stowasser M et al. The Primary Aldosteronism Surgical Outcome Score for the Prediction of Clinical Outcomes After Adrenalectomy for Unilateral Primary Aldosteronism. Ann Surg. 2020;272(6):1125-1132. doi: 10.1097/SLA.0000000000003200.
17) Stats NZ. Subnational population estimates (DHB, DHB constituency), by age and sex, at 30 June 1996-2021 (2021 boundaries) [Internet]. Wellington (NZ): Statistics NZ [cited 2022 July 30]. Available from: https://nzdotstat.stats.govt.nz/wbos/Index.aspx?DataSetCode=TABLECODE7509.
18) Manolopoulou J, Fischer E, Dietz A et al. Clinical validation for the aldosterone-to-renin ratio and aldosterone suppression testing using simultaneous fully automated chemiluminescence immunoassays. J Hypertens. 2015;33(12):2500-2511. doi: 10.1097/HJH.0000000000000727.
19) Williams TA, Lenders JWM, Mulatero P et al. Outcomes after adrenalectomy for unilateral primary aldosteronism: an international consensus on outcome measures and analysis of remission rates in an international cohort. Lancet Diabetes Endocrinol. 2017;5(9):689-699. doi: 10.1016/S2213-8587(17)30135-3.
20) Lim YY, Shen J, Fuller PJ, Yang J. Current pattern of primary aldosteronism diagnosis: Delayed and complicated. Aust J Gen Pract. 2018;47(10):712-718. doi: 10.31128/AJGP-05-18-4587.
21) WHO Collaborating Centre for Drug Statistics Methodology. ATC/DDD Index 2022 [Internet]. Oslo (NO): World Health Organization; 2022 [cited 2022 Jan 1]. Available from: www.whocc.no/atc_ddd_index/.
22) Suurd DPD, Visscher WP, Vorselaars WMCM et al. A simplified primary aldosteronism surgical outcome score is a useful prediction model when target organ damage is unknown - Retrospective cohort study. Ann Med Surg (Lond). 2021;65:102333. doi: 10.1016/j.amsu.2021.102333.
23) McLean RM, Williams S, Mann JI et al. Blood pressure and hypertension in New Zealand: results from the 2008/09 Adult Nutrition Survey. N Z Med J. 2013;126(1372):66-79.
24) Selak V, Poppe K, Grey C et al. Ethnic differences in cardiovascular risk profiles among 475,241 adults in primary care in Aotearoa, New Zealand. N Z Med J. 2020;133(1521):14-27.
25) Endocrinology Subcommittee for the Pharmacology and Therapeutics Advisory Committee. Record of the Endocrinology Subcommittee of PTAC Meeting held on 30 March 2021. Item 12: Eplerenone for primary aldosteronism [Internet]. Wellington (NZ): Pharmac; 2021 [cited 2023 Mar 18]. Available from: https://pharmac.govt.nz/assets/2021-03-30-Endocrinology-Subcommittee-Records-Web-Version.pdf.
26) Parthasarathy HK, Ménard J, White WB et al. A double-blind, randomized study comparing the antihypertensive effect of eplerenone and spironolactone in patients with hypertension and evidence of primary aldosteronism. J Hypertens. 2011;29(5):980-90. doi: 10.1097/HJH.0b013e3283455ca5.
27) PHARMAC. Community Schedule [Internet]. Wellington (NZ): Pharmac; 2023 [cited 2023 Feb 20]. Available from: https://schedule.pharmac.govt.nz/ScheduleOnline.php.
28) Herd A, Harman R, Taylor E. Surgical outcomes following laparoscopic adrenalectomy for treatment of Conn’s syndrome (primary hyperaldosteronism) between 1999 and 2006. N Z Med J. 2010;123(1324):50-56.
29) Libianto R, Russell GM, Stowasser M et al. Detecting primary aldosteronism in Australian primary care: a prospective study. Med J Aust. 2022;216(8):408-412. doi: 10.5694/mja2.51438.
30) O’Toole SM, Hornby C, Sze WC et al. Performance evaluation of scoring systems for predicting post-operative hypertension cure in primary aldosteronism. Clin Endocrinol. 2021;95(4):576-586. doi: 10.1111/cen.14534.
31) Rossi GP, Rossitto G, Amar L et al. Clinical Outcomes of 1625 Patients With Primary Aldosteronism Subtyped With Adrenal Vein Sampling. Hypertension. 2019;74(4):800-808. doi: 10.1161/HYPERTENSIONAHA.119.13463.
