- Email: [email protected]
Challenges in obesity and primary aldosteronism: Diagnosis and treatment
Surgery xxx (2019) 1e7
Contents lists available at ScienceDirect
Surgery journal homepage: www.elsevier.com/locate/surg
Challenges in obesity and primary aldosteronism: Diagnosis and treatment Victoria M. Gershuni, MD, MSGMa, Daniel S. Herman, MD, PhDb, Rachel R. Kelz, MD, MSCE, MBAa, Robert E. Roses, MDa, Debbie L. Cohen, MDc, Scott O. Trerotola, MDd, Douglas L. Fraker, MDa, Heather Wachtel, MDa,* a
Department of Surgery, Division of Endocrine and Oncologic Surgery, Hospital of the University of Pennsylvania, Philadelphia, PA Department of Pathology and Laboratory Medicine, Hospital of the University of Pennsylvania, Philadelphia, PA Department of Medicine, Division of Renal, Electrolyte and Hypertension, Hospital of the University of Pennsylvania, Philadelphia, PA d Department of Radiology, Division of Vascular and Interventional Radiology, Hospital of the University of Pennsylvania, Philadelphia, PA b c
a r t i c l e i n f o
a b s t r a c t
Article history: Accepted 15 March 2019 Available online xxx
Background: Obese patients may have unrecognized primary aldosteronism due to high rates of concomitant hypertension. We hypothesized that obesity impacts the diagnosis and management of patients with primary aldosteronism. Methods: We conducted a retrospective analysis of all primary aldosteronism patients (n ¼ 418) who underwent adrenal vein sampling (1997e2017). Patients were classified by body mass index as obese (body mass index 35) or nonobese (body mass index <35) and diagnostic evaluation was compared between groups. Within the operative cohort (n ¼ 285), primary outcomes were changes in both blood pressure and antihypertensive medications after adrenalectomy. Secondary outcome was clinical resolution by Primary Aldosteronism Surgery Outcomes criteria. Results: Thirty-five percent of patients were obese. Obese patients were more likely to be male (67.8% vs 56.1%, P ¼ .025), somewhat younger (51.5 vs 54.4 years old, P < .012), and require more preoperative antihypertensive medications (6.7 vs 5.7, P ¼ .04) than nonobese patients. Obese patients had lesser rates of radiologic evidence of adrenal tumors (68.4 vs 77.9%, P ¼ .038) despite similar rates of lateralization on adrenal vein sampling. In the operative subset, obese patients had somewhat smaller tumors on final pathology (1.1 vs 1.5 cm, P ¼ .014) but similar rates of complete and partial clinical resolution (P ¼ 1.000). Conclusion: Obese primary aldosteronism patients have lesser rates of localization by imaging, likely due to smaller tumor size, however, experience similar benefit from adrenalectomy. © 2019 Elsevier Inc. All rights reserved.
Introduction Hypertension is a well-established risk factor for cardiovascular morbidity and mortality. As of 2017, a startling 108.2 million adults in the United States were hypertensive.1 Primary aldosteronism (PA) is the most common cause of secondary hypertension affecting 3% to 20% of hypertensives.2e4 Increased circulating levels of aldosterone are associated with morbidity and mortality in excess of projected risk due to hypertension.5,6 Treatment of PA can decrease cardiovascular risk, however, prolonged duration disease
Presented at the American Association of Endocrine Surgeons, Los Angeles, CA, April 2019. * Reprint requests: Heather Wachtel, MD, 3400 Spruce St, 4 Silverstein Pavilion, Philadelphia, PA 19104. E-mail address: [email protected] (H. Wachtel). https://doi.org/10.1016/j.surg.2019.03.036 0039-6060/© 2019 Elsevier Inc. All rights reserved.
of the disease is associated with poorer clinical outcomes, likely due to irreversible end-organ dysfunction.7,8 Similarly, the prevalence of obesity has reached record levels, affecting nearly 40% of US adults, an estimated 93.3 million patients in 2016.9 Hypertension and are obesity are intimately linked; in the Framingham Heart Study, obesity accounted for approximately 26% and 28% of hypertension in men and women, respectively. The underlying relationship is incompletely understood but is posited to be due to increased systemic vascular resistance, with subsequent increased cardiac output demand.10 An extensive literature exists on the complex interaction between obesity and aldosterone. Obese patients have high aldosterone levels independent of adrenal pathology.11 Adiposity is linked directly to plasma aldosterone levels12 and has been associated with obesity-associated hypertension in certain populations.13,14 Consistent with these observations, aldosterone levels
2
V.M. Gershuni et al. / Surgery xxx (2019) 1e7
decrease after weight loss.15 Recent evidence suggests a mechanistic relationship between the adipokine leptin and the secretion of aldosterone, with a potential positive feedback loop that contributes to obesity-related cardiovascular disease.16 The complexities of aldosterone hormone regulation in the obese inform our incomplete understanding of primary aldosteronism (PA) in obesity. Studies conflict regarding the accuracy of biochemical testing for PA in obese subjects.17,18 Furthermore, the biologic challenges of diagnosing PA in the obese is compounded by social factors. The negative health impact of stigmatized conditions such as obesity has been well documented. Poorer clinical outcomes are mediated by a range of factors, encompassing multiple patient behaviors, such as avoidance of healthcare and decreased compliance and provider behaviors such as lesser rates of recommended screening.19 In the context of these multiple biologic and social determinants, we hypothesized that obesity may impact the diagnosis and management of patients with PA. We therefore examined our large institutional cohort to identify factors that may influence patient outcomes. Methods Study population This study was approved by the Institutional Review Board of the University of Pennsylvania. We performed a retrospective cohort study. All patients (n ¼ 446) with PA who underwent adrenal vein sampling (AVS) at the University of Pennsylvania Health System between 1997 and 2017 were screened for inclusion. Patient data were collected prospectively and maintained in quality assurance databases. All patients had cross-sectional imaging performed. Anthropometric measurements included height and weight which were used to calculate body mass index (BMI, kg/m2). Subjects who did not have height or weight data (n ¼ 28) were excluded from the final study population (n ¼ 418). Subjects were classified as obese (n ¼ 131) or nonobese (n ¼ 287), as defined by us as a BMI cutoff of 35 kg/m2. Subgroup analysis was performed on patients who underwent adrenalectomy at our institution (n ¼ 285). Variables and definitions Abstracted data included patient demographics, measurements of blood pressure (BP), biochemical markers, antihypertensive medications, imaging results, AVS results, and histopathologic diagnosis. Subjects with a biochemical diagnosis of PA (defined as an increased plasma aldosterone concentration with a suppressed plasma renin level and an aldosterone-renin ratio (ARR) of >20 (ng/dL)/(ng/mL/h) or high clinical concern for PA were referred for AVS. Standard criteria for BP were used. Hypertension was defined as 140/90 mm Hg or a mean arterial pressure (MAP) >110 mm Hg. Glomerular filtration rate (GFR) was calculated using the Modification of Diet in Renal Disease Study equation (MDRD) consistent with standard practice. Defined daily dose (DDD) was calculated using the World Health Organization calculator.20 A modified Elixhauser Comorbidity Index was calculated for all patients using International Classification of Diseases, Ninth and Tenth editions, codes for 31 defined comorbidities.21 AVS As published previously by our group, AVS was performed using a modified Mayo protocol.22 Selectivity index (adrenal or caval cortisol) and lateralization index ([greater aldosterone or cortisol (A/C) ratio/(lesser A/C ratio)]) were calculated. AVS was considered successful with a selectivity index of 3. Subjects with
lateralization index 4 were considered lateralizing and were referred for operative management. Outcome measures Diagnostic evaluation was compared in all subjects included in the study. Within the subgroup who underwent adrenalectomy, the primary outcomes after adrenalectomy were change in BP by MAP and change in antihypertensive medications by DDD. Secondary outcomes after adrenalectomy included clinical outcomes. Clinical outcomes were also classified as complete, partial, or absent clinical response according to the Primary Aldosteronism Surgery Outcomes (PASO) consensus criteria.23 Biochemical outcome data were limited and therefore, were not included in the final analysis. BP and antihypertensive medication outcomes were assessed at 2 weeks, 6 to 12 months, and most recent follow-up >12 months after adrenalectomy. Statistics Descriptive statistics were performed; 2-sample comparisons were performed with parametric and nonparametric tests, as appropriate. Statistical analysis was performed using STATA 15.1 (Stata Corporation, College Station, TX). Continuous data are presented as mean ± standard deviation or median values (interquartile range) as appropriate. Results Patients with PA Of the 418 patients with PA who underwent AVS, the mean age was 53.5 ± 10.8 years old, and 59.8% (n ¼ 250) were male (Table I). The mean BMI was 32.3 ± 6.7 kg/m2 .The median duration of hypertension at time of presentation was 10 years.5e17 Approximately three-quarters of the population had hypokalemia documented at presentation (n ¼ 277), and the median ARR was 123.3 (ng/dL)/(ng/mL/h). From this total cohort, 131 (31.3%) were categorized as obese as determined by a BMI 35. The mean BMI in the obese versus nonobese groups was 28.7 and 40.1 kg/m2, respectively (P < .0001). When compared to nonobese patients, obese patients had a greater burden of comorbid disease, as measured by the modified Elixhauser score (median 2 vs 1, P ¼ .02). The obese group was also somewhat younger (51.5 vs 54.4 years old, P ¼ .012), more frequently male (67.8 vs 56.1%, P ¼ .025), and on more antihypertensive medications at time of diagnosis (DDD; 6.7 vs 5.7, P ¼ .04). Renal function was similar by MDRD equation (glomerular filtration rate: 75.5 vs 76.1 mL/min/1.73 m2, P ¼ .73). On biochemical analysis, there was no difference between groups in median ARR (126.3 vs 120 [ng/dL]/[ng/mL/h], P ¼ .87). The findings on cross sectional imaging were divergent between the 2 groups. Obese patients were less likely to have a localizing lesion on cross-sectional imaging than nonobese patients (68.4 vs 77.9%, P ¼ .038). When a lesion was identified, there was no difference in median radiographic tumor size (1.3 vs 1.4 cm, P ¼ .77). There was statistically significant discordance between the fraction of patients localized by abdominal imaging and lateralized by AVS in the obese cohort (Fig 1). Patients with BMI <35 had no statistically significant difference in rates of radiologic localization and rates of positive AVS lateralization, while subjects with BMI 35 had lesser rates of imaging localization compared to AVS lateralization (P < .05). Obese patients lateralized on AVS with near-equivalent rates as nonobese patients (74.2 vs 77.5%, P ¼ .47), and had a similar median lateralization index (7.9 vs 8.8, P ¼ .35).
