Systematic review of surgery and outcomes in patients with primary aldosteronism

Abstract

Background

Primary aldosteronism (PA) is the most common cause of secondary hypertension. The main aims of this paper were to review outcome after surgical versus medical treatment of PA and partial versus total adrenalectomy in patients with PA.

Methods

Relevant medical literature from PubMed, the Cochrane Library and Embase OvidSP from 1985 to June 2014 was reviewed.

Results

Of 2036 records, 43 articles were included in the final analysis. Twenty-one addressed surgical versus medical treatment of PA, four considered partial versus total adrenalectomy for unilateral PA, and 18 series reported on surgical outcomes. Owing to the heterogeneity of protocols and reported outcomes, only a qualitative analysis was performed. In six studies, surgical and medical treatment had comparable outcomes concerning blood pressure, whereas six showed better outcome after surgery. No differences were seen in cardiovascular complications, but surgery was associated with the use of fewer antihypertensive medications after surgery, improved quality of life, and (possibly) lower all-cause mortality compared with medical treatment. Randomized studies indicate a role for partial adrenalectomy in PA, but the high rate of multiple adenomas or adenoma combined with hyperplasia in localized disease is disconcerting. Surgery for unilateral dominant PA normalized BP in a mean of 42 (range 20–72) per cent and the biochemical profile in 96–100 per cent of patients. The mean complication rate in 1056 patients was 4·7 per cent.

Conclusion

Recommendations for treatment of PA are hampered by the lack of randomized trials, but support surgical resection of unilateral disease. Partial adrenalectomy may be an option in selected patients.

Introduction

Primary aldosteronism (PA), or Conn's syndrome, is the most common cause of secondary hypertension, with an estimated prevalence of 5–13 per cent¹. PA is caused by overproduction of aldosterone from one or both of the adrenal glands. It typically presents with hypertension but not always with hypokalaemia². Compared with patients with essential hypertension and comparable BP levels, patients with PA have an increased cardiovascular and cerebrovascular risk, and more commonly impaired renal function^1,3–6.

The most common (more than 90 per cent) diseases underlying PA are idiopathic bilateral hyperplasia and aldosterone-producing adenoma, although other rare causes exist⁷. The recommended screening method for PA is measurement of aldosterone and renin in plasma, and subsequent calculation of the aldosterone to renin ratio (ARR)^8–10. After confirmatory testing, identification of the PA subtype is recommended for most patients because unilateral dominant lesions are most commonly treated by surgery. Adrenal venous sampling is the standard investigation for subtype evaluation and can be performed with low morbidity. Patients with bilaterally increased aldosterone production and patients with unilateral disease who are not candidates for surgery are treated with mineralocorticoid (aldosterone) receptor antagonists (spironolactone or eplerenone)¹¹. In unilateral dominant disease, laparoscopic or retroperitoneoscopic surgical approaches have become the preferred methods owing to small tumour size and low frequency of adrenocortical cancer in PA. The concept that unilateral disease is caused by a single adenoma has made partial adrenalectomy a possible alternative to total adrenalectomy¹². This has been challenged by detailed histopathological studies showing hyperplasia, combined adenoma and hyperplasia, or multiple adenomas in many patients in whom adrenal venous sampling showed unilateral disease^13,14.

Recent data from the German Conn Registry showed an overall adrenal venous sampling rate in patients with PA of 32 per cent (range 19–84 per cent in the various centres). Adrenalectomy rates ranged from 15 to 44 per cent¹⁵, suggesting large variation in practice. This implies that a significant number of patients with unilateral PA are not considered for surgery and therefore do not undergo subtype evaluation. To elucidate the outcome after surgical versus medical treatment of patients with PA, and after partial versus total adrenalectomy in patients with PA, a systematic review of the literature was performed. A secondary aim of the study was to assess overall complication and cure rates after surgery for PA.