Contact diana@nzma.org.nz
for the PDF of this article
Increasing rates of referrals for investigation of primary aldosteronism at a tertiary centre
View Article PDF
Primary aldosteronism (PA) is a pathology of autonomous secretion of aldosterone from one or both adrenal glands, classically associated with hypertension and hypokalemia. PA appears to carry a greater burden of cardiovascular disease than essential hypertension and is independently associated with end organ damage, including cardiovascular disease, stroke, left ventricular hypertrophy, microalbuminuria and atrial fibrillation.[[1,2]] The prevalence of PA in hypertensive adults is at least 6%, with some estimates exceeding 20% in those with refractory hypertension.[[1,3–4]]
In Aotearoa New Zealand, screening is generally performed with a plasma aldosterone to renin ratio (ARR). PA is suggested by a high aldosterone and appropriately suppressed renin level, with thresholds differing depending on local guidelines, although the sensitivity and specificity of the ARR is variable depending on the assay and threshold used. The diagnosis is confirmed by dynamic testing. A saline infusion test (SIT) is most commonly performed in Aotearoa New Zealand, whereby two litres of normal saline is infused intravenously over 4 hours, and aldosterone levels are measured both before and after the infusion. The expansion in intravascular volume should suppress renin, and thus aldosterone production, in healthy individuals. In patients with PA, the aldosterone concentration remains inappropriately high at the 4-hour mark. Cortisol may also be measured, as a non-suppressed cortisol at 4 hours may indicate a confounding ACTH effect and produce false positive results. For patients in whom the SIT suggests inappropriate aldosterone secretion, adrenal vein sampling (AVS) may then be performed by an interventional radiologist. AVS may be used to identify unilateral disease in patients who are surgical candidates, as unilateral PA can be cured by surgically removing the pathologic adrenal gland. Unilateral disease is typically due to a functioning adenoma, while bilateral disease may be due to adrenal hyperplasia. Contemporary studies show a higher frequency of unilateral adenomas than reported historically with the availability of improved imaging and biochemical assessments, suggesting a higher proportion of unilateral disease in patients with confirmed PA.[[5]] Computed tomography (CT) can be used to identify an adenoma, but caution needs to be applied to interpreting the significance of an adenoma, particularly in older patients where incidental adrenal adenomas are common.[[6]] The exact role of AVS compared with CT to inform the decision to undertake adrenalectomy remains to be clarified.[[7]] Patients with bilateral disease or those who choose not to undertake surgery may be treated with mineralocorticoid antagonists (MRAs) such as spironolactone or eplerenone; however, eplerenone is not funded for this indication in Aotearoa New Zealand. Both surgical and medical therapies have been shown to be effective at reducing blood pressure, and there are likely additional benefits to cardiovascular health independent of blood pressure improvements.[[8–10]]
Despite a readily available screening test, PA has historically been underdiagnosed in Aotearoa New Zealand and internationally.[[11,12]] Lower frequency of screening may be due to the misconception that hypokalemia is a prerequisite for PA or that it would be managed similarly to essential hypertension once diagnosed.[[11,13]] The requirement for medication changes before dynamic testing also discourages clinicians from screening, especially since the ARR may produce frequent false results even when interfering medications have been removed.[[14,15]]
It is well documented in overseas cohorts that certain characteristics increase the likelihood of PA, including hypokalemia, a higher ARR and a greater antihypertensive requirement.[[8,10]] Other characteristics such as a shorter duration of hypertension, lower BMI and absence of end organ damage can predict the likelihood of success from adrenalectomy.[[16]] However, the predictive value of clinical characteristics in predicting diagnosis or response to treatment has not been studied in Aotearoa New Zealand.
In this audit, we described the characteristics of patients referred to our centre for confirmatory testing for PA. We analysed whether any of these characteristics predicted a diagnosis of PA after a positive screening test. For those with an eventual positive diagnosis, we described their surgical and medical outcomes and whether the Primary Aldosteronism Surgical Outcome (PASO) score was able to predict clinical success of surgery.[[16]] These relationships may aid clinicians in diagnosing PA, and among patients who have confirmed disease, may inform the decision of whether to offer surgical intervention.