V.M. Gershuni et al. / Surgery xxx (2019) 1e7
3
Table I Demographic and clinical characteristics of 418 patients with primary aldosteronism who underwent adrenal vein sampling and categorized by BMI status
Mean age, y (SD) Male sex (%) Mean BMI, kg/m2 (SD) Median Elixhauser Score (IQI) Median duration of hypertension, y (IQI) Hypokalemia (%) Preoperative AHM by DDD Median creatinine, mg/dL (IQI) Median GFR (IQI) Median PAC, ng/dL (IQI) Median PRA, ng/mL/h (IQI) Median ARR, (ng/dL)/(ng/mL/h) (IQI) Localizing tumor on imaging (%) Tumor laterality on imaging Right (%) Left (%) Bilateral (%) None (%) Median tumor size on imaging, cm (IQI) Lateralizing AVS (%) Median lateralization index (IQI)
Total cohort (n ¼ 418)
Obese (n ¼ 131)
Nonobese (n ¼ 287)
P value
53.5 (10.8) 250 (59.8) 32.3 (6.7) 1 (0,3) 10 (5,17) 277 (74.7) 6 (3.4, 8.4) 1.0 (0.8, 1.2) 76 (59.8, 89.5) 26.6 (17.8e40.2) 0.19 (0.1e0.4) 123.3 (66.8e233.5) 313 (74.9)
51.5 (9.4) 90 (67.8) 40.1 (4.3) 2 (0,3) 10 (6,15) 92 (78.0) 6.7 (3.7, 9) 1.0 (0.9, 1.2) 75.5 (59.2, 88.6) 24 (17, 42) 0.19 (0.1, 0.3) 126.3 (62.1, 230) 91 (68.4)
54.4 (11.3) 160 (56.1) 28.7 (4.0) 1 (0,3) 10 (5,20) 185 (73.1) 5.7 (3.2, 8) 0.99 (0.8, 1.2) 76.1 (59.9, 90.1) 27 (18.9, 37.5) 0.19 (0.1, 0.4) 120 (68.4, 233.7) 222 (77.9)
.012 .025 <.001 .02 .96 .32 .04 .19 .73 .97 .92 .87 .038 .16
105 (25.1) 208 (49.8) 19 (4.6) 86 (20.6) 1.3 (1.1, 1.9) 318 (76.4) 8.7 (3.4, 21.2)
32 (24.1) 59 (44.4) 6 (4.5) 36 (27.1) 1.3 (1.1, 1.9) 98 (74.2) 7.9 (3.1, 20.3)
73 (25.6) 149 (52.3) 13 (4.6) 50 (17.5) 1.4 (1.1, 1.9) 220 (77.5) 8.8 (3.6, 21.2)
.77 .47 .35
IQI, interquartile interval; GFR, glomerular filtration rate; PAC, plasma aldosterone concentration; PRA, plasma renin level; AHM, anti-hypertensive medications; SD, standard deviation.
Fig 1. Characteristics on imaging grouped by BMI demonstrating increased localization with AVS among patients with greater BMI. Difference becomes statistically significant at BMI 35.
AVS identified potentially curable unilateral PA in 73.8% of obese patients with non-localizing imaging (31 out of 42). Patients undergoing adrenalectomy Of the 418 patients who underwent AVS, 285 underwent adrenalectomy at our institution (Table II). Within this operative cohort, obese patients were somewhat younger (mean age: 49.3 vs 52.4 years, P ¼ .02) and more likely to be male (75 vs 57.6%, P ¼ .004). Obese patients had a greater mean BMI of 40.0 kg/m2 compared to 28.9 kg/m2 in the nonobese group (P < .0001) and seemed to have a greater burden of disease as measured by the median modified
Elixhauser score (2 vs 1, P ¼ .053). There was no apparent difference in the median duration of hypertension at presentation (10 vs 10 years, P ¼ .98), use of antihypertensive medications at diagnosis (median medications 5.4 vs 4.8 DDD, P ¼ .40), incidence of hypokalemia (78.4 vs 76.5%, P ¼ .72), or biochemical markers of PA. Median plasma renin level was suppressed but similar in both groups (0.16 vs 0.2 ng/mL/h, P ¼ .14) and the median plasma aldosterone concentration (24.4 vs 28 ng/dL, P ¼ .84) and ARR (151.3 vs 125.4 [ng/dL]/[ng/mL/h], P ¼ .19) were increased similarly in both groups. The obese group was less likely to have evidence of a localized lesion on cross-sectional imaging (70 vs 83%, P ¼ .013). Furthermore, obese patients were less likely to have evidence of either
4
V.M. Gershuni et al. / Surgery xxx (2019) 1e7
Table II Preoperative characteristics of PA patients undergoing adrenalectomy (n ¼ 285) by BMI status
Mean age, y (SD) Male sex (%) Mean BMI, kg/m2 (SD) Median Elixhauser Score (IQI) Median duration of hypertension, years (IQI) Hypokalemia (%) Preoperative AHM by DDD Median creatinine, mg/dL (IQI) Median GFR (IQI) Median PAC, ng/dL (IQI) Median PRA, ng/mL/h (IQI) Median ARR, (ng/dL)/(ng/ml/h) (IQI) Localizing tumor on imaging (%) Tumor laterality on imaging Right (%) Left (%) Bilateral (%) None (%) Median tumor size on imaging, cm (IQI) Median lateralization index (IQI)
Obese (n ¼ 89)
Nonobese (n ¼ 196)
P value
49.3 (8.6) 67 (75) 40.0 (4.2) 2 (0,3) 10 (6,15)
52.4 (11.2) 113 (57.6) 28.9 (4.2) 1 (0,3) 10 (5,20)
.021 .004 <.001 .053 .98
69 (78) 5.4 (2.7, 8.4) 1.0 (0.9, 1.2) 78.0 (61.2, 90.1) 24.4 (17.6, 44.6) 0.16 (0.1, 0.3) 151.3 (82.1, 298) 62 (70)
150 (76.5) 4.8 (2.5, 7.5) 1.0 (0.8, 1.2) 76.1 (59.8, 88.8) 28 (20.4, 40) 0.2 (0.1, 0.4) 125.4 (71.8, 240) 162 (82.7)
.72 .40 .37 .44 .84 .14 .19 .013 0.048
21 (24) 41 (46) 4 (5) 23 (26) 1.3 (1, 1.9)
58 (29.6) 104 (53.0) 10 (5.1) 34 (12.2) 1.5 (1.1, 1.9)
.38
10.7 (5.9, 23.4)
12.7 (6.8, 24.4)
.38
IQI, Interquartile interval; GFR, glomerular filtration rate; PAC, plasma aldosterone concentration; PRA, plasma renin level; AHM, anti-hypertensive medications; SD, standard deviation.