Methods

Standard Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines¹⁶ were followed. The two main study objectives were formulated according to the PICO (Patient, Intervention, Comparison, Outcomes) model: surgical versus medical treatment in PA (P, patients with PA; I, adrenalectomy; C, medical treatment; O, normalized potassium, normalized or improved BP, cardiovascular events, quality of life, renal function, complications and adverse effects); and partial versus total adrenalectomy in PA (P, patients with lateralized PA; I, partial adrenalectomy; C, total adrenalectomy; O, as in the first objective, and normalized ARR). Outcome measures were analysed from larger surgical series comprising 70 or more patients.

A detailed web-based search of the databases PubMed, Cochrane Library and Embase OvidSP was last performed on 25 June 2014. Search strategies combined the following terms: hyperaldosteronism, hyperaldosteronaemia, aldosteronism, aldosteronaemia, adrenalectomy(-ies), surgery, surgical, resection (Appendix S1, supporting information). The search was limited to articles published after 1985 in English, German or the Scandinavian languages, to studies in humans, and to patients older than 18 years. Based on the title of the studies and abstracts, reviews, letters, editorials and case reports were excluded from analysis. Full articles of interest were retrieved and analysed by at least two of the authors. Disagreement was resolved by a third author. Small clinical series, publications without relevant outcome data, and duplicates were excluded. For articles that reported data on the same patient cohort and the same outcome, only the latest publication with extractable data was included. Main findings from relevant articles were extracted and tabulated. Improved BP was defined as fewer medications to maintain the same BP, or lower systolic or diastolic BP, no higher than 140/90 mmHg, with unchanged or less medication. Complications were classified according to the Clavien–Dindo system¹⁷.

Statistical analysis

Outcome data were categorized (normalized ARR (yes or no), normalized potassium (yes or no), normalized or improved BP (normalized, improved or not improved)), and percentages calculated. Mean values for improvement in BP after surgery were calculated, considering the sizes of the different individual series.

Results

Results of the web-based search are shown in Fig. 1. After exclusion of duplicates, and screening of titles and abstracts, 68 full-text articles were retrieved and further analysed. Of these, 25 were excluded owing to lack of relevant data (16 articles), outcomes not reported uniformly or not separated for treatment modality (7), or overlapping publications (2). Forty-three publications were included in the qualitative analysis; 21^3,4,18–36 addressed adrenalectomy versus medical treatment for PA, four^12,37–39 total versus partial adrenalectomy for lateralized PA, and 18^{13,38,40–55} reported outcomes after surgery for PA in patient series comprising 70 or more patients. Owing to wide variations in definitions of biochemical diagnosis of PA, heterogeneity in investigations and treatment protocols, and lack of standardized reporting of outcomes, data could not be extracted for a quantitative meta-analysis.

Fig. 1

PRISMA flow diagram showing selection of articles for review. PA, primary aldosteronism

Surgical versus medical treatment of primary aldosteronism

Data on surgical versus medical treatment of PA are summarized in Table 1. In six studies, five prospective^{3,19,22,29,35} and one retrospective³⁶, the effects on improvement of BP and hypokalaemia were similar in surgically and medically treated patients. In another six studies, four prospective^23,25,26,32 and two retrospective^4,18, the effects on BP and hypokalaemia were better in surgically treated patients, or they needed fewer antihypertensive drugs.