Methods
Study population
This audit was reviewed and approved by Capital and Coast District Health Board (CCDHB) Clinical Audit and Research Committee. All patients who attended our unit for an SIT, AVS, adrenalectomy or initial post-intervention follow-up for PA between 1 January 2016 and 31 November 2021 were included. At the time of data collection, our unit covered the regions previously known as Capital and Coast, Hutt Valley and Wairarapa District Health Boards (DHBs), collectively referred to as 3DHB. We additionally performed AVS and adrenalectomies for patients in the mid to lower North Island (previously Hawke’s Bay, Whanganui and MidCentral DHBs). Investigations performed up to 19 April 2022 were included. Patients were identified with a searchable template for SIT and AVS results on the electronic medical record (EMR), and a clinician-recorded list of adrenalectomies was used. Email referrals were manually checked to ensure no eligible patients were missed. Patients referred locally for investigation of PA usually have an SIT performed after an ARR result above the local threshold of 30.5 pmol/mIU or high clinical suspicion, such as resistant hypertension and hypokalemia despite a normal ARR. Patients from other regions were typically referred for AVS after case confirmation, though occasionally referred for SIT or directly for adrenalectomy.
Data collection
Demographic information, clinical and biochemical characteristics and antihypertensive medications were preferentially collected from the referral letter and supplemented by the EMR. Ethnicity was analysed as total response ethnicity under Level 1 ethnic codes and compared to the ethnicity distribution of our catchment based on 2018 Census data.[[17]] AVS and SIT results were extracted from templates filled out in the EMR. Data was stored securely in REDCap on hospital servers.
Diagnosis, subtyping and treatment
Plasma aldosterone concentration for ARR, SIT and AVS was measured in EDTA plasma using the IDS-iSYS® Immunochemiluminometric assay with a reference range of 103 to 1,197 pmol/L when upright. Plasma renin concentration was measured in EDTA plasma using the IDS-iSYS® chemiluminescence immunoassay with a reference range of 5.3 to 99.1 mIU/L when upright. A screening ARR cutoff of 30.5 pmol/mIU is recommended by the manufacturer, with a quoted sensitivity of 98.9% and specificity of 78.9% when units are converted from the validation study.[[18]] Screening ARRs were taken by local laboratories and recommended to be tested while sitting and before 10 am. There was no rapid cortisol assay available during AVS procedures to assess successful cannulation.
Local SIT protocols aligned with Endocrine Society recommendations. MRAs were stopped 6 weeks prior to testing, and all other antihypertensive medications except alpha-blockers and calcium channel blockers were stopped 2 weeks prior, unless continuation was clinically necessary. The protocol for AVS procedures was the same but specified a 4-week washout for MRAs. The test was performed supine with potassium, creatinine and aldosterone measured at 0 minutes and 4 hours. Renin was additionally measured at 0 minutes for a baseline ARR. Protocols did not change over the study period. All SIT and AVS results were interpreted through discussion at a departmental meeting that included at least one consultant endocrinologist. Cortisol was not routinely measured as part of an SIT during the study period. Local SIT protocols use the following 4-hour aldosterone levels as thresholds for the diagnosis of PA:
• <140 pmol/L—aldosterone suppressed; PA excluded.
• 140–280 pmol/L—indeterminate; diagnosis based on renin, aldosterone and potassium levels over duration of the test. May be repeated after optimising interfering factors.
• >280 pmol/L—aldosterone not suppressed; PA confirmed.
Due to the increasing frequency of adrenal incidentalomas with increasing patient age, AVS was required in patients aged over 40 years. AVS were Synacthen-stimulated and performed by a single operator, except in three unsuccessful cases where it was documented that an alternative operator had attempted the procedure. A selectivity index >5:1 confirmed cannulation success and a lateralisation index of >4:1 confirmed unilateral secretion. In patients under 40 years old, radiological lateralisation was accepted without AVS. Based on lateralisation data and patient preference, either surgical or medical therapy was offered as clinically indicated. Patients outside of our catchment were included in surgical outcome analyses if they received adrenalectomy at our centre, and follow-up was available. Patients who were diagnosed and/or followed up at our centre were included in the medical treatment outcome analyses if follow-up was available. Follow-up data was preferentially collected from the EMR through clinic appointment letters 6–12 months after intervention, but interactions outside the window were used when this was unavailable. The primary care and electronic prescription records were used to supplement hospital records or as the main data source when clinic letters were unavailable.
Clinical and biochemical success criteria[[19,20]]
Defined daily dose (DDD) was used to quantify total antihypertensive medication requirement and determine clinical success. DDD was calculated using the World Health Organization ATC/DDD Index 2021.[[21]] Changes in DDD were significant if they were more than 0.5 times the pre-intervention DDD. Changes in blood pressure were significant if systolic blood pressure reduced by >20mmHg or diastolic by >10mmHg. Normotension was defined as office readings of <140/90mmHg or home readings of <135/85mmHg. Antihypertensive medications used for other indications such as alpha-blockers for prostatic hypertrophy or beta-blockers for atrial fibrillation were included in the DDD. Pre-treatment blood pressure readings and DDD were taken from the closest time to initiating an MRA or surgery as possible, typically the clinic appointment prior to intervention, but the referral blood pressure and DDD was used if unavailable.