Table III Operative findings in 285 PA patients undergoing adrenalectomy by BMI status
Operative approach Laparoscopic (%) Median tumor size, cm (IQI) Pathology Adenoma (%) Nodular hyperplasia (%)
Obese (n ¼ 88)
Nonobese (n ¼ 197)
99% 1.1 (0.8, 1.8)
99.0% 1.5 (1, 2)
92% 8%
95.9% 4.1%
P value 1.0 .014 .24
IQI, interquartile interval.
unilateral or bilateral lesions compared to nonobese patients (26 vs 12.2%, P ¼ .048). Within the operative cohort of patients who had evidence of a tumor on preoperative imaging, the median radiographic size was near equivalent in both groups (1.3 vs 1.5 cm, P ¼ .38). On evaluation with AVS, there was no difference in median lateralization index (10.7 vs 12.7, P ¼ .38) between obese and nonobese subjects. There were no differences in rates of laparoscopic approach (99 vs 99.0%, P ¼ 1.0) in the obese and nonobese groups. On final surgical pathology, obese patients had smaller tumors (1.1 vs 1.5 cm, P ¼ .014) with no difference in histopathologic subtype (92 vs 95.9% adenomas, P ¼ .24; Table III). In comparison, the preoperatively assessed radiographic tumor size in this cohort was 1.3 cm in obese patients, and 1.5 cm in nonobese patients (Table II). Postoperative outcomes Although the preoperative median DDD of antihypertensive medications trended toward greater values in the obese group (6.7 vs 5.7, P ¼ .06), there was no difference in baseline MAP (108.6 vs 107.4, P ¼ .46; Fig 2). At 6-month follow-up, both groups exhibited an equivalent improvement in mean MAP (98.1 vs 97.2 mm Hg, P ¼ .69), having decreased 16 mm Hg in the obese group and 18 mm Hg in the nonobese group. Likewise, both obese and nonobese patients were able to decrease their need for antihypertensive
medications by approximately one-half (P ¼ .20) and required a similar median DDD of antihypertensive medications for control of BP (3.2 vs 2.7, P ¼ .36). At long-term median follow-up (17 vs 19.5 months, P ¼ .56), obese and nonobese patients both benefited from adrenalectomy as demonstrated by a marked decrease from baseline of MAP (P < .0001) and DDD (P < .001). Obese patients tended to have a slightly greater MAP at long-term follow-up compared to nonobese (100.2 vs 96.7, P ¼ .09) despite similar decreases in antihypertensive medication requirements (0.75 vs 0.67 DDD, P ¼ .63). Using the classification system of the PASO group, complete versus partial versus absent clinical response was determined based on systolic and diastolic BP and the decrease in antihypertensive medications (Table IV). The majority of patients in both groups achieved partial (68.1%) or complete long-term resolution (27.7%) with an average decrease in systolic and diastolic BP of (17 vs 22 mm Hg, P ¼ .55, and 8 vs 6 mm Hg, P ¼ .23, respectively). There seems to be a gradual evolution toward PASO response with more patients falling into the complete response category at a greater duration of time from adrenalectomy. At the earlier time point of 6-months of follow-up, all patients demonstrated a response to adrenalectomy, but obese patients had greater rates of complete response than nonobese patients (8.1 vs 3.0%, P ¼ .014). Discussion In this large, single institution experience of patients with PA, obese patients were more likely to be young, male, have greater rates of comorbid disease at presentation, and a greater medication requirement despite similar biochemical profiles. Obese patients were less likely to have localization by cross-sectional imaging despite equivalent rates of functional lateralization by AVS. The findings on cross sectional imaging were validated on final pathology, where the obese group had smaller tumor sizes which potentially were not identified radiographically. Nevertheless, obese patients experienced early and durable benefit from adrenalectomy, with improvement in control potentially of BP and decreased need for antihypertensive medications. Current guidelines recommend cross-sectional imaging for all patients with biochemically diagnosed PA; AVS is standard of care for subtype differentiation, but may be bypassed for patients 35 years old with an adenoma >10 mm in size and a normal contralateral adrenal gland.24 This study suggests that AVS is the key diagnostic study in obese subjects who may be less likely to have evidence of a lesion on computed tomography. The small tumor size observed in our obese patient population approaches the 10 mm tumor size which is considered localizing. Our institutional practice has been to refer all patients for AVS. In centers where AVS is not readily available, it is easy to speculate that patients who have nonlocalizing imaging may experience a delay in referral for consideration of adrenalectomy or inappropriate exclusion from surgical evaluation. The findings of this study suggest that obese patients may be particularly susceptible to delays in care based on the potential for a lack of findings on imaging. Our group showed recently that an AVS-first, imaging-second approach could decrease the need for abdominal imaging in 43% of patients.22 Given the low rates of localization in obese patients, it is interesting to consider whether this approach would be of particular benefit in this patient population.25 To our knowledge, this is the first study to specifically examine the management of PA in obesity. The observed smaller tumor size in obese patients is provocative and may suggest a difference in underlying biology. The relationship between PA and obesity is incompletely understood. Tirosh et al reported a nonlinear relationship between obesity and plasma aldosterone levels, plasma
V.M. Gershuni et al. / Surgery xxx (2019) 1e7
5
Fig 2. Postoperative improvement in (A) MAP and (B) antihypertensive medication by DDD. *Indicates P < .01.
Table IV Postoperative clinical success (by PASO criteria*) by BMI status (n ¼ 285) Obese Clinical success at 6-months postoperatively Complete 6 (8.1%) Partial 68 (91.9%) Absent 0 (0%) Clinical success at long-term follow-up Complete 13 (27.7%) Partial 32 (68.1%) Absent 2 (4.3%)
Nonobese
P value
3 (3.0%) 170 (98.3%) 0 (0%)
.014
26 (27.7%) 64 (68.1%) 4 (4.3%)
1.0
PASO e Primary Aldosteronism Surgical Outcome; BMI e Body Mass Index. Absent clinical success: Unchanged or increased blood pressure levels with either the same amount or an increase in antihypertensive medication. * PASO criteria were defined as follows: Complete clinical success: normal blood pressure (defined by the European Society of Hypertension Guidelines) without the aid of antihypertensive medication. Partial clinical success: The same blood pressure (defined as a difference (pre- versus postoperatively) of systolic BP <20 mm Hg and diastolic BP <10 mm Hg) with less antihypertensive medication (unchanged medication defined as a change of < ± 0.5 DDD compared to presurgery; less antihypertensive medication is defined as: a decrease of 0.5 DDD compared to presurgery; increased antihypertensive medication is defined as: an increase of 0.5 DDD compared to presurgery); or a reduction in blood pressure (defined as a difference in systolic BP 20 mm Hg or diastolic BP 10 mm Hg) with either the same amount or less antihypertensive medication.
renin activity, and the aldosterone to renin ratio. In their cohort, patients with BMI >30 kg/m2 were less likely to be diagnosed with PA accurately based on biochemical assessment alone.17 Conversely, in a large, multi-institutional prospective study of hypertensive patients, Rossi et al identified a step-wise increase in
plasma aldosterone with BMI quartile, and no association between BMI and the ARR which is used to diagnose PA.18 Although this study does not address the question of diagnosis in our cohort of patients with PA, AVS lateralized at similar rates. As with all retrospective studies, there are inherent limitations in this study. Notwithstanding, the results of our analysis are consistent with other series and existing literature in the PA population. Although the data were collected prospectively, collection occurred over 20 years, and data are heterogeneous; screening and diagnostic practices may have changed over the course of the collection period. Furthermore, these data are only able to capture the obese PA patients in our health system who were evaluated by AVS, so by default, we are limited in the ability to fully describe discordance between imaging and AVS. We are in the process of looking into whether screening practices for PA differ in the obese population at our institution. In conclusion, obese patients with PA have lesser rates of tumor localization on cross-sectional imaging, likely because they have smaller tumors; however, obese patients as a group experience equivalent and durable clinical benefit from adrenalectomy. The interaction between obesity and PA is a complex area with many unanswered questions. Future study is needed to encompass knowledge of screening rates, reliability of diagnosis, and molecular and histopathologic understandings of disease.