Table 1

Studies comparing medical and surgical treatment of primary aldosteronism

Reference	Location	Design	Study period	Screening/confirmatory test/subtype evaluation	ADX	Medical treatment	Follow-up (years)	Main outcome variables	Main findings
Miyake et al.18 (2014)	Japan (multicentre)	Retrospective	2003–2007	n.a.	733	626; type of treatment n.a.	n.a.	BP and K	A somewhat better effect on BP and hypokalaemia for ADX
Zacharieva et al.19 (2006)	Sofia, Bulgaria	Prospective	n.a.	ARR/none/PT	30	34; all spironolactone (≥ 100 mg)	3·0 ADX; 0·5 medical treatment	BP and K	Similar effect on BP and hypokalaemia in the two groups
Catena et al.3 (2008)	Udine, Italy	Prospective	1994–2001	ARR/SIT/AVS (26%), NP-59 (87%)	24	31; all spironolactone (≥ 100 mg)	7·4	Cardiovascular outcome	CHD, cerebrovascular events and arrhythmias no different between groups
Mulatero et al.4 (2013)	Torino, Italy	Retrospective	1992–2009	ARR/SIT/AVS (33%)	57	213; all spironolactone (dose n.a.)	12	Cardiovascular outcome	CHD, cerebrovascular events and arrhythmias no different between groups
Reincke et al.20 (2012)	Germany (multicentre)	Cross-sectional	1994–2010	ARR/SIT, FST or CST/AVS (selective)	157	180; all MRA (type or dose n.a.)	10	Cardiovascular outcome	Increased cardiovascular mortality in PA; all-cause mortality increased in medical treatment group
Catena et al.21 (2007)	Udine, Italy	Prospective	n.a.	ARR/SIT/AVS (26%), NP-59 (87%)	24	30; all spironolactone (≥ 100 mg)	6·4	Left ventricular mass	Left ventricular mass decreased in both groups
Rossi et al.22 (2013)	Padua, Italy	Prospective	1992–2012	ARR/CST/AVS (100%)	110	70; all MRA (type or dose n.a.)	3·0	Left ventricular mass	Left ventricular mass decreased in both groups
Giacchetti et al.23 (2007)	Padua, Italy	Prospective	2003–2004	ARR/SIT/AVS (selective)	25	36; 27 MRA (type and dose n.a.)	2·9 ADX; 4·6 medical treatment	Left ventricular mass/glucose metabolism	Glucose metabolism improved and left ventricular mass decreased in both groups
Catena et al.24 (2006)	Udine, Italy	Prospective	n.a.	ARR/SIT/AVS and/or NP-59 (100%)	20	27; all spironolactone (50–300 mg/day)	5·7	Glucose metabolism	Fasting glucose and insulin sensitivity improved similarly in the two groups
Fourkiotis et al.25 (2013)	Germany (multicentre)	Prospective	2008–2011	ARR/SIT, FST, CST, OST/AVS (54–87%)	86	83; 65 spironolactone (64 ± 6 mg/day), 18 eplerenone (88 ± 11 mg/day)	5·5	Renal function	GFR and albumin excretion decreased to comparable degree in the two groups
Iwakura et al.26 (2014)	Sendai, Japan	Prospective	2007–2010	ARR/CST/AVS (100%)	102	111; all MRA (type or dose n.a.)	1·0 ADX; 0·7 medical treatment	Renal function	Prevalence of chronic kidney disease increased, GFR and albumin excretion decreased to comparable degree in the two groups
Reincke et al.27 (2009)	Germany (multicentre)	Case–control	1990–1999	ARR (88%)/SIT, FST, CST or OST (62%)/AVS (34%)	51	63; all spironolactone (25–150 mg/day)	n.a.	Renal function	GFR decreased and serum creatinine increased in both groups
Sechi et al.28 (2006)	Udine, Italy	Prospective	1994–2001	ARR/SIT/AVS (24%), NP-59 (88%)	22	28; all spironolactone (50–300 mg/day)	6·4	Renal function	GFR and albumin excretion decreased to comparable degree in the two groups
Sechi et al.29 (2009)	Udine, Italy	Prospective	n.a.	ARR/SIT/AVS (22%), NP-59 (87%)	24	30; all spironolactone (50–300 mg/day)	1·0	Renal function	Intrarenal vascular resistance increased and urinary protein losses decreased similarly in the two groups
Wu et al.30 (2011)	Taiwan (multicentre)	Prospective	2007–2009	ARR/CST + SIT/AVS (54%), NP-59 (42%)	63	61; all spironolactone (50 mg/day)	1·0	Renal function	GFR and albumin decreased only after ADX
Wu et al.31 (2011)	Taiwan (multicentre)	Prospective	2003–2007	ARR + CST/SIT/AVS (52%), NP-59 (69%)	185	101; all spironolactone (dose n.a.)	2·0 ADX; 1·0 medical treatment	Renal function	GFR decreased and serum creatinine increased in both groups
Ahmed et al.32 (2011)	Brisbane, Australia	Prospective	2009–2010	ARR/FST/AVS (100%)	22	21; 12 spironolactone (12·5–25 mg/day)	0·5	QoL	QoL improved more slowly and to a lesser degree in the medical treatment group
Kunzel et al.33 (2012)	Germany (multicentre)	Prospective	2008–2009	n.a.	49	56; 49 spironolactone (15–200 mg/day), 7 eplerenone (25–150 mg/day)	4·3 ADX; 5·4 medical treatment	QoL	QoL worse in female patients treated with MRA than in those having ADX
Apostolopoulou et al.34 (2014)	Munich, Germany	Cross-sectional	2008–2010	ARR/SIT or FS/AVS (72%)	49	56; 49 spironolactone (15–200 mg/day), 7 eplerenone (25–150 mg/day)	4·3 ADX; 5·4 medical treatment	Anxiety and depression	Scores for depression and anxiety worse in female patients treated medically than in those having ADX
Hanusch et al.35 (2014)	Germany (multicentre)	Prospective and cross-sectional	2008–2011	ARR/SIT, FST, CST, OST/AVS (54–87%)	39	57; 39 spironolactone (50–240 mg/day), 13 eplerenone (25–200 mg/day)	5·3	Sleep	No difference in sleep quality between the groups
Kline et al.36 (2013)	Calgary, Canada	Retrospective	2005–2011	ARR/none/AVS (96%)	38	39; 37 MRA (type and dose n.a.)	0·5 ADX; 1·1 medical treatment	Follow-up time/visits	Follow-up time shorter and clinical visits fewer in ADX