For surgically treated patients, complete clinical success required normotension and no antihypertensive medication at follow-up, and partial clinical success required a reduction in blood pressure and/or DDD compared to pre-operative status. Patients treated with an MRA were deemed to have had complete clinical success if they remained normotensive on MRA therapy without other antihypertensive medication. Partial clinical success required a reduction in blood pressure and/or DDD after starting an MRA. Complete biochemical success required a normal ARR at follow-up without hypokalemia, while partial biochemical success required reduction in plasma aldosterone of >50% from pre-surgery with no hypokalemia.
PASO surgical outcome score
View Figures 1–3, Tables 1–4.
The PASO surgical outcome score can be used to predict likelihood of complete clinical success for surgically treated patients with unilateral disease, where a favorable outcome is more likely with a shorter duration of hypertension, female sex, lower BMI, lower antihypertensive medication requirement, absence of target organ damage (TOD) and larger adrenal lesion. A score greater than 16 predicts probable clinical success.
Onset of hypertension was determined by the earliest date of the patient taking antihypertensive medications or a recorded diagnosis of hypertension. Sex and BMI were typically provided in the referral letter. TOD was present if the patient had documented microalbuminuria or met criteria for left ventricular hypertrophy on ECG or echocardiogram, as documented in the EMR. When unavailable or when neither of these criteria were met, TOD was entered as absent while BMI, duration of hypertension and adrenal lesion diameter were entered as the median value.[[22]]
Analyses
When patient information was unavailable for a particular analysis, the patient was excluded from that analysis only. Except when discussing treatment outcomes and numbers of procedures performed, patients outside of our region were excluded from analyses as they typically had been diagnosed prior to referral and were heterogeneous to those identified locally.
Independent t-Tests were used to describe differences in means between groups unless results were non-parametric, in which case a Mann–Whitney U test was used. Paired t-Tests were used to describe mean changes before and after intervention unless results were non-parametric, in which case a Wilcoxon Rank-Sum Test was used. Differences in proportions were calculated using an n-1 Chi-squared test. Statistical significance was set at a p-value of 0.05.
Results
180 patients met the criteria for inclusion, 24 being out of our catchment. Among all eligible patients, from 14 October 2014 to 24 March 2022, we performed 173 SIT and 80 AVS procedures, as well as 27 adrenalectomies (Figure 2). This includes 4 SIT, 19 AVS and one adrenalectomy performed for patients out of catchment. Twenty patients had more than one SIT and 16 patients had more than one AVS performed at our centre.
156 eligible patients were identified within our catchment. Baseline characteristics are shown in Table 1. Most of these referrals came from general practitioners (57%), followed by cardiologists (9.6%), renal physicians (5.8%), general medicine physicians (5.8%), general surgeons (1.9%) and other specialists (1.3%). Eleven percent were identified by an endocrinologist. The remainder (7.7%) had unknown referrers. The reported ethnicity of referred patients in comparison to ethnicity rates in our region is shown in Figure 3.
Indications for referral
Forty patients out of 140 identified in our catchment with adequate referral information did not meet any Endocrine Society guidelines for PA screening (29%), as outlined in Table 2. This included two normotensive patients, both of whom were eventually diagnosed with normotensive PA. Three patients had an ARR below the referral threshold of 30.5 pmol/mIU, but there was high clinical suspicion of PA. Two of these patients went on to be diagnosed with PA.
One patient with a suppressed SIT was categorised as having PA due to a convincingly high baseline ARR and aldosterone. One patient with a non-suppressed SIT was determined not to have PA due to an inappropriately high 4-hour renin.
Subtypes of PA
Sixty-three eligible patients within our catchment (40%) eventually received a diagnosis of PA, 76 patients (49%) had the diagnosis excluded, and the remaining 17 patients were still under investigation at completion of data collection or had indeterminate investigations, so were excluded from Table 1. We performed 169 SIT for 150 unique patients within our catchment, contributing to 58 confirmed PA diagnoses, not including five patients diagnosed with PA without SIT, in keeping with Endocrine Society guidelines. This resulted in a positive SIT rate of 34%.
Of those who received a positive diagnosis, 22 had unilateral disease (35%), 20 had bilateral disease (32%) and 21 were unconfirmed (33%), where AVS was still being awaited (9.5%), not attempted (9.5%) or unsuccessful (14%). Six patients under 40 years old with unilateral findings on imaging were subtyped as unilateral PA and referred for surgery without AVS in keeping with local protocols.