Conflict of interest/Disclosure The authors report no proprietary or commercial interest in any product mentioned or discussed in this article.
6
V.M. Gershuni et al. / Surgery xxx (2019) 1e7
References 1. Dorans KS, Mills KT, Liu Y, He J. Trends in prevalence and control of hypertension according to the 2017 American College of Cardiology/American Heart Association (ACC/AHA) Guideline. J Am Heart Assoc. 2018;7. 2. Chao CT, Wu VC, Kuo CC, et al. Diagnosis and management of primary aldosteronism: an updated review. Ann Med. 2013;45:375e383. 3. Williams TA, Reincke M. Management of endocrine disease: Diagnosis and management of primary aldosteronism: the Endocrine Society guideline 2016 revisited. Eur J Endocrinol. 2018;179:R19eR29. 4. Rossi GP, Bernini G, Caliumi C, et al. A prospective study of the prevalence of primary aldosteronism in 1,125 hypertensive patients. J Am Coll Cardiol. 2006;48:2293e2300. 5. Savard S, Amar L, Plouin PF, Steichen O. Cardiovascular complications associated with primary aldosteronism: a controlled cross-sectional study. Hypertension. 2013;62:331e336. 6. Tomaschitz A, Pilz S, Ritz E, Meinitzer A, Boehm BO, Marz W. Plasma aldosterone levels are associated with increased cardiovascular mortality: the Ludwigshafen Risk and Cardiovascular Health (LURIC) study. European Heart Journal. 2010;31:1237e1247. 7. Wachtel H, Cerullo I, Bartlett EK, et al. Long-term blood pressure control in patients undergoing adrenalectomy for primary hyperaldosteronism. Surgery. 2014;156:1394e1402; discussion 402-403. 8. Rossi GP, Sacchetto A, Visentin P, et al. Changes in left ventricular anatomy and function in hypertension and primary aldosteronism. Hypertension. 1996;27: 1039e1045. 9. Ostchega Y, Zhang G, Hughes JP, Nwankwo T. Factors Associated with hypertension control in US adults using 2017 ACC/AHA guidelines: National Health and Nutrition Examination Survey 1999e2016. Am J Hypertens. 2018;31: 886e894. 10. Schmieder RE, Messerli FH. Does obesity influence early target organ damage in hypertensive patients? Circulation. 1993;87:1482e1488. 11. Vecchiola A, Lagos CF, Carvajal CA, Baudrand R, Fardella CE. Aldosterone production and signaling dysregulation in obesity. Curr Hypertens Rep. 2016;18:20. 12. Harada E, Mizuno Y, Katoh D, et al. Increased urinary aldosterone excretion is associated with subcutaneous not visceral, adipose tissue area in obese individuals: a possible manifestation of dysfunctional subcutaneous adipose tissue. Clin Endocrinol (Oxf). 2013;79:510e516.
13. Goodfriend TL, Calhoun DA. Resistant hypertension, obesity, sleep apnea, and aldosterone: theory and therapy. Hypertension. 2004;43:518e524. 14. Goodfriend TL, Egan BM, Kelley DE. Plasma aldosterone, plasma lipoproteins, obesity and insulin resistance in humans. Prostaglandins Leukot Essent Fatty Acids. 1999;60:401e405. 15. Briones AM, Nguyen Dinh Cat A, Callera GE, et al. Adipocytes produce aldosterone through calcineurin-dependent signaling pathways: implications in diabetes mellitus-associated obesity and vascular dysfunction. Hypertension. 2012;59:1069e1078. 16. Faulkner JL, Bruder-Nascimento T, Belin de Chantemele EJ. The regulation of aldosterone secretion by leptin: implications in obesity-related cardiovascular disease. Curr Opin Nephrol Hypertens. 2018;27:63e69. 17. Tirosh A, Hannah-Shmouni F, Lyssikatos C, et al. Obesity and the diagnostic accuracy for primary aldosteronism. J Clin Hypertens (Greenwich). 2017;19: 790e797. 18. Rossi GP, Belfiore A, Bernini G, et al. Body mass index predicts plasma aldosterone concentrations in overweight-obese primary hypertensive patients. J Clin Endocrinol Metab. 2008;93:2566e2571. 19. Phelan SM, Burgess DJ, Yeazel MW, et al. Impact of weight bias and stigma on quality of care and outcomes for patients with obesity. Obes Rev. 2015;16: 319e326. 20. WHO. Available at: https://www.whocc.no/atc_ddd_index/. Accessed August 20, 2019. 21. Elixhauser A, Steiner C, Harris DR, Coffey RM. Comorbidity measures for use with administrative data. Med Care. 1998;36:8e27. 22. Asmar M, Wachtel H, Yan Y, Fraker DL, Cohen D, Trerotola SO. Reversing the established order: Should adrenal venous sampling precede cross-sectional imaging in the evaluation of primary aldosteronism? J Surg Oncol. 2015;112: 144e148. 23. 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;5:689e699. 24. 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 Metabol. 2016;101:1889e1916. 25. Machann J, Horstmann A, Born M, Hesse S, Hirsch FW. Diagnostic imaging in obesity. Best Pract Res Clin Endocrinol Metab. 2013;27:261e277.
Discussion Dr Abbey Fingeret (Omaha, NE): In your cohort, you did not find any difference between laterality. What proportion of your patients had nonselective adrenal vein sampling? And then what was your protocol for managing those patients? How did you incorporate them into this data set? Dr Victoria M. Gershuni: All patients underwent adrenal vein sampling. Of those in both groups approximately 75% did lateralize. So it was similar in both. For those that did not lateralize, we proceeded with our biochemical workup as indicated, and we actually did have multiple attempts at lateralization. So if they failed on their primary attempt, we sometimes brought them back for multiple attempts to try to determine lateralization. For the selectivity index we used a cut-off of 2 for all patients, in accordance with what's published in the literature, and we used a lateralization index of 4. Abbey Fingeret (Omaha, NE): In this cohort, patients that were labeled as nonlateralized could have been nonselective or nonlateralized? Dr Victoria M. Gershuni: Yes. Dr Travis Mckenzie (Rochester, MN): As you know, not all CT scans are equivalent. Did you look at, for instance, slice thickness differences between groups? Dr Victoria M. Gershuni: Unfortunately, we were unable to assess this in our large cohort, but it is expected that over time the slice thickness would have decreased with improvements in technology. So somebody who was diagnosed in 1997 perhaps had a different opportunity to detect a lesion on CT than somebody in 2017. That's a very good point.
Dr Eren Berber (Cleveland Clinic): Do you do AVS in all patients with primary hyperaldosteronism? Dr Victoria M. Gershuni: Yes, we do. All patients undergo AVS Of course, there are some patients who will refuse or cannot obtain the test. That, fortunately, is a very small number. In this cohort, there were only 17 patients who were unable to have AVS performed over those 20 years, and they were excluded from our analysis. Dr Barry Inabnet (New York, NY): I think that if the patient’s BMI is 35 or greater they have a very high chance of hypertension, and we know that the male patient is more likely to have metabolic syndrome with sleep apnea, diabetes, hypertension, so forth. Since they developed hypertension related to their obesity, maybe they are getting imaged earlier, diagnosed earlier, and that's why your tumors are smaller. That's an interesting observation. Dr Victoria M. Gershuni: Definitely. I appreciate the comment because it's something we have talked about. Our initial hypothesis was that obese patients were not being screened as frequently. However, what we are seeing, which is implied by the smaller tumor size, is that perhaps they really are getting screened. However, it just may not be in correct proportion with the vast number of patients that are susceptible to the disease. Dr Barry Inabnet (New York, NY): My question is about how you will use these data in your clinical pathway because there is another group of operations that is very effective for the obese hypertensive patient, and that's weight loss surgery. How will you work with your metabolic surgeons to factor these data that you presented into the pathway for managing the obese hypertensive
V.M. Gershuni et al. / Surgery xxx (2019) 1e7
patient? Should we be screening all of these patients for primary aldosteronism, or should they undergo weight loss surgery and then be screened? Dr Victoria M. Gershuni: That's an excellent question.It's something that we are talking about going forward, and we are actually working with our hypertension medical experts to understand who is getting screened. Currently, as you can imagine,
7
some patients may not come to an endocrinologist or an endocrine surgeon. Instead, they are seen in a bariatric surgery clinic prior. So, we are in the middle of parsing that out, and hopefully we will have a better explanation of what is the appropriate pathway in the future. I think it's important to note that these young males who do have multiple comorbidities may have an underlying reason for their hypertension in addition to their obesity.