ADX, adrenalectomy; n.a., not available; K, potassium; ARR, aldosterone to renin ratio; PT, posture test; SIT, sodium infusion test; AVS, adrenal venous sampling; NP-59, iodocholesterol scintigraphy; CHD, coronary heart disease; FST, fludrocortisone suppression test; CST, captopril suppression test; MRA, mineralocorticoid receptor antagonist; PA, primary aldosteronism; OST, oral sodium loading test; GFR, glomerular filtration rate; QoL, quality of life; FS, furosemide stimulation.

Reincke and colleagues²⁰ reported reduced all-cause mortality following adrenalectomy compared with medical treatment. However, comparing medically and surgically treated patients with PA with regard to coronary heart disease, cerebrovascular events and arrhythmias, outcomes were no different between the two groups of patients^3,4 (Table 1). Similar improvement in glucose metabolism in surgically and medically treated PA was also reported²⁴. Echocardiographic characteristics were evaluated in three studies^21–23 in patients with PA; all showed a reduction in left ventricular mass after adrenalectomy and medical treatment (Table 1). The beneficial effect on ventricular mass seemed to appear earlier after surgical treatment²¹.

The decrease in renal function, as measured by glomerular filtration rate (GFR), and the decrease in urinary albumin excretion were similar in surgically and medically treated patients^25–28,31 (Table 1). Similarly, the low intrarenal vascular resistance observed before treatment of PA (the plausible reason for increased GFR before treatment) increased to similar levels after treatment in both groups²⁹. In one study³⁰, GFR and albumin excretion decreased only in patients treated with adrenalectomy, probably explained by the low spironolactone dose used in the medically treated group (50 mg/day).

General quality of life^32,33, and coexisting depression and anxiety³⁴ in patients with PA, evaluated in three studies, showed better outcome after surgical compared with medical treatment.