Eighty AVS procedures were performed to subtype eligible patients who had confirmed or probable PA, including 14 second-attempt AVS and two third-attempt AVS. The success rate was 64% among patients’ first AVS procedures and 58% when including second and third procedures. Twenty-five of these procedures were performed for patients outside of our catchment, and three unsuccessful AVS attempts were documented as not being performed by the primary operator. AVS failure related to failure to successfully cannulate the right adrenal vein on all but two occasions. There were no significant AVS complications documented in the EMR over the study period.
Treatment outcomes
Clinical and biochemical treatment outcomes are shown in Table 4. Twenty-five patients out of 27 had follow-up available after adrenalectomy performed in our region (mean follow-up 256 days, range 16 to 1,456 days). Of these 25 with available follow-up information, 24 adrenalectomies were performed by a single endocrine surgeon, including three in private. Two additional patients who met eligibility criteria for inclusion in the study were excluded from analyses of treatment success as they had adrenalectomy performed outside of our centre. Twenty-four patients (96%) achieved clinical success, including seven patients (28%) with complete clinical success and 17 (68%) with partial clinical success. Only one patient (4%) had absent clinical success. Complete biochemical success occurred in 20 patients (80%), with one patient each having partial biochemical success and absent biochemical success (4% each). Three patients did not have a post-operative ARR available (12%).
A PASO surgical outcome score of >16 correctly predicted complete clinical success of adrenalectomy for unilateral PA with a sensitivity of 57% and a specificity of 89%. The positive predictive value was 67% and the negative predictive value was 84%, giving the model an accuracy of 80% in our sample.
Of the 20 patients who underwent documented medical treatment with an MRA and had follow-up information available, 14 (70%) achieved clinical success, including seven patients (35%) with complete clinical success and seven (35%) with partial clinical success. Five (25%) had absent clinical success. The remaining one patient had an increased BP but reduced antihypertensive medication dosage. Three additional patients had follow-up with inadequate information to determine clinical success.
Adverse events of surgical treatment
Of the 23 patients with available surgical information who underwent adrenalectomy under the care of an endocrine surgeon in our centre, 12 (52%) experienced self-limiting post-operative numbness. One patient experienced permanent sensory loss at the T12 dermatome. Another was readmitted for transient post-operative pain. One patient experienced a major adverse event (4%) with a post-operative acute kidney injury causing deterioration of chronic kidney disease. There were no recorded cases of significant intraoperative complications or perioperative mortality.
Adverse events of medical treatment
One patient of eight who tried eplerenone experienced severe vomiting, but no other patients had documented adverse effects. Thirteen out of 33 patients who tried spironolactone experienced adverse drug reactions (39%), specifically gynaecomastia (24%), low libido (6%), hypotension (6%), fatigue (3%), constipation (3%) and unspecified intolerance (6%). Two patients (6%) stopped spironolactone for conception. One additional patient not included in the above group denied MRA therapy due to the potential anti-androgenic side effects.
Discussion
We can expect that almost 8,000 people are living with PA in our catchment size of 409,260 individuals aged 20 years or over, given the national prevalence of hypertension of 31% and a conservative estimate that PA is present in 6% of adults with hypertension based on international data.[[17,23]] We are not aware of any recent data on the likely prevalence of PA in Aotearoa New Zealand. Over the 5-year study period we only had 63 confirmed diagnoses of PA in our catchment and 156 eligible patients identified. While this indicates that PA is likely to be heavily underdiagnosed in our region, the number of referrals and SITs undertaken have both increased five-fold from 2015 to 2021 with a clear upward trend, even with the interruptions to services due to COVID-19 in 2020 and 2021. Rates of diagnosis have also generally shown a positive trend over the same period.
Compared to the total response ethnicity of adults aged 20 and over based on 2018 Census data for the region, there were significantly more referrals than expected for patients of Asian ethnicity and less than expected for patients of European ethnicity. The proportion of referred patients who identified as Māori was similar to the proportion of Māori in the region. There are no other data describing ethnic differences in PA prevalence in Aotearoa New Zealand, including the effect of PA on Māori, which is particularly crucial to establish given the inequities in cardiovascular disease which affect Māori.[[24]]
We found that certain characteristics may offer value in predicting a diagnosis of PA independently of having a raised ARR. In our sample of patients referred in for assessment of potential PA, hypokalemia or potassium supplementation were not sensitive indicators of PA but were relatively specific. Other findings may increase the likelihood of PA, including higher antihypertensive requirement, ARR level and cardiovascular disease. Despite 29% of referred patients not having clear evidence of meeting the Endocrine Society criteria for PA screening, we support the use of these criteria as well as using clinical judgement for case detection, given that 18% of patients within our catchment with eventual PA did not clearly meet these criteria at referral.