Contents lists available at ScienceDirect
Surgery journal homepage: www.elsevier.com/locate/surg
Challenges in obesity and primary aldosteronism: Diagnosis and treatment Victoria M. Gershuni, MD, MSGMa, Daniel S. Herman, MD, PhDb, Rachel R. Kelz, MD, MSCE, MBAa, Robert E. Roses, MDa, Debbie L. Cohen, MDc, Scott O. Trerotola, MDd, Douglas L. Fraker, MDa, Heather Wachtel, MDa,* a
Department of Surgery, Division of Endocrine and Oncologic Surgery, Hospital of the University of Pennsylvania, Philadelphia, PA Department of Pathology and Laboratory Medicine, Hospital of the University of Pennsylvania, Philadelphia, PA Department of Medicine, Division of Renal, Electrolyte and Hypertension, Hospital of the University of Pennsylvania, Philadelphia, PA d Department of Radiology, Division of Vascular and Interventional Radiology, Hospital of the University of Pennsylvania, Philadelphia, PA b c
a r t i c l e i n f o
a b s t r a c t
Article history: Accepted 15 March 2019 Available online xxx
Background: Obese patients may have unrecognized primary aldosteronism due to high rates of concomitant hypertension. We hypothesized that obesity impacts the diagnosis and management of patients with primary aldosteronism. Methods: We conducted a retrospective analysis of all primary aldosteronism patients (n ¼ 418) who underwent adrenal vein sampling (1997e2017). Patients were classified by body mass index as obese (body mass index 35) or nonobese (body mass index <35) and diagnostic evaluation was compared between groups. Within the operative cohort (n ¼ 285), primary outcomes were changes in both blood pressure and antihypertensive medications after adrenalectomy. Secondary outcome was clinical resolution by Primary Aldosteronism Surgery Outcomes criteria. Results: Thirty-five percent of patients were obese. Obese patients were more likely to be male (67.8% vs 56.1%, P ¼ .025), somewhat younger (51.5 vs 54.4 years old, P < .012), and require more preoperative antihypertensive medications (6.7 vs 5.7, P ¼ .04) than nonobese patients. Obese patients had lesser rates of radiologic evidence of adrenal tumors (68.4 vs 77.9%, P ¼ .038) despite similar rates of lateralization on adrenal vein sampling. In the operative subset, obese patients had somewhat smaller tumors on final pathology (1.1 vs 1.5 cm, P ¼ .014) but similar rates of complete and partial clinical resolution (P ¼ 1.000). Conclusion: Obese primary aldosteronism patients have lesser rates of localization by imaging, likely due to smaller tumor size, however, experience similar benefit from adrenalectomy. © 2019 Elsevier Inc. All rights reserved.
Introduction Hypertension is a well-established risk factor for cardiovascular morbidity and mortality. As of 2017, a startling 108.2 million adults in the United States were hypertensive.1 Primary aldosteronism (PA) is the most common cause of secondary hypertension affecting 3% to 20% of hypertensives.2e4 Increased circulating levels of aldosterone are associated with morbidity and mortality in excess of projected risk due to hypertension.5,6 Treatment of PA can decrease cardiovascular risk, however, prolonged duration disease
Presented at the American Association of Endocrine Surgeons, Los Angeles, CA, April 2019. * Reprint requests: Heather Wachtel, MD, 3400 Spruce St, 4 Silverstein Pavilion, Philadelphia, PA 19104. E-mail address: [email protected] (H. Wachtel). https://doi.org/10.1016/j.surg.2019.03.036 0039-6060/© 2019 Elsevier Inc. All rights reserved.
of the disease is associated with poorer clinical outcomes, likely due to irreversible end-organ dysfunction.7,8 Similarly, the prevalence of obesity has reached record levels, affecting nearly 40% of US adults, an estimated 93.3 million patients in 2016.9 Hypertension and are obesity are intimately linked; in the Framingham Heart Study, obesity accounted for approximately 26% and 28% of hypertension in men and women, respectively. The underlying relationship is incompletely understood but is posited to be due to increased systemic vascular resistance, with subsequent increased cardiac output demand.10 An extensive literature exists on the complex interaction between obesity and aldosterone. Obese patients have high aldosterone levels independent of adrenal pathology.11 Adiposity is linked directly to plasma aldosterone levels12 and has been associated with obesity-associated hypertension in certain populations.13,14 Consistent with these observations, aldosterone levels
2
V.M. Gershuni et al. / Surgery xxx (2019) 1e7
decrease after weight loss.15 Recent evidence suggests a mechanistic relationship between the adipokine leptin and the secretion of aldosterone, with a potential positive feedback loop that contributes to obesity-related cardiovascular disease.16 The complexities of aldosterone hormone regulation in the obese inform our incomplete understanding of primary aldosteronism (PA) in obesity. Studies conflict regarding the accuracy of biochemical testing for PA in obese subjects.17,18 Furthermore, the biologic challenges of diagnosing PA in the obese is compounded by social factors. The negative health impact of stigmatized conditions such as obesity has been well documented. Poorer clinical outcomes are mediated by a range of factors, encompassing multiple patient behaviors, such as avoidance of healthcare and decreased compliance and provider behaviors such as lesser rates of recommended screening.19 In the context of these multiple biologic and social determinants, we hypothesized that obesity may impact the diagnosis and management of patients with PA. We therefore examined our large institutional cohort to identify factors that may influence patient outcomes. Methods Study population This study was approved by the Institutional Review Board of the University of Pennsylvania. We performed a retrospective cohort study. All patients (n ¼ 446) with PA who underwent adrenal vein sampling (AVS) at the University of Pennsylvania Health System between 1997 and 2017 were screened for inclusion. Patient data were collected prospectively and maintained in quality assurance databases. All patients had cross-sectional imaging performed. Anthropometric measurements included height and weight which were used to calculate body mass index (BMI, kg/m2). Subjects who did not have height or weight data (n ¼ 28) were excluded from the final study population (n ¼ 418). Subjects were classified as obese (n ¼ 131) or nonobese (n ¼ 287), as defined by us as a BMI cutoff of 35 kg/m2. Subgroup analysis was performed on patients who underwent adrenalectomy at our institution (n ¼ 285). Variables and definitions Abstracted data included patient demographics, measurements of blood pressure (BP), biochemical markers, antihypertensive medications, imaging results, AVS results, and histopathologic diagnosis. Subjects with a biochemical diagnosis of PA (defined as an increased plasma aldosterone concentration with a suppressed plasma renin level and an aldosterone-renin ratio (ARR) of >20 (ng/dL)/(ng/mL/h) or high clinical concern for PA were referred for AVS. Standard criteria for BP were used. Hypertension was defined as 140/90 mm Hg or a mean arterial pressure (MAP) >110 mm Hg. Glomerular filtration rate (GFR) was calculated using the Modification of Diet in Renal Disease Study equation (MDRD) consistent with standard practice. Defined daily dose (DDD) was calculated using the World Health Organization calculator.20 A modified Elixhauser Comorbidity Index was calculated for all patients using International Classification of Diseases, Ninth and Tenth editions, codes for 31 defined comorbidities.21 AVS As published previously by our group, AVS was performed using a modified Mayo protocol.22 Selectivity index (adrenal or caval cortisol) and lateralization index ([greater aldosterone or cortisol (A/C) ratio/(lesser A/C ratio)]) were calculated. AVS was considered successful with a selectivity index of 3. Subjects with
lateralization index 4 were considered lateralizing and were referred for operative management. Outcome measures Diagnostic evaluation was compared in all subjects included in the study. Within the subgroup who underwent adrenalectomy, the primary outcomes after adrenalectomy were change in BP by MAP and change in antihypertensive medications by DDD. Secondary outcomes after adrenalectomy included clinical outcomes. Clinical outcomes were also classified as complete, partial, or absent clinical response according to the Primary Aldosteronism Surgery Outcomes (PASO) consensus criteria.23 Biochemical outcome data were limited and therefore, were not included in the final analysis. BP and antihypertensive medication outcomes were assessed at 2 weeks, 6 to 12 months, and most recent follow-up >12 months after adrenalectomy. Statistics Descriptive statistics were performed; 2-sample comparisons were performed with parametric and nonparametric tests, as appropriate. Statistical analysis was performed using STATA 15.1 (Stata Corporation, College Station, TX). Continuous data are presented as mean ± standard deviation or median values (interquartile range) as appropriate. Results Patients with PA Of the 418 patients with PA who underwent AVS, the mean age was 53.5 ± 10.8 years old, and 59.8% (n ¼ 250) were male (Table I). The mean BMI was 32.3 ± 6.7 kg/m2 .The median duration of hypertension at time of presentation was 10 years.5e17 Approximately three-quarters of the population had hypokalemia documented at presentation (n ¼ 277), and the median ARR was 123.3 (ng/dL)/(ng/mL/h). From this total cohort, 131 (31.3%) were categorized as obese as determined by a BMI 35. The mean BMI in the obese versus nonobese groups was 28.7 and 40.1 kg/m2, respectively (P < .0001). When compared to nonobese patients, obese patients had a greater burden of comorbid disease, as measured by the modified Elixhauser score (median 2 vs 1, P ¼ .02). The obese group was also somewhat younger (51.5 vs 54.4 years old, P ¼ .012), more frequently male (67.8 vs 56.1%, P ¼ .025), and on more antihypertensive medications at time of diagnosis (DDD; 6.7 vs 5.7, P ¼ .04). Renal function was similar by MDRD equation (glomerular filtration rate: 75.5 vs 76.1 mL/min/1.73 m2, P ¼ .73). On biochemical analysis, there was no difference between groups in median ARR (126.3 vs 120 [ng/dL]/[ng/mL/h], P ¼ .87). The findings on cross sectional imaging were divergent between the 2 groups. Obese patients were less likely to have a localizing lesion on cross-sectional imaging than nonobese patients (68.4 vs 77.9%, P ¼ .038). When a lesion was identified, there was no difference in median radiographic tumor size (1.3 vs 1.4 cm, P ¼ .77). There was statistically significant discordance between the fraction of patients localized by abdominal imaging and lateralized by AVS in the obese cohort (Fig 1). Patients with BMI <35 had no statistically significant difference in rates of radiologic localization and rates of positive AVS lateralization, while subjects with BMI 35 had lesser rates of imaging localization compared to AVS lateralization (P < .05). Obese patients lateralized on AVS with near-equivalent rates as nonobese patients (74.2 vs 77.5%, P ¼ .47), and had a similar median lateralization index (7.9 vs 8.8, P ¼ .35).