Partial versus total adrenalectomy for localized primary aldosteronism

Whether to perform partial or total adrenalectomy in patients with PA was studied in two randomized clinical trials (RCTs) and in prospective and retrospective series (Table 2). In one RCT, where unilateral disease was confirmed with norcholesterol scintigraphy and CT, Nakada and co-workers¹² found no difference in BP improvement, potassium values or plasma aldosterone between patients randomized to open adrenalectomy or tumour enucleation via a high lateral lumbar incision. Details of randomization and blinding were not provided. More recently, Fu et al.³⁷ published a large randomized trial on retroperitoneoscopic partial or total adrenalectomy in PA. No differences in perioperative course or postoperative BP, ARR or potassium levels were found between the groups of patients. Adrenal venous sampling was performed only in selected patients, and blinding of investigators for the treatment group was not ascertained during follow-up.

Table 2

Studies comparing total and partial adrenalectomy in localized primary aldosteronism

Reference	Location	n	Design	Screening/confirmatory test/subtype evaluation	Procedure	Normalized ARR (%)	Hypertension cured/improved (%)	Normalized K (%)	Follow-up (years)	Comments
Fu et al.37 (2011)	Beijing, China	212	RCT	ARR/SIT/AVS selectively	Total ADX 108	100	70/30	100	> 0·5	No blinding during surgery or follow-up
					Partial ADX 104	100	72/28	100
Nakada et al.12 (1995)	Yamagata, Japan NP-59 all	48	RCT	ARR/n.a./AVS selectively,	Total ADX 22	n.a.	n.a.	n.a.	5·2	Details of randomization not provided. Similar improvement with regard to ARR, K and hypertension. Postop. response to angiotensin II infusion similar to that in normal subjects after partial ADX, blunted after total ADX
					Partial ADX 26	n.a.	n.a.	n.a.
Walz et al.38 (2008)	Essen, Germany	183	Prospective	ARR/n.a./AVS selectively	Total ADX 136	n.a.	22/63	n.a.	4·9
					Partial ADX 47	n.a.	57/35	n.a.
Ishidoya et al.39 (2005)	Sendai, Japan	92	Retrospective	n.a./n.a./AVS selectively	Total ADX 63	100	n.a./100	n.a.	2·4	Operations by 10 surgeons, 6·9% multiple lesions in partial specimens, 27% in total ADX
					Partial ADX 29	93	n.a./93	n.a.	5·2

ARR, aldosterone to renin ratio; K, potassium; RCT, randomized clinical trial; SIT, sodium infusion test; AVS, adrenal venous sampling; ADX, adrenalectomy; n.a., not available; NP-59, iodocholesterol scintigraphy.

Overall outcomes following adrenalectomy for primary aldosteronism

A summary of overall outcomes after adrenalectomy for PA is shown in Table 3 and detailed in Table S1 (supporting information). Various techniques were utilized, but the lateral transperitoneal and posterior retroperitoneal approaches were most common. Open procedures were reported only in the very early series. Based on seven studies^{44–46,50–52,55}, including 992 patients, biochemical cure rates ranged from 96 to 100 per cent. Based on 2482 patients from 16 studies^{38,40–47,49–55}, BP was normalized in a mean of 42 (range 20–72) per cent of patients. The cure rate was higher in series with a longer follow-up period. Positive predictive factors for cure were short duration of hypertension, histologically verified aldosterone-producing adenoma, female sex, young age, preserved kidney function, low number of antihypertensive drugs, and mutations in KCNJ5 (inwardly rectifying potassium channel, subfamily J, member 5 gene).