We observed high rates of complete or partial clinical success in our cohort. However, 42% of male patients experienced gynaecomastia with spironolactone. PA patients may benefit from eplerenone being subsidised as it has lower rates of adverse effects, including in our cohort, but thought to have similar or only slightly lower efficacy in treating PA.[[25,26]] Currently eplerenone is not funded to use for PA in Aotearoa New Zealand, despite being funded for use in congestive heart failure, so would cost PA patients $300 per year at a typical dose of 50mg.[[27]]
Burrello et al. have validated a PASO score of >16 to predict complete clinical success of adrenalectomy with an accuracy of 79% (sensitivity of 71%, specificity 84%) with an original cohort of 380 patients, which largely matches what was observed in our sample.[[16]] Our sensitivity was somewhat lower (57%), although we had a very small sample size of only seven patients with complete clinical success.
Little has been published on PA in Aotearoa New Zealand, which is a barrier to generating awareness and understanding of this condition that requires disease-specific therapy to be managed effectively.[[11]] Our study has several strengths which aim to remedy this. It is the first in Aotearoa New Zealand to examine the increasing rates of referrals and testing for PA. Our findings reinforce the importance of identifying PA with targeted screening, including in patients with longstanding hypertension. On average, our cohort had almost 10 years of established hypertension before being diagnosed with PA, but most still received effective intervention.
Our study is also the first in Aotearoa New Zealand to discuss the predictive value of clinical and biochemical characteristics in the diagnosis of PA, as well as attempting to validate the PASO predictive score. Herd et al. have previously described the effectiveness of surgical treatment of PA in Aotearoa New Zealand, but did not go so far as to describe the characteristics that predict a positive diagnosis.[[28]] A handful of papers published in other journals have discussed case reports, the point prevalence of PA and role of dynamic testing in Aotearoa New Zealand, but none have described the entire course of diagnosis and treatment, medical treatment outcomes, or increasing referrals. This cohort is comparable with cohorts described in the international literature, noting similar rates of complete clinical success from surgery (28% in our cohort vs 37%), medical treatment (both 35%) and suppressed SITs (46% vs 44%).[[19,20,29]]
Our study has several limitations. Firstly, patients only became eligible when undertaking specialist investigation for PA, so the characteristics we described in Table 1 may therefore be less valid in a primary care setting. Secondly, the study was limited to a single centre and retrospective, which contributed to missing information. For example, 52% of patients did not have sufficient information to exclude or confirm presence of TOD for the PASO score, so the imputation of this value as absent reduces the predictive validity in our cohort. Follow-up occurred outside of the recommended 6–12 month window in 52% of surgically treated patients, which may distort rates of clinical success.[[30]] Our AVS success rates being lower than expected from international literature also reduced observed rates of unilateral disease.[[31]] Finally, we did not routinely measure cortisol as part of our SIT protocols, which may have introduced a confounding effect of aldosterone levels from ACTH.[[8]] We hope to eventually move to using mass spectrometry for these assays when it becomes available in Aotearoa New Zealand, given the inaccuracy of immunoassays in diagnosing PA.[[5]]
Future research should aim to overcome these limitations by describing the status of PA from a primary care point of view where patients are most frequently identified, which may also allow estimates of prevalence, including ethnic differences and inequities. Prospectively describing the medical and surgical outcomes of PA patients would more effectively describe the success rates and harms of treatment, including providing further evidence to support funding eplerenone.
Summary
Abstract
Aim
To describe the frequency and characteristics of patients referred for specialist investigation of primary aldosteronism (PA) in the lower North Island over a 5-year period, and the outcomes of those who received treatment.
Method
Patients who underwent confirmatory testing or treatment for PA at Wellington Regional Hospital were retrospectively identified and data were collected from electronic clinical records.
Results
There has been a five-fold increase in both referrals and confirmatory testing for PA in 2021 compared to 2015. Compared to patients without PA, those eventually diagnosed with PA had a higher ARR, serum sodium, antihypertensive requirement and cardiovascular disease prevalence, as well as lower serum renin, potassium and GFR (all p <0.05), but similar blood pressure. Complete or partial clinical success was achieved in 96% of surgically treated patients compared with 70% of medically treated patients. Thirty-nine percent of patients experienced minor adverse effects with spironolactone and only one significant adverse event was experienced perioperatively.