V.M. Gershuni et al. / Surgery xxx (2019) 1e7
3
Table I Demographic and clinical characteristics of 418 patients with primary aldosteronism who underwent adrenal vein sampling and categorized by BMI status
Mean age, y (SD) Male sex (%) Mean BMI, kg/m2 (SD) Median Elixhauser Score (IQI) Median duration of hypertension, y (IQI) Hypokalemia (%) Preoperative AHM by DDD Median creatinine, mg/dL (IQI) Median GFR (IQI) Median PAC, ng/dL (IQI) Median PRA, ng/mL/h (IQI) Median ARR, (ng/dL)/(ng/mL/h) (IQI) Localizing tumor on imaging (%) Tumor laterality on imaging Right (%) Left (%) Bilateral (%) None (%) Median tumor size on imaging, cm (IQI) Lateralizing AVS (%) Median lateralization index (IQI)
Total cohort (n ¼ 418)
Obese (n ¼ 131)
Nonobese (n ¼ 287)
P value
53.5 (10.8) 250 (59.8) 32.3 (6.7) 1 (0,3) 10 (5,17) 277 (74.7) 6 (3.4, 8.4) 1.0 (0.8, 1.2) 76 (59.8, 89.5) 26.6 (17.8e40.2) 0.19 (0.1e0.4) 123.3 (66.8e233.5) 313 (74.9)
51.5 (9.4) 90 (67.8) 40.1 (4.3) 2 (0,3) 10 (6,15) 92 (78.0) 6.7 (3.7, 9) 1.0 (0.9, 1.2) 75.5 (59.2, 88.6) 24 (17, 42) 0.19 (0.1, 0.3) 126.3 (62.1, 230) 91 (68.4)
54.4 (11.3) 160 (56.1) 28.7 (4.0) 1 (0,3) 10 (5,20) 185 (73.1) 5.7 (3.2, 8) 0.99 (0.8, 1.2) 76.1 (59.9, 90.1) 27 (18.9, 37.5) 0.19 (0.1, 0.4) 120 (68.4, 233.7) 222 (77.9)
.012 .025 <.001 .02 .96 .32 .04 .19 .73 .97 .92 .87 .038 .16
105 (25.1) 208 (49.8) 19 (4.6) 86 (20.6) 1.3 (1.1, 1.9) 318 (76.4) 8.7 (3.4, 21.2)
32 (24.1) 59 (44.4) 6 (4.5) 36 (27.1) 1.3 (1.1, 1.9) 98 (74.2) 7.9 (3.1, 20.3)
73 (25.6) 149 (52.3) 13 (4.6) 50 (17.5) 1.4 (1.1, 1.9) 220 (77.5) 8.8 (3.6, 21.2)
.77 .47 .35
IQI, interquartile interval; GFR, glomerular filtration rate; PAC, plasma aldosterone concentration; PRA, plasma renin level; AHM, anti-hypertensive medications; SD, standard deviation.
Fig 1. Characteristics on imaging grouped by BMI demonstrating increased localization with AVS among patients with greater BMI. Difference becomes statistically significant at BMI 35.
AVS identified potentially curable unilateral PA in 73.8% of obese patients with non-localizing imaging (31 out of 42). Patients undergoing adrenalectomy Of the 418 patients who underwent AVS, 285 underwent adrenalectomy at our institution (Table II). Within this operative cohort, obese patients were somewhat younger (mean age: 49.3 vs 52.4 years, P ¼ .02) and more likely to be male (75 vs 57.6%, P ¼ .004). Obese patients had a greater mean BMI of 40.0 kg/m2 compared to 28.9 kg/m2 in the nonobese group (P < .0001) and seemed to have a greater burden of disease as measured by the median modified
Elixhauser score (2 vs 1, P ¼ .053). There was no apparent difference in the median duration of hypertension at presentation (10 vs 10 years, P ¼ .98), use of antihypertensive medications at diagnosis (median medications 5.4 vs 4.8 DDD, P ¼ .40), incidence of hypokalemia (78.4 vs 76.5%, P ¼ .72), or biochemical markers of PA. Median plasma renin level was suppressed but similar in both groups (0.16 vs 0.2 ng/mL/h, P ¼ .14) and the median plasma aldosterone concentration (24.4 vs 28 ng/dL, P ¼ .84) and ARR (151.3 vs 125.4 [ng/dL]/[ng/mL/h], P ¼ .19) were increased similarly in both groups. The obese group was less likely to have evidence of a localized lesion on cross-sectional imaging (70 vs 83%, P ¼ .013). Furthermore, obese patients were less likely to have evidence of either
4
V.M. Gershuni et al. / Surgery xxx (2019) 1e7
Table II Preoperative characteristics of PA patients undergoing adrenalectomy (n ¼ 285) by BMI status
Mean age, y (SD) Male sex (%) Mean BMI, kg/m2 (SD) Median Elixhauser Score (IQI) Median duration of hypertension, years (IQI) Hypokalemia (%) Preoperative AHM by DDD Median creatinine, mg/dL (IQI) Median GFR (IQI) Median PAC, ng/dL (IQI) Median PRA, ng/mL/h (IQI) Median ARR, (ng/dL)/(ng/ml/h) (IQI) Localizing tumor on imaging (%) Tumor laterality on imaging Right (%) Left (%) Bilateral (%) None (%) Median tumor size on imaging, cm (IQI) Median lateralization index (IQI)
Obese (n ¼ 89)
Nonobese (n ¼ 196)
P value
49.3 (8.6) 67 (75) 40.0 (4.2) 2 (0,3) 10 (6,15)
52.4 (11.2) 113 (57.6) 28.9 (4.2) 1 (0,3) 10 (5,20)
.021 .004 <.001 .053 .98
69 (78) 5.4 (2.7, 8.4) 1.0 (0.9, 1.2) 78.0 (61.2, 90.1) 24.4 (17.6, 44.6) 0.16 (0.1, 0.3) 151.3 (82.1, 298) 62 (70)
150 (76.5) 4.8 (2.5, 7.5) 1.0 (0.8, 1.2) 76.1 (59.8, 88.8) 28 (20.4, 40) 0.2 (0.1, 0.4) 125.4 (71.8, 240) 162 (82.7)
.72 .40 .37 .44 .84 .14 .19 .013 0.048
21 (24) 41 (46) 4 (5) 23 (26) 1.3 (1, 1.9)
58 (29.6) 104 (53.0) 10 (5.1) 34 (12.2) 1.5 (1.1, 1.9)
.38
10.7 (5.9, 23.4)
12.7 (6.8, 24.4)
.38
IQI, Interquartile interval; GFR, glomerular filtration rate; PAC, plasma aldosterone concentration; PRA, plasma renin level; AHM, anti-hypertensive medications; SD, standard deviation.