Table 3

Summary of outcomes of surgery for primary aldosteronism

Reference	n	Screening/confirmatory test/subtype evaluation	Normalized ARR/K (%)	HT cured/improved (%)	Mean follow-up (years)	Factors predictive of cure
Zhang et al.40 (2013)	376	ARR/SIT/AVS selectively	n.a./n.a.	55/37	3·6	Duration of HT, plasma aldosterone level
Quillo et al.13 (2011)*	215	ARR/OST/AVS selectively	†	†	1·1	n.a.
Meria et al.47 (2003)	212	ARR/n.a./AVS selectively	Mean normal/100	58/42	3·7	n.a.
Lin et al.48 (2007)	195	n.a.	n.a./n.a.	n.a.	n.a.	n.a.
Walz et al38 (2008)	183	ARR/n.a./AVS selectively	n.a./96	30/57	4·9	Age, sex, duration of HT, APA
Ishidoya et al.52 (2011)	174	ARR/CST/AVS all	99/n.a.	20/69	0·08	n.a.
Letavernier et al.50 (2008)	168	ARR/sARR/AVS selectively	100‡ /n.a.	32/42	0·6	Urinary aldosterone excretion, serum K level
Jiang et al.51 (2014)	164	ARR/SIT + CST/CT	99/99	54/43	4·0	n.a.
van der Linden et al.44 (2012)	156	ARR/sARR/AVS selectively	96§/98	44/56	0·5	Systolic BP, number of drugs, duration of HT, urinary aldosterone excretion
Wu et al.42 (2009)	150	ARR/CST ± SIT/AVS and NP-59 selectively	n.a./n.a.	66/33	4·9	Duration of HT, BMI, age, diastolic BP, renal function
Lim et al.49 (2014)*	133	ARR/OST/AVS selectively	n.a./97	42/46¶	6·8	n.a.
Utsumi et al.45 (2014)	132	ARR/SIT, CST, FS/AVS and/or NP-59 all	100/n.a.	42/56	0·5	Age, duration of HT, number of drugs, sex
Zhang et al.53(2009)	127	n.a.	n.a./n.a.	59/41	3·0	n.a.
Wang et al.43 (2012)	124	ARR/SIT/AVS selectively	n.a./n.a.	55/35	4·9	Duration of HT, number of drugs, response to spironolactone, genotype, APA
Zarnegar et al.41 (2007)	102	ARR/n.a./AVS selectively	n.a./n.a.	39/38	0·5	Normalized BP, BMI (men), age, number of drugs
Proye et al.46 (1998)	100	ARR/n.a./AVS and NP-59 selectively	100/100	56/44	5·4	Response to spironolactone, duration of HT, age, family history of HT, unilateral disease
Lumachi et al.55 (2005)	98	ARR/OST/AVS and NP-59 selectively	100/100	72/24	6·7	Age, duration of HT
Ip et al.54 (2013)	83	n.a.	n.a./n.a.	25/75	0·08	KCNJ5 mutation

*

Overlapping series.

†

Cure in 98 per cent defined as normalization of aldosterone and/or BP off all medications.

‡

Postoperative aldosterone to renin ratio (ARR) available for 81 of 168 patients.

§

Postoperative ARR available for 116 patients.

¶

Cure in 95·5 per cent defined as ‘a combination of the postoperative plasma aldosterone concentration, cure of hypokalaemia, and improvement or cure of hypertension’; BP data based on 127 patients. K, potassium; SIT, sodium infusion test; AVS, adrenal venous sampling; n.a., not available; HT, hypertension; OST, oral sodium loading test; APA, aldosterone-producing adenoma; CST, captopril suppression test; sARR, supine aldosterone to renin ratio; NP-59, iodocholesterol scintigraphy; BMI, body mass index; FS, furosemide stimulation.

Surgical complications were reported in six^{38,43,47,48,50,52} of 18 studies. The mean complication rate in 1056 patients was 4·7 per cent with a range in the various studies of 2–10 per cent, the higher figure from an early series⁴⁷ in the investigated time interval. No deaths and only two grade III complications¹⁷ were noted (Table S2, supporting information). No difference in outcome was seen for laparoscopic or retroperitoneoscopic adrenalectomy, and hospital stay was short for the endoscopic techniques.

Discussion

In this study, surgical treatment of PA has been reviewed systematically, with special reference to outcomes after surgical versus medical treatment, and after partial versus total adrenalectomy. The literature search revealed only two RCTs, both studying partial versus total adrenalectomy. There was wide variation in biochemical definitions of the diagnosis of PA, diagnostic procedures and outcome measures. The outcome measures varied from rate of normalization of serum potassium concentration, mean change, normalization or improvement in BP, change in the number of antihypertensive medications, and change or normalization of the ARR. These outcome measures of treatment are valid, but, as there was no consistency in reported outcomes among trials, a formal meta-analysis could not be performed. Instead, qualitative data from the retrieved articles were summarized.