Conclusion
The rate of referrals and confirmatory testing for PA are increasing in our region. Adrenalectomy and mineralocorticoid antagonist therapy are both safe and effective treatments, although minor adverse effects were common with spironolactone.
Author Information
Acknowledgements
Correspondence
Correspondence Email
Competing Interests
1) Parasiliti-Caprino M, Lopez C, Prencipe N et al. Prevalence of primary aldosteronism and association with cardiovascular complications in patients with resistant and refractory hypertension. J Hypertens. 2020;38(9):1841-1848. doi: 10.1097/HJH.0000000000002441.
2) Milliez P, Girerd X, Plouin PF et al. Evidence for an increased rate of cardiovascular events in patients with primary aldosteronism. J Am Coll Cardiol. 2005;45(8):1243-1248. doi: 10.1016/j.jacc.2005.01.015.
3) Monticone S, Burrello J, Tizzani D et al. Prevalence and clinical manifestations of primary aldosteronism encountered in primary care practice. J Am Coll Cardiol. 2017;69(14):1811-1820. doi: 10.1016/j.jacc.2017.01.052.
4) Brown JM, Siddiqui M, Calhoun DA et al. The unrecognised prevalence of primary aldosteronism: a cross-sectional study. Ann Intern Med. 2020;173(1):10-20. doi: 10.7326/M20-0065.
5) Eisenhofer G, Kurlbaum M, Peitzsch M et al. The Saline Infusion Test for Primary Aldosteronism: Implications of Immunoassay Inaccuracy. J Clin Endocrinol Metab. 2022;107(5):e2027-e2036. doi: 10.1210/clinem/dgab924.
6) Jing Y, Hu J, Luo R et al. Prevalence and characteristics of adrenal tumors in an unselected screening population: A cross-sectional study. Ann Intern Med. 2022;175(10):1383-1391. doi: 10.7326/M22-1619.
7) Dekkers T, Prejbisz A, Kool LJS et al. Adrenal vein sampling versus CT scan to determine treatment in primary aldosteronism: an outcome-based randomised diagnostic trial. Lancet Diabetes Endocrinol. 2016;4(9):739-746. doi: 10.1016/S2213-8587(16)30100-0.
8) Funder JW, Carey RM, Mantero F et al. The management of primary aldosteronism: case detection, diagnosis, and treatment: an endocrine society clinical practice guideline. J Clin Endocrinol Metab. 2016;101(5):1889-1916. doi: 10.1210/jc.2015-4061.
9) Muth A, Ragnarsson O, Johannsson G et al. Systematic review of surgery and outcomes in patients with primary aldosteronism. Br J Surg. 2015;102(4):307-317. doi: 10.1002/bjs.9744.
10) Catena C, Colussi G, Nadalini E et al. Cardiovascular outcomes in patients with primary aldosteronism after treatment. Arch Intern Med. 2008;168(1):80-85. doi: 10.1001/archinternmed.2007.33.
11) van der Merwe W, van der Merwe V. May we at least have a civilised discussion about primary aldosteronism in New Zealand? N Z Med J. 2015;128(1419):63-64.
12) Mulatero P, Stowasser M, Loh KC et al. Increased diagnosis of primary aldosteronism, including surgically correctable forms, in centers from five continents. J Clin Endocrinol Metab. 2004;89(3):1045-1050. doi: 10.1210/jc.2003-031337.
13) Kaplan NM. Hypokalemia in the hypertensive patient, with observations on the incidence of primary aldosteronism. Ann Intern Med. 1967;66(6):1079-1090. doi: 10.7326/0003-4819-66-6-1079.
14) Nainani AK, Yang J, Peters S, Russell G. ‘I can’t understand why others don’t screen more’: a qualitative study exploring why Australian general practitioners screen for primary aldosteronism. BMJ Open. 2022;12(6):e061671. doi: 10.1136/bmjopen-2022-061671.
15) Stowasser M, Ahmed AH, Pimenta E et al. Factors affecting the aldosterone/renin ratio. Horm Metab Res. 2012;44(3):170-176. doi: 10.1055/s-0031-1295460.
16) Burrello J, Burrello A, Stowasser M et al. The Primary Aldosteronism Surgical Outcome Score for the Prediction of Clinical Outcomes After Adrenalectomy for Unilateral Primary Aldosteronism. Ann Surg. 2020;272(6):1125-1132. doi: 10.1097/SLA.0000000000003200.
17) Stats NZ. Subnational population estimates (DHB, DHB constituency), by age and sex, at 30 June 1996-2021 (2021 boundaries) [Internet]. Wellington (NZ): Statistics NZ [cited 2022 July 30]. Available from: https://nzdotstat.stats.govt.nz/wbos/Index.aspx?DataSetCode=TABLECODE7509.