Table III Operative findings in 285 PA patients undergoing adrenalectomy by BMI status
Operative approach Laparoscopic (%) Median tumor size, cm (IQI) Pathology Adenoma (%) Nodular hyperplasia (%)
Obese (n ¼ 88)
Nonobese (n ¼ 197)
99% 1.1 (0.8, 1.8)
99.0% 1.5 (1, 2)
92% 8%
95.9% 4.1%
P value 1.0 .014 .24
IQI, interquartile interval.
unilateral or bilateral lesions compared to nonobese patients (26 vs 12.2%, P ¼ .048). Within the operative cohort of patients who had evidence of a tumor on preoperative imaging, the median radiographic size was near equivalent in both groups (1.3 vs 1.5 cm, P ¼ .38). On evaluation with AVS, there was no difference in median lateralization index (10.7 vs 12.7, P ¼ .38) between obese and nonobese subjects. There were no differences in rates of laparoscopic approach (99 vs 99.0%, P ¼ 1.0) in the obese and nonobese groups. On final surgical pathology, obese patients had smaller tumors (1.1 vs 1.5 cm, P ¼ .014) with no difference in histopathologic subtype (92 vs 95.9% adenomas, P ¼ .24; Table III). In comparison, the preoperatively assessed radiographic tumor size in this cohort was 1.3 cm in obese patients, and 1.5 cm in nonobese patients (Table II). Postoperative outcomes Although the preoperative median DDD of antihypertensive medications trended toward greater values in the obese group (6.7 vs 5.7, P ¼ .06), there was no difference in baseline MAP (108.6 vs 107.4, P ¼ .46; Fig 2). At 6-month follow-up, both groups exhibited an equivalent improvement in mean MAP (98.1 vs 97.2 mm Hg, P ¼ .69), having decreased 16 mm Hg in the obese group and 18 mm Hg in the nonobese group. Likewise, both obese and nonobese patients were able to decrease their need for antihypertensive
medications by approximately one-half (P ¼ .20) and required a similar median DDD of antihypertensive medications for control of BP (3.2 vs 2.7, P ¼ .36). At long-term median follow-up (17 vs 19.5 months, P ¼ .56), obese and nonobese patients both benefited from adrenalectomy as demonstrated by a marked decrease from baseline of MAP (P < .0001) and DDD (P < .001). Obese patients tended to have a slightly greater MAP at long-term follow-up compared to nonobese (100.2 vs 96.7, P ¼ .09) despite similar decreases in antihypertensive medication requirements (0.75 vs 0.67 DDD, P ¼ .63). Using the classification system of the PASO group, complete versus partial versus absent clinical response was determined based on systolic and diastolic BP and the decrease in antihypertensive medications (Table IV). The majority of patients in both groups achieved partial (68.1%) or complete long-term resolution (27.7%) with an average decrease in systolic and diastolic BP of (17 vs 22 mm Hg, P ¼ .55, and 8 vs 6 mm Hg, P ¼ .23, respectively). There seems to be a gradual evolution toward PASO response with more patients falling into the complete response category at a greater duration of time from adrenalectomy. At the earlier time point of 6-months of follow-up, all patients demonstrated a response to adrenalectomy, but obese patients had greater rates of complete response than nonobese patients (8.1 vs 3.0%, P ¼ .014). Discussion In this large, single institution experience of patients with PA, obese patients were more likely to be young, male, have greater rates of comorbid disease at presentation, and a greater medication requirement despite similar biochemical profiles. Obese patients were less likely to have localization by cross-sectional imaging despite equivalent rates of functional lateralization by AVS. The findings on cross sectional imaging were validated on final pathology, where the obese group had smaller tumor sizes which potentially were not identified radiographically. Nevertheless, obese patients experienced early and durable benefit from adrenalectomy, with improvement in control potentially of BP and decreased need for antihypertensive medications. Current guidelines recommend cross-sectional imaging for all patients with biochemically diagnosed PA; AVS is standard of care for subtype differentiation, but may be bypassed for patients 35 years old with an adenoma >10 mm in size and a normal contralateral adrenal gland.24 This study suggests that AVS is the key diagnostic study in obese subjects who may be less likely to have evidence of a lesion on computed tomography. The small tumor size observed in our obese patient population approaches the 10 mm tumor size which is considered localizing. Our institutional practice has been to refer all patients for AVS. In centers where AVS is not readily available, it is easy to speculate that patients who have nonlocalizing imaging may experience a delay in referral for consideration of adrenalectomy or inappropriate exclusion from surgical evaluation. The findings of this study suggest that obese patients may be particularly susceptible to delays in care based on the potential for a lack of findings on imaging. Our group showed recently that an AVS-first, imaging-second approach could decrease the need for abdominal imaging in 43% of patients.22 Given the low rates of localization in obese patients, it is interesting to consider whether this approach would be of particular benefit in this patient population.25 To our knowledge, this is the first study to specifically examine the management of PA in obesity. The observed smaller tumor size in obese patients is provocative and may suggest a difference in underlying biology. The relationship between PA and obesity is incompletely understood. Tirosh et al reported a nonlinear relationship between obesity and plasma aldosterone levels, plasma
V.M. Gershuni et al. / Surgery xxx (2019) 1e7
5
Fig 2. Postoperative improvement in (A) MAP and (B) antihypertensive medication by DDD. *Indicates P < .01.
Table IV Postoperative clinical success (by PASO criteria*) by BMI status (n ¼ 285) Obese Clinical success at 6-months postoperatively Complete 6 (8.1%) Partial 68 (91.9%) Absent 0 (0%) Clinical success at long-term follow-up Complete 13 (27.7%) Partial 32 (68.1%) Absent 2 (4.3%)
Nonobese
P value
3 (3.0%) 170 (98.3%) 0 (0%)
.014
26 (27.7%) 64 (68.1%) 4 (4.3%)
1.0
PASO e Primary Aldosteronism Surgical Outcome; BMI e Body Mass Index. Absent clinical success: Unchanged or increased blood pressure levels with either the same amount or an increase in antihypertensive medication. * PASO criteria were defined as follows: Complete clinical success: normal blood pressure (defined by the European Society of Hypertension Guidelines) without the aid of antihypertensive medication. Partial clinical success: The same blood pressure (defined as a difference (pre- versus postoperatively) of systolic BP <20 mm Hg and diastolic BP <10 mm Hg) with less antihypertensive medication (unchanged medication defined as a change of < ± 0.5 DDD compared to presurgery; less antihypertensive medication is defined as: a decrease of 0.5 DDD compared to presurgery; increased antihypertensive medication is defined as: an increase of 0.5 DDD compared to presurgery); or a reduction in blood pressure (defined as a difference in systolic BP 20 mm Hg or diastolic BP 10 mm Hg) with either the same amount or less antihypertensive medication.
renin activity, and the aldosterone to renin ratio. In their cohort, patients with BMI >30 kg/m2 were less likely to be diagnosed with PA accurately based on biochemical assessment alone.17 Conversely, in a large, multi-institutional prospective study of hypertensive patients, Rossi et al identified a step-wise increase in
plasma aldosterone with BMI quartile, and no association between BMI and the ARR which is used to diagnose PA.18 Although this study does not address the question of diagnosis in our cohort of patients with PA, AVS lateralized at similar rates. As with all retrospective studies, there are inherent limitations in this study. Notwithstanding, the results of our analysis are consistent with other series and existing literature in the PA population. Although the data were collected prospectively, collection occurred over 20 years, and data are heterogeneous; screening and diagnostic practices may have changed over the course of the collection period. Furthermore, these data are only able to capture the obese PA patients in our health system who were evaluated by AVS, so by default, we are limited in the ability to fully describe discordance between imaging and AVS. We are in the process of looking into whether screening practices for PA differ in the obese population at our institution. In conclusion, obese patients with PA have lesser rates of tumor localization on cross-sectional imaging, likely because they have smaller tumors; however, obese patients as a group experience equivalent and durable clinical benefit from adrenalectomy. The interaction between obesity and PA is a complex area with many unanswered questions. Future study is needed to encompass knowledge of screening rates, reliability of diagnosis, and molecular and histopathologic understandings of disease.