Comparisons of surgical and medical treatment of PA were difficult in the absence of RCTs. Many studies came from the same centres, included relatively small numbers of patients, and were designed primarily to compare patients with PA with patients with essential hypertension. In many studies, medical treatment was not predefined for dose, type of medication or target BP. Some of the studies that demonstrated superior effect of surgery over medically treated groups of patients received either no treatment with¹⁸, or low doses of ²⁵, mineralocorticoid receptor antagonists, or the medical treatment was not reported⁴.

A further limitation was that the majority of surgically treated patients supposedly had aldosterone-producing adenomas and most medically treated patients had idiopathic bilateral hyperplasia; thus, a comparison was made between two different subtypes of PA that may not be entirely comparable. Comparisons between surgically and medically treated patients with PA should, therefore, be interpreted with caution.

Although several outcome variables, such as cardiovascular risk and albuminuria, appeared to improve to a comparable level after medical and surgical treatment, there are other important aspects that differ between the two types of treatment. Patients treated medically need more antihypertensive drugs^22,23,36, and require a longer follow-up and more clinical visits at specialized referral centres than those treated surgically³⁶. Quality of life^32,33 as well as scores for anxiety and depression³⁴ seem to be worse in medically treated patients with PA. Side-effects are common in patients receiving treatment with mineralocorticoid receptor antagonists, especially spironolactone^3,23,26, contributing to worse quality of life, increased healthcare consumption, and increased risk of non-compliance. Adrenalectomy for PA is also less expensive than long-term medical treatment^56,57.

Two RCTs^12,37 together with a large non-randomized series³⁸ reported comparable outcomes after partial and total adrenalectomy for disease lateralized before surgery. These studies are in contrast to other clinical findings³⁹ and studies of adrenal histology in patients with PA. By use of conventional histopathology, immunohistochemistry and in situ hybridization techniques, hyperplasia, multiple adenoma, or combined hyperplasia and adenoma has been identified in 16–27 per cent of patients with unilateral dominant disease^13,14,58,59.

A number of large case series have confirmed that surgical treatment of PA is a safe treatment with low morbidity and short hospital stay (Table 3; Tables S1 and S2, supporting information). No major differences in cure rate between techniques were seen, but some authors^{38,40–44,46,50,54,55,60} have identified predictive characteristics for cure. Reported complication rates ranged between 2 and 10 per cent, with the highest rate in an early series⁴⁷, possibly reflecting a laparoscopic learning curve. As complication rates were reported in less than half of the studies, data are uncertain, but major complications were rare.

Recommendations on treatment are hampered by the lack of systematic reporting of clearly defined outcomes and RCTs. The present data support surgical resection of unilateral disease, which can be performed with low morbidity. Two RCTs^12,37 showed no difference between partial or total adrenalectomy, but histological findings may suggest an increased risk of treatment failure after partial resection. Surgical treatment improves quality of life, is cost-effective and curative with normalization of aldosterone and renin levels. A suggested algorithm for diagnosis and treatment has been compiled based on consensus documents and guidelines, together with the results of the present study (Fig. S1, supporting information)^{1,7–10,61,62}.

Acknowledgements

A.M. and O.R. contributed equally to this paper. The expert assistance of T. Svanberg at the Clinical Library and Health Technology Assessment Centre at Sahlgrenska University Hospital is greatly appreciated.

Disclosure: The authors declare no conflict of interest.

 

Supporting information

Additional supporting information may be found in the online version of this article:

Appendix S1 Search strategies (Word document)

Table S1 Details of outcome of surgery for primary aldosteronism (Word document)

Table S2 Complications after surgery for primary aldosteronism as reported in major case series (Word document)

Fig. S1 Suggested algorithm for diagnosis and treatment of primary aldosteronism (JPEG file)

Snapshot quiz 15/3

 

Author notes

Cutting edge articles are invited by the BJS Editorial Team, and focus on how current research and innovation will affect future clinical practice.