18) Manolopoulou J, Fischer E, Dietz A et al. Clinical validation for the aldosterone-to-renin ratio and aldosterone suppression testing using simultaneous fully automated chemiluminescence immunoassays. J Hypertens. 2015;33(12):2500-2511. doi: 10.1097/HJH.0000000000000727.
19) Williams TA, Lenders JWM, Mulatero P et al. Outcomes after adrenalectomy for unilateral primary aldosteronism: an international consensus on outcome measures and analysis of remission rates in an international cohort. Lancet Diabetes Endocrinol. 2017;5(9):689-699. doi: 10.1016/S2213-8587(17)30135-3.
20) Lim YY, Shen J, Fuller PJ, Yang J. Current pattern of primary aldosteronism diagnosis: Delayed and complicated. Aust J Gen Pract. 2018;47(10):712-718. doi: 10.31128/AJGP-05-18-4587.
21) WHO Collaborating Centre for Drug Statistics Methodology. ATC/DDD Index 2022 [Internet]. Oslo (NO): World Health Organization; 2022 [cited 2022 Jan 1]. Available from: www.whocc.no/atc_ddd_index/.
22) Suurd DPD, Visscher WP, Vorselaars WMCM et al. A simplified primary aldosteronism surgical outcome score is a useful prediction model when target organ damage is unknown - Retrospective cohort study. Ann Med Surg (Lond). 2021;65:102333. doi: 10.1016/j.amsu.2021.102333.
23) McLean RM, Williams S, Mann JI et al. Blood pressure and hypertension in New Zealand: results from the 2008/09 Adult Nutrition Survey. N Z Med J. 2013;126(1372):66-79.
24) Selak V, Poppe K, Grey C et al. Ethnic differences in cardiovascular risk profiles among 475,241 adults in primary care in Aotearoa, New Zealand. N Z Med J. 2020;133(1521):14-27.
25) Endocrinology Subcommittee for the Pharmacology and Therapeutics Advisory Committee. Record of the Endocrinology Subcommittee of PTAC Meeting held on 30 March 2021. Item 12: Eplerenone for primary aldosteronism [Internet]. Wellington (NZ): Pharmac; 2021 [cited 2023 Mar 18]. Available from: https://pharmac.govt.nz/assets/2021-03-30-Endocrinology-Subcommittee-Records-Web-Version.pdf.
26) Parthasarathy HK, Ménard J, White WB et al. A double-blind, randomized study comparing the antihypertensive effect of eplerenone and spironolactone in patients with hypertension and evidence of primary aldosteronism. J Hypertens. 2011;29(5):980-90. doi: 10.1097/HJH.0b013e3283455ca5.
27) PHARMAC. Community Schedule [Internet]. Wellington (NZ): Pharmac; 2023 [cited 2023 Feb 20]. Available from: https://schedule.pharmac.govt.nz/ScheduleOnline.php.
28) Herd A, Harman R, Taylor E. Surgical outcomes following laparoscopic adrenalectomy for treatment of Conn’s syndrome (primary hyperaldosteronism) between 1999 and 2006. N Z Med J. 2010;123(1324):50-56.
29) Libianto R, Russell GM, Stowasser M et al. Detecting primary aldosteronism in Australian primary care: a prospective study. Med J Aust. 2022;216(8):408-412. doi: 10.5694/mja2.51438.
30) O’Toole SM, Hornby C, Sze WC et al. Performance evaluation of scoring systems for predicting post-operative hypertension cure in primary aldosteronism. Clin Endocrinol. 2021;95(4):576-586. doi: 10.1111/cen.14534.
31) Rossi GP, Rossitto G, Amar L et al. Clinical Outcomes of 1625 Patients With Primary Aldosteronism Subtyped With Adrenal Vein Sampling. Hypertension. 2019;74(4):800-808. doi: 10.1161/HYPERTENSIONAHA.119.13463.
Contact diana@nzma.org.nz
for the PDF of this article
Increasing rates of referrals for investigation of primary aldosteronism at a tertiary centre
Primary aldosteronism (PA) is a pathology of autonomous secretion of aldosterone from one or both adrenal glands, classically associated with hypertension and hypokalemia. PA appears to carry a greater burden of cardiovascular disease than essential hypertension and is independently associated with end organ damage, including cardiovascular disease, stroke, left ventricular hypertrophy, microalbuminuria and atrial fibrillation.
Subscriber Content
The full contents of this pages only available to subscribers.
Login, subscribe or email nzmj@nzma.org.nz to purchase this article.