Conflict of interest/Disclosure The authors report no proprietary or commercial interest in any product mentioned or discussed in this article.
6
V.M. Gershuni et al. / Surgery xxx (2019) 1e7
References 1. Dorans KS, Mills KT, Liu Y, He J. Trends in prevalence and control of hypertension according to the 2017 American College of Cardiology/American Heart Association (ACC/AHA) Guideline. J Am Heart Assoc. 2018;7. 2. Chao CT, Wu VC, Kuo CC, et al. Diagnosis and management of primary aldosteronism: an updated review. Ann Med. 2013;45:375e383. 3. Williams TA, Reincke M. Management of endocrine disease: Diagnosis and management of primary aldosteronism: the Endocrine Society guideline 2016 revisited. Eur J Endocrinol. 2018;179:R19eR29. 4. Rossi GP, Bernini G, Caliumi C, et al. A prospective study of the prevalence of primary aldosteronism in 1,125 hypertensive patients. J Am Coll Cardiol. 2006;48:2293e2300. 5. Savard S, Amar L, Plouin PF, Steichen O. Cardiovascular complications associated with primary aldosteronism: a controlled cross-sectional study. Hypertension. 2013;62:331e336. 6. Tomaschitz A, Pilz S, Ritz E, Meinitzer A, Boehm BO, Marz W. Plasma aldosterone levels are associated with increased cardiovascular mortality: the Ludwigshafen Risk and Cardiovascular Health (LURIC) study. European Heart Journal. 2010;31:1237e1247. 7. Wachtel H, Cerullo I, Bartlett EK, et al. Long-term blood pressure control in patients undergoing adrenalectomy for primary hyperaldosteronism. Surgery. 2014;156:1394e1402; discussion 402-403. 8. Rossi GP, Sacchetto A, Visentin P, et al. Changes in left ventricular anatomy and function in hypertension and primary aldosteronism. Hypertension. 1996;27: 1039e1045. 9. Ostchega Y, Zhang G, Hughes JP, Nwankwo T. Factors Associated with hypertension control in US adults using 2017 ACC/AHA guidelines: National Health and Nutrition Examination Survey 1999e2016. Am J Hypertens. 2018;31: 886e894. 10. Schmieder RE, Messerli FH. Does obesity influence early target organ damage in hypertensive patients? Circulation. 1993;87:1482e1488. 11. Vecchiola A, Lagos CF, Carvajal CA, Baudrand R, Fardella CE. Aldosterone production and signaling dysregulation in obesity. Curr Hypertens Rep. 2016;18:20. 12. Harada E, Mizuno Y, Katoh D, et al. Increased urinary aldosterone excretion is associated with subcutaneous not visceral, adipose tissue area in obese individuals: a possible manifestation of dysfunctional subcutaneous adipose tissue. Clin Endocrinol (Oxf). 2013;79:510e516.
13. Goodfriend TL, Calhoun DA. Resistant hypertension, obesity, sleep apnea, and aldosterone: theory and therapy. Hypertension. 2004;43:518e524. 14. Goodfriend TL, Egan BM, Kelley DE. Plasma aldosterone, plasma lipoproteins, obesity and insulin resistance in humans. Prostaglandins Leukot Essent Fatty Acids. 1999;60:401e405. 15. Briones AM, Nguyen Dinh Cat A, Callera GE, et al. Adipocytes produce aldosterone through calcineurin-dependent signaling pathways: implications in diabetes mellitus-associated obesity and vascular dysfunction. Hypertension. 2012;59:1069e1078. 16. Faulkner JL, Bruder-Nascimento T, Belin de Chantemele EJ. The regulation of aldosterone secretion by leptin: implications in obesity-related cardiovascular disease. Curr Opin Nephrol Hypertens. 2018;27:63e69. 17. Tirosh A, Hannah-Shmouni F, Lyssikatos C, et al. Obesity and the diagnostic accuracy for primary aldosteronism. J Clin Hypertens (Greenwich). 2017;19: 790e797. 18. Rossi GP, Belfiore A, Bernini G, et al. Body mass index predicts plasma aldosterone concentrations in overweight-obese primary hypertensive patients. J Clin Endocrinol Metab. 2008;93:2566e2571. 19. Phelan SM, Burgess DJ, Yeazel MW, et al. Impact of weight bias and stigma on quality of care and outcomes for patients with obesity. Obes Rev. 2015;16: 319e326. 20. WHO. Available at: https://www.whocc.no/atc_ddd_index/. Accessed August 20, 2019. 21. Elixhauser A, Steiner C, Harris DR, Coffey RM. Comorbidity measures for use with administrative data. Med Care. 1998;36:8e27. 22. Asmar M, Wachtel H, Yan Y, Fraker DL, Cohen D, Trerotola SO. Reversing the established order: Should adrenal venous sampling precede cross-sectional imaging in the evaluation of primary aldosteronism? J Surg Oncol. 2015;112: 144e148. 23. 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;5:689e699. 24. 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 Metabol. 2016;101:1889e1916. 25. Machann J, Horstmann A, Born M, Hesse S, Hirsch FW. Diagnostic imaging in obesity. Best Pract Res Clin Endocrinol Metab. 2013;27:261e277.
Discussion Dr Abbey Fingeret (Omaha, NE): In your cohort, you did not find any difference between laterality. What proportion of your patients had nonselective adrenal vein sampling? And then what was your protocol for managing those patients? How did you incorporate them into this data set? Dr Victoria M. Gershuni: All patients underwent adrenal vein sampling. Of those in both groups approximately 75% did lateralize. So it was similar in both. For those that did not lateralize, we proceeded with our biochemical workup as indicated, and we actually did have multiple attempts at lateralization. So if they failed on their primary attempt, we sometimes brought them back for multiple attempts to try to determine lateralization. For the selectivity index we used a cut-off of 2 for all patients, in accordance with what's published in the literature, and we used a lateralization index of 4. Abbey Fingeret (Omaha, NE): In this cohort, patients that were labeled as nonlateralized could have been nonselective or nonlateralized? Dr Victoria M. Gershuni: Yes. Dr Travis Mckenzie (Rochester, MN): As you know, not all CT scans are equivalent. Did you look at, for instance, slice thickness differences between groups? Dr Victoria M. Gershuni: Unfortunately, we were unable to assess this in our large cohort, but it is expected that over time the slice thickness would have decreased with improvements in technology. So somebody who was diagnosed in 1997 perhaps had a different opportunity to detect a lesion on CT than somebody in 2017. That's a very good point.
Dr Eren Berber (Cleveland Clinic): Do you do AVS in all patients with primary hyperaldosteronism? Dr Victoria M. Gershuni: Yes, we do. All patients undergo AVS Of course, there are some patients who will refuse or cannot obtain the test. That, fortunately, is a very small number. In this cohort, there were only 17 patients who were unable to have AVS performed over those 20 years, and they were excluded from our analysis. Dr Barry Inabnet (New York, NY): I think that if the patient’s BMI is 35 or greater they have a very high chance of hypertension, and we know that the male patient is more likely to have metabolic syndrome with sleep apnea, diabetes, hypertension, so forth. Since they developed hypertension related to their obesity, maybe they are getting imaged earlier, diagnosed earlier, and that's why your tumors are smaller. That's an interesting observation. Dr Victoria M. Gershuni: Definitely. I appreciate the comment because it's something we have talked about. Our initial hypothesis was that obese patients were not being screened as frequently. However, what we are seeing, which is implied by the smaller tumor size, is that perhaps they really are getting screened. However, it just may not be in correct proportion with the vast number of patients that are susceptible to the disease. Dr Barry Inabnet (New York, NY): My question is about how you will use these data in your clinical pathway because there is another group of operations that is very effective for the obese hypertensive patient, and that's weight loss surgery. How will you work with your metabolic surgeons to factor these data that you presented into the pathway for managing the obese hypertensive
V.M. Gershuni et al. / Surgery xxx (2019) 1e7
patient? Should we be screening all of these patients for primary aldosteronism, or should they undergo weight loss surgery and then be screened? Dr Victoria M. Gershuni: That's an excellent question.It's something that we are talking about going forward, and we are actually working with our hypertension medical experts to understand who is getting screened. Currently, as you can imagine,
7
some patients may not come to an endocrinologist or an endocrine surgeon. Instead, they are seen in a bariatric surgery clinic prior. So, we are in the middle of parsing that out, and hopefully we will have a better explanation of what is the appropriate pathway in the future. I think it's important to note that these young males who do have multiple comorbidities may have an underlying reason for their hypertension in addition to their obesity